Beej's Guide to C Programming

Brian “Beej Jorgensen” Hall

1 Foreword

No point in wasting words here, folks, let’s jump straight into the C code:

    E((ck?main((z?(stat(M,&t)?P+=a+'{'?0:3:
    execv(M,k),a=G,i=P,y=G&255,
    sprintf(Q,y/'@'-3?A(*L(V(%d+%d)+%d,0)

And they lived happily ever after. The End.

What’s this? You say something’s still not clear about this whole C programming language thing?

Well, to be quite honest, I’m not even sure what the above code does. It’s a snippet from one of the entries in the 2001 International Obfuscated C Code Contest ¹, a wonderful competition wherein the entrants attempt to write the most unreadable C code possible, with often surprising results.

The bad news is that if you’re a beginner in this whole thing, all C code you see probably looks obfuscated! The good news is, it’s not going to be that way for long.

What we’ll try to do over the course of this guide is lead you from complete and utter sheer lost confusion on to the sort of enlightened bliss that can only be obtained though pure C programming. Right on.

1.1 Audience

This guide assumes that you’ve already got some programming knowledge under your belt from another language, such as Python ², JavaScript ³, Java ⁴, Rust ⁵, Go ⁶, Swift ⁷, etc. (Objective-C ⁸ devs will have a particularly easy time of it!)

We’re going to assume you know what variables are, what loops do, how functions work, and so on.

If that’s not you for whatever reason the best I can hope to provide is some pastey entertainment for your reading pleasure. The only thing I can reasonably promise is that this guide won’t end on a cliffhanger…or will it?

1.2 Platform and Compiler

I’ll try to stick to Plain Ol’-Fashioned ISO-standard C ⁹. Well, for the most part. Here and there I might go crazy and start talking about POSIX ¹⁰ or something, but we’ll see.

Unix users (e.g. Linux, BSD, etc.) try running cc or gcc from the command line–you might already have a compiler installed. If you don’t, search your distribution for installing gcc or clang.

Windows users should check out Visual Studio Community ¹¹. Or, if you’re looking for a more Unix-like experience (recommended!), install WSL ¹² and gcc.

Mac users will want to install XCode ¹³, and in particular the command line tools.

There are a lot of compilers out there, and virtually all of them will work for this book. And for those not in the know, a C++ compiler will compile C most code, so it’ll work for the purposes of this guide.

1.3 Official Homepage

This official location of this document is http://beej.us/guide/bgc/¹⁴. Maybe this’ll change in the future, but it’s more likely that all the other guides are migrated off Chico State computers.

1.4 Email Policy

I’m generally available to help out with email questions so feel free to write in, but I can’t guarantee a response. I lead a pretty busy life and there are times when I just can’t answer a question you have. When that’s the case, I usually just delete the message. It’s nothing personal; I just won’t ever have the time to give the detailed answer you require.

As a rule, the more complex the question, the less likely I am to respond. If you can narrow down your question before mailing it and be sure to include any pertinent information (like platform, compiler, error messages you’re getting, and anything else you think might help me troubleshoot), you’re much more likely to get a response.

If you don’t get a response, hack on it some more, try to find the answer, and if it’s still elusive, then write me again with the information you’ve found and hopefully it will be enough for me to help out.

Now that I’ve badgered you about how to write and not write me, I’d just like to let you know that I fully appreciate all the praise the guide has received over the years. It’s a real morale boost, and it gladdens me to hear that it is being used for good! :-) Thank you!

1.5 Mirroring

You are more than welcome to mirror this site, whether publicly or privately. If you publicly mirror the site and want me to link to it from the main page, drop me a line at beej@beej.us.

1.6 Note for Translators

If you want to translate the guide into another language, write me at beej@beej.us and I’ll link to your translation from the main page. Feel free to add your name and contact info to the translation.

Please note the license restrictions in the Copyright and Distribution section, below.

1.7 Copyright and Distribution

With specific exceptions for source code and translations, below, this work is licensed under the Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA.

One specific exception to the “No Derivative Works” portion of the license is as follows: this guide may be freely translated into any language, provided the translation is accurate, and the guide is reprinted in its entirety. The same license restrictions apply to the translation as to the original guide. The translation may also include the name and contact information for the translator.

The C source code presented in this document is hereby granted to the public domain, and is completely free of any license restriction.

Educators are freely encouraged to recommend or supply copies of this guide to their students.

Contact beej@beej.us for more information.

2 Hello, World!

2.1 What to Expect from C

“Where do these stairs go?”
“They go up.”

—Ray Stantz and Peter Venkman, Ghostbusters

C is a low-level language.

It didn’t used to be. Back in the day when people carved punch cards out of granite, C was an incredible way to be free of the drudgery of lower-level languages like assembly ¹⁵.

But now in these modern times, current-generation languages offer all kinds of features that didn’t exist in 1972 when C was invented. This means C is a pretty basic language with not a lot of features. It can do anything, but it can make you work for it.

So why would we even use it today?

As a learning tool: not only is C a venerable piece of computing history, but it is connected to the bare metal ¹⁶ in a way that present-day languages are not. When you learn C, you learn about how software interfaces with computer memory at a low level. There are no seatbelts. You’ll write software that crashes, I assure you. And that’s all part of the fun!
As a useful tool: C still is used for certain applications, such as building operating systems ¹⁷ or in embedded systems ¹⁸. (Though the Rust ¹⁹ programming language is eyeing both these fields!)

If you’re familiar with another language, a lot of things about C are easy. C inspired many other languages, and you’ll see bits of it in Go, Rust, Swift, Python, JavaScript, Java, and all kinds of other languages. Those parts will be familiar.

The one thing about C that hangs people up is pointers. Virtually everything else is familiar, but pointers are the weird one. The concept behind pointers is likely one you already know, but C forces you to be explicit about it, using operators you’ve likely never seen before.

It’s especially insidious because once you grok²⁰ pointers, they’re suddenly easy. But up until that moment, they’re slippery eels.

Everything else in C is just memorizing another way (or sometimes the same way!) of doing something you’ve done already. Pointers are the weird bit.

So get ready for a rollicking adventure as close to the core of the computer as you can get without assembly, in the most influential computer language of all time²¹. Hang on!

2.2 Hello, World!

This is the canonical example of a C program. Everyone uses it. (Note that the numbers to the left are for reader reference only, and are not part of the source code.)

/* Hello world program */

#include <stdio.h>

int main(void)
{
    printf("Hello, World!\n");  // Actually do the work here
}

We’re going to don our long-sleeved heavy-duty rubber gloves, grab a scalpel, and rip into this thing to see what makes it tick. So, scrub up, because here we go. Cutting very gently…

Let’s get the easy thing out of the way: anything between the digraphs /* and */ is a comment and will be completely ignored by the compiler. Same goes for anything on a line after a //. This allows you to leave messages to yourself and others, so that when you come back and read your code in the distant future, you’ll know what the heck it was you were trying to do. Believe me, you will forget; it happens.

Now, what is this #include? GROSS! Well, it tells the C Preprocessor to pull the contents of another file and insert it into the code right there.

Wait—what’s a C Preprocessor? Good question. There are two stages (well, technically there are more than two, but hey, let’s pretend there are two and have a good laugh) to compilation: the preprocessor and the compiler. Anything that starts with pound sign, or “octothorpe”, (#) is something the preprocessor operates on before the compiler even gets started. Common preprocessor directives, as they’re called, are #include and #define. More on that later.

Before we go on, why would I even begin to bother pointing out that a pound sign is called an octothorpe? The answer is simple: I think the word octothorpe is so excellently funny, I have to gratuitously spread its name around whenever I get the opportunity. Octothorpe. Octothorpe, octothorpe, octothorpe.

So anyway. After the C preprocessor has finished preprocessing everything, the results are ready for the compiler to take them and produce assembly code ²², machine code ²³, or whatever it’s about to do. Don’t worry about the technical details of compilation for now; just know that your source runs through the preprocessor, then the output of that runs through the compiler, then that produces an executable for you to run. Octothorpe.

What about the rest of the line? What’s <stdio.h>? That is what is known as a header file. It’s the dot-h at the end that gives it away. In fact it’s the “Standard I/O” (stdio) header file that you will grow to know and love. It contains preprocessor directives and function prototypes (more on that later) for common input and output needs. For our demo program, we’re outputting the string “Hello, World!”, so we in particular need the function prototype for the printf() function from this header file. Basically, if we tried to use printf() without #include <stdio.h>, the compiler would have complained to us about it.

How did I know I needed to #include <stdio.h> for printf()? Answer: it’s in the documentation. If you’re on a Unix system, man printf and it’ll tell you right at the top of the man page what header files are required. Or see the reference section in this book. :-)

Holy moly. That was all to cover the first line! But, let’s face it, it has been completely dissected. No mystery shall remain!

So take a breather…look back over the sample code. Only a couple easy lines to go.

Welcome back from your break! I know you didn’t really take a break; I was just humoring you.

The next line is main(). This is the definition of the function main(); everything between the squirrelly braces ({ and }) is part of the function definition.

How do you call a different function, anyway? The answer lies in the printf() line, but we’ll get to that in a minute.

Now, the main function is a special one in many ways, but one way stands above the rest: it is the function that will be called automatically when your program starts executing. Nothing of yours gets called before main(). In the case of our example, this works fine since all we want to do is print a line and exit.

Oh, that’s another thing: once the program executes past the end of main(), down there at the closing squirrelly brace, the program will exit, and you’ll be back at your command prompt.

So now we know that that program has brought in a header file, stdio.h, and declared a main() function that will execute when the program is started. What are the goodies in main()?

I am so happy you asked. Really! We only have the one goodie: a call to the function printf(). You can tell this is a function call and not a function definition in a number of ways, but one indicator is the lack of squirrelly braces after it. And you end the function call with a semicolon so the compiler knows it’s the end of the expression. You’ll be putting semicolons after most everything, as you’ll see.

You’re passing one parameter to the function printf(): a string to be printed when you call it. Oh, yeah—we’re calling a function! We rock! Wait, wait—don’t get cocky. What’s that crazy \n at the end of the string? Well, most characters in the string look just like they are stored. But there are certain characters that you can’t print on screen well that are embedded as two-character backslash codes. One of the most popular is \n (read “backslash-N”) that corresponds to the newline character. This is the character that causing further printing to continue on the next line instead of the current. It’s like hitting return at the end of the line.

So copy that code into a file called hello.c and build it. On a Unix-like platform (e.g. Linux, BSD, Mac, or WSL), you’ll build with a command like so:

    gcc -o hello hello.c

(This means “compile hello.c, and output an executable called hello”.)

After that’s done, you should have a file called hello that you can run with this command:

    ./hello

(The leading ./ tells the shell to “run from the current directory”.)

And see what happens:

    Hello, World!

It’s done and tested! Ship it!

2.3 Compilation

Let’s talk a bit more about how to build C programs, and what happens behind the scenes there.

Like other languages, C has source code. But, depending on what language you’re coming from, you might never have had to compile your source code into an executable.

Compilation is the process of taking your C source code and turning it into a program that your operating system can execute.

JavaScript and Python devs aren’t used to a separate compilation step at all–though behind the scenes it’s happening! Python compiles your source code into something called bytecode that the Python virtual machine can execute. Java devs are used to compilation, but that produces bytecode for the Java Virtual Machine.

When compiling C, machine code is generated. This is the 1s and 0s that can be executed directly by the CPU.

Languages that typically aren’t compiled are called interpreted languages. But as we mentioned with Java and Python, they also have a compilation step. And there’s no rule saying that C can’t be interpreted. (There are C interpreters out there!) In short, it’s a bunch of gray areas. Compilation in general is just taking source code and turning it into another, more easily-executed form.

The C compiler is the program that does the compilation.

As we’ve already said, gcc is a compiler that’s installed on a lot of Unix-like operating systems ²⁴. And it’s commonly run from the command line in a terminal, but not always. You can run it from your IDE, as well.

But we’ll do some command line examples here because there are too many IDEs to cover. Search the Internet for your IDE and “how to compile C” for more information.

So how do we do command line builds?

2.4 Building with `gcc`

If you have a source file called hello.c in the current directory, you can build that into a program called hello with this command typed in a terminal:

    gcc -o hello hello.c

The -o means “output to this file”²⁵. And there’s hello.c at the end, the name of the file we want to compile.

If your source is broken up into multiple files, you can compile them all together (almost as if they were one file, but the rules are actually more complex than that) by putting all the .c files on the command line:

    gcc -o awesomegame ui.c characters.c npc.c items.c

and they’ll all get built together into a big executable.

That’s enough to get started—later we’ll talk details about multiple source files, object files, and all kinds of fun stuff.

2.5 C Versions

C has come a long way over the years, and it had many named version numbers to describe which dialect of the language you’re using.

These generally refer to the year of the specification.

The most famous are C89, C99, and C11. We’ll focus on the latter in this book.

But here’s a more complete table:

Version	Description
K&R C	1978, the original. Named after Brian Kernighan and Dennis Ritchie. Ritchie designed and coded the language, and Kernighan co-authored the book on it. You rarely see original K&R code today. If you do, it’ll look odd, like Middle English looks odd to modern English readers.
C89, ANSI C, C90	In 1989, the American National Standards Institute (ANSI) produced a C language specification that set the tone for C that persists to this day. A year later, the reins were handed to the International Organization for Standardization (ISO) that produced the identical C90.
C95	A rarely-mentioned addition to C89 that included wide character support.
C99	The first big overhaul with lots of language additions. The thing most people will rememeber is the addition of `//`-style comments. This is the most popular version of C in use as of this writing.
C11	This major version update includes Unicode support and multi-threading. Be advised that if you start using these language features, you might be sacrificing portability with places that are stuck in C99 land. But, honestly, 1999 is getting to be a while back now.
C17, C18	Bugfix update to C11. C17 seems to be the official name, but the publication was delayed until 2018. As far as I can tell, these two are interchangeable, with C17 being preferred.
C2x	What’s coming next! Expected to eventually become C21.

You can force GCC to use one of these standards with the -std= command line argument. If you want it to be picky about the standard, add -pedantic.

For example:

    gcc -std=c99 -pedantic foo.c

For this book, I compile programs for C18 with all warnings set:

    gcc -Wall -Wextra -std=c18 -pedantic foo.c

3 Variables and Statements

“It takes all kinds to make a world, does it not, Padre?”

“So it does, my son, so it does.”

—Pirate Captain Thomas Bartholomew Red to the Padre, Pirates

There sure can be lotsa stuff in a C program.

Yup.

And for various reasons, it’ll be easier for all of us if we classify some of the types of things you can find in a program, so we can be clear what we’re talking about.

3.1 Variables

It’s said that “variables hold values”. But another way to think about it is that a variable is a human-readable name that refers to some data in memory.

We’re going to take a second here and take a peek down the rabbit hole that is pointers. Don’t worry about it.

You can think of memory as a big array of bytes²⁶ Data is stored in this “array”²⁷. If a number is larger than a single byte, it is stored in multiple bytes. Because memory is like an array, each byte of memory can be referred to by its index. This index into memory is also called an address, or a location, or a pointer.

When you have a variable in C, the value of that variable is in memory somewhere, at some address. Of course. After all, where else would it be? But it’s a pain to refer to a value by its numeric address, so we make a name for it instead, and that’s what the variable is.

The reason I’m bringing all this up is twofold:

It’s going to make it easier to understand pointers later.
Also, it’s going to make it easier to understand pointers later.

So a variable is a name for some data that’s stored in memory at some address.

3.1.1 Variable Names

You can use any characters in the range 0-9, A-Z, a-z, and underscore for variable names, with the following rules:

You can’t start a variable with a digit 0-9.
You can’t start a variable name with two underscores.
You can’t start a variable name with an underscore followed by a capital A-Z.

For Unicode, things get a little different, but the basic idea is that you can start or continue the variable name with one of the characters listed in C11 §D.1, and you can continue but not start a variable name with any of the characters listed in C11 §D.2.

Since those are just number ranges, I’m not going to reproduce them here. If you’re in an environment that supports Unicode, just try it and see if it works.

Just don’t start a variable name with the “Combining Left Harpoon Above” character and you’ll be fine.

3.1.2 Variable Types

Depending on which languages you already have in your toolkit, you might or might not be familiar with the idea of types. But C’s kinda picky about them, so we should do a refresher.

Some example types:

Type	Example	C Type
Integer	`3490`	`int`
Floating point	`3.14159`	`float`
Character (single)	`'c'`	`char`
String	`"Hello, world!"`	`char *`²⁸

C makes an effort to convert automatically between most numeric types when you ask it to. But other than that, all conversions are manual, notably between string and numeric.

Almost all of the types in C are variants on these types.

Before you can use a variable, you have to declare that variable and tell C what type the variable holds. Once declared, the type of variable cannot be changed later at runtime. What you set it to is what it is until it falls out of scope and is reabsorbed into the universe.

Let’s take our previous “Hello, world” code and add a couple variables to it:

#include <stdio.h>

int main(void)
{
    int i;    /* holds signed integers, e.g. -3, -2, 0, 1, 10 */
    float f;  /* holds signed floating point numbers, e.g. -3.1416 */

    printf("Hello, World!\n"); /* ah, blessed familiarity */
}

There! We’ve declared a couple of variables. We haven’t used them yet, and they’re both uninitialized. One holds an integer number, and the other holds a floating point number (a real number, basically, if you have a math background).

Uninitialized variables have indeterminate value²⁹. They have to be initialized or else you must assume they contain some nonsense number.

This is one of the places C can “get you”. Much of the time, in my experience, the indeterminate value is zero… but it can vary from run to run! Never assume the value will be zero, even if you see it is. Always explicitly initialize variables to some value before you use them!

What’s this? You want to store some numbers in those variables? Insanity!

Let’s go ahead and do that:

int main(void)
{
    int i;

    i = 2; // Assign the value 2 into the variable i

    printf("Hello, World!\n");
}

Killer. We’ve stored a value. Let’s print it.

We’re going to do that by passing two amazing parameters to the printf() function. The first argument is a string that describes what to print and how to print it (called the format string), and the second is the value to print, namely whatever is in the variable i.

printf() hunts through the format string for a variety of special sequences which start with a percent sign (%) that tell it what to print. For example, if it finds a %d, it looks to the next parameter that was passed, and prints it out as an integer. If it finds a %f, it prints the value out as a float. If it finds a %s, it prints a string.

As such, we can print out the value of various types like so:

int main(void)
{
    int i = 2;
    float f = 3.14;
    char *s = "Hello, world!";  // char * ("char pointer") is the string type

    printf("%s  i = %d and f = %f!\n", s, i, f);
}

And the output will be:

    Hello, world!  i = 2 and f = 3.14!

In this way, printf() might be similar to various types of format or parameterized strings in other languages you’re familiar with.

3.1.3 Boolean Types

C has Boolean types, true or false?

Historically, C didn’t have a Boolean type, and some might argue it still doesn’t.

In C, 0 means “false”, and non-zero means “true”.

So 1 is true. And 37 is true. And 0 is false.

You can just declare Boolean types as ints:

    int x = 1;
    
    if (x) {
        printf("x is true!\n");
    }

If you #include <stdbool.h>, you also get access to some symbolic names that might make things look more familiar, namely a bool type and true and false values:

#include <stdio.h>
#include <stdbool.h>

int main(void) {
    bool x = true;

    if (x) {
        printf("x is true!\n");
    }
}

But these are identical to using integer values for true and false. They’re just a facade to make things look nice.

3.2 Operators and Expressions

C operators should be familiar to you from other languages. Let’s blast through some of them here.

(There are a bunch more details than this, but we’re going to do enough in this section to get started.)

3.2.1 The `sizeof` Operator

This operator tells you the size (in bytes) that a particular variable or data type uses in memory.

More particularly, it tells you the size (in bytes) that the type of a particular expression (which might be just a single variable) uses in memory.

This can be different on different systems, except for char (which is always 1 byte).

And this might not seem very useful now, but we’ll be making reference to it here and there, so it’s worth covering.

You can take the sizeof a variable or expression:

    int a = 999;
    
    // %zu is the format specifier for type size_t ("t" is for "type", but
    // it's pronounced "size tee"), which is what is returned by sizeof.
    // More on size_t later.
    
    printf("%zu", sizeof a);      // Prints 4 on my system
    printf("%zu", sizeof(2 + 7)); // Prints 4 on my system
    printf("%zu", sizeof 3.14);   // Prints 8 on my system

Remember: it’s the size in bytes of the type of the expression, not the size of the expression itself. That’s why the size of 2+7 is the same as the size of a—they’re both type int. We’ll revisit this number 4 in the very next block of code…

…Where we’ll see you can take the sizeof a type (note the parentheses are required around a type name, unlike an expression):

    printf("%zu", sizeof(int));   // Prints 4 on my system
    printf("%zu", sizeof(char));  // Prints 1 on all systems

It’s important to note that sizeof is a compile-time operation³⁰. The result of the expression is determined entirely at compiletime, not at runtime.

We’ll make use of this later on.

3.2.2 Arithmetic

Hopefully these are familiar:

    i = i + 3;  // addition (+) and assignment (=) operators, add 3 to i
    i = i - 8;  // subtraction, subtract 8 from i
    i = i * 9;  // multiplication
    i = i / 2;  // division
    i = i % 5;  // modulo (division remainder)

There are shorthand variants for all of the above. Each of those lines could more tersely be written as:

    i += 3;  // Same as "i = i + 3", add 3 to i
    i -= 8;  // Same as "i = i - 8"
    i *= 9;  // Same as "i = i * 9"
    i /= 2;  // Same as "i = i / 2"
    i %= 5;  // Same as "i = i % 5"

There is no exponentiation. You’ll have to use one of the pow() function variants from math.h.

Let’s get into some of the weirder stuff you might not have in your other languages!

3.2.3 Ternary Operator

C also includes the ternary operator. This is an expression whose value depends on the result of a conditional embedded in it.

    // If x > 10, add 17 to y. Otherwise add 37 to y.
    
    y += x > 10? 17: 37;

What a mess! You’ll get used to it the more you read it. To help out a bit, I’ll rewrite the above expression using if statements:

    // This expression:
    
    y += x > 10? 17: 37;
    
    // is equivalent to this non-expression:
    
    if (x > 10)
        y += 17;
    else
        y += 37;

Or, another example that prints if a number stored in x is odd or even:

    printf("The number %d is %s.\n", x, x % 2 == 0?"even": "odd")

The %s format specifier in printf() means print a string. If the expression x % 2 evaluates to 0, the value of the entire ternary expression evaluates to the string "even". Otherwise it evaluates to the string "odd". Pretty cool!

It’s important to note that the ternary operator isn’t flow control like the if statement is. It’s just an expression that evaluates to a value.

3.2.4 Pre-and-Post Increment-and-Decrement

Now, let’s mess with another thing that you might not have seen.

These are the legendary post-increment and post-decrement operators:

    i++;        // Add one to i (post-increment)
    i--;        // Subtract one from i (post-decrement)

Very commonly, these are just used as shorter versions of:

    i += 1;        // Add one to i
    i -= 1;        // Subtract one from i

but they’re more subtly different than that, the clever scoundrels.

Let’s take a look at this variant, pre-increment and pre-decrement:

    ++i;        // Add one to i (pre-increment)
    --i;        // Subtract one from i (pre-decrement)

With pre-increment and pre-decrement, the value of the variable is incremented or decremented before the expression is evaluated. Then the expression is evaluated with the new value.

With post-increment and post-decrement, the value of the expression is first computed with the value as-is, and then the value is incremented or decremented after the value of the expression has been determined.

You can actually embed them in expressions, like this:

    i = 10;
    j = 5 + i++;  // Compute 5 + i, _then_ increment i
    
    printf("%d, %d\n", i, j);  // Prints 11, 15

Let’s compare this to the pre-increment operator:

    i = 10;
    j = 5 + ++i;  // Increment i, _then_ compute 5 + i
    
    printf("%d, %d\n", i, j);  // Prints 11, 16

This technique is used frequently with array and pointer access and manipulation. It gives you a way to use the value in a variable, and also increment or decrement that value before or after it is used.

But by far the most common place you’ll see this is in a for loop:

    for (i = 0; i < 10; i++)
        printf("i is %d\n");

But more on that later.

3.2.5 The Comma Operator

This is an uncommonly-used way to separated expressions that will run left to right:

    x = 10, y = 20;  // First assign 10 to x, then 20 to y

Seems a bit silly, since you could just replace the comma with a semicolon, right?

    x = 10; y = 20;  // First assign 10 to x, then 20 to y

But that’s a little different. The latter is two separate expressions, while the former is a single expression!

With the comma operator, the value of the comma expression is the value of the rightmost expression:

    x = 1, 2, 3;
    
    printf("x is %d\n", x);  // Prints 3, because 3 is rightmost in the comma list

But even that’s pretty contrived. One common place the comma operator is used is in for loops to do multiple things in each section of the statement:

    for (i = 0, j = 10; i < 100; i++, j++)
        printf("%d, %d\n", i, j);

We’ll revisit that later.

3.2.6 Conditional Operators

For Boolean values, we have a raft of standard operators:

    a == b;  // True if a is equivalent to b
    a != b;  // True if a is not equivalent to b
    a < b;   // True if a is less than b
    a > b;   // True if a is greater than b
    a <= b;  // True if a is less than or equal to b
    a >= b;  // True if a is greater than or equal to b

Don’t mix up assignment = with comparison ==! Use two equals to compare, one to assign.

We can use the comparison expressions with if statements:

    if (a <= 10)
        printf("Success!\n");

3.2.7 Boolean Operators

We can chain together or alter conditional expressions with Boolean operators for and, or, and not.

Operator	Boolean meaning
`&&`	and
`\|\|`	or
`!`	not

An example of Boolean “and”:

    // Do something if x less than 10 and y greater than 20:
    
    if (x < 10 && y > 20)
        printf("Doing something!\n");

An example of Boolean “not”:

    if (!(x < 12))
        printf("x is not less than 12\n");

! has higher precedence than the other Boolean operators, so we have to use parentheses in that case.

Of course, that’s just the same as:

    if (x >= 12)
        printf("x is not less than 12\n");

but I needed the example!

3.3 Flow Control

Booleans are all good, but of course we’re nowhere if we can’t control program flow. Let’s take a look at a number of constructs: if, for, while, and do-while.

First, a general forward-looking note about statements and blocks of statements brought to you by your local friendly C developer:

After something like an if or while statement, you can either put a single statement to be executed, or a block of statements to all be executed in sequence.

Let’s start with a single statement:

    if (x == 10) printf("x is 10");

This is also sometimes written on a separate line. (Whitespace is largely irrelevant in C—it’s not like Python.)

    if (x == 10)
        printf("x is 10\n");

But what if you want multiple things to happen due to the conditional? You can use squirrelly braces to mark a block or compound statement.

    if (x == 10) {
        printf("x is 10\n");
        printf("And also this happens when x is 10\n");
    }

It’s a really common style to always use squirrelly braces even if they aren’t necessary:

    if (x == 10) {
        printf("x is 10\n");
    }

Some devs feel the code is easier to read and avoids errors like this where things visually look like they’re in the if block, but actually they aren’t.

    // BAD ERROR EXAMPLE
    
    if (x == 10)
        printf("x is 10\n");
        printf("And also this happens ALWAYS\n");  // Surprise!! Unconditional!

while and for and the other looping constructs work the same way as the examples above. If you want to do multiple things in a loop or after an if, wrap them up in squirrelly braces.

In other words, the if is going to run the one thing after the if. And that one thing can be a single statement or a block of statements.

3.3.1 The `if` statement

We’ve already been using if for multiple examples, since it’s likely you’ve seen it in a language before, but here’s another:

    int i = 10;
    
    if (i > 10) {
        printf("Yes, i is greater than 10.\n");
        printf("And this will also print if i is greater than 10.\n");
    }
    
    if (i <= 10) printf("i is less than or equal to 10.\n");

In the example code, the message will print if i is greater than 10, otherwise execution continues to the next line. Notice the squirrley braces after the if statement; if the condition is true, either the first statement or expression right after the if will be executed, or else the collection of code in the squirlley braces after the if will be executed. This sort of code block behavior is common to all statements.

3.3.2 The `while` statement

while is your average run-of-the-mill looping construct. Do a thing while a condition expression is true.

Let’s do one!

    // print the following output:
    //
    //   i is now 0!
    //   i is now 1!
    //   [ more of the same between 2 and 7 ]
    //   i is now 8!
    //   i is now 9!
    
    i = 0;
    
    while (i < 10) {
        printf("i is now %d!\n", i);
        i++;
    }
    
    printf("All done!\n");

That gets you a basic loop. C also has a for loop which would have been cleaner for that example.

A not-uncommon use of while is for infinite loops where you repeat while true:

    while (1) {
        printf("1 is always true, so this repeats forever.\n");
    }

3.3.3 The `do-while` statement

So now that we’ve gotten the while statement under control, let’s take a look at its closely related cousin, do-while.

They are basically the same, except if the loop condition is false on the first pass, do-while will execute once, but while won’t execute at all. Let’s see by example:

    /* using a while statement: */
    
    i = 10;
    
    // this is not executed because i is not less than 10:
    while(i < 10) {
        printf("while: i is %d\n", i);
        i++;
    }
    
    /* using a do-while statement: */
    
    i = 10;
    
    // this is executed once, because the loop condition is not checked until
    // after the body of the loop runs:
    
    do {
        printf("do-while: i is %d\n", i);
        i++;
    } while (i < 10);
    
    printf("All done!\n");

Notice that in both cases, the loop condition is false right away. So in the while, the loop fails, and the following block of code is never executed. With the do-while, however, the condition is checked after the block of code executes, so it always executes at least once. In this case, it prints the message, increments i, then fails the condition, and continues to the “All done!” output.

The moral of the story is this: if you want the loop to execute at least once, no matter what the loop condition, use do-while.

All these examples might have been better done with a for loop. Let’s do something less deterministic—repeat until a certain random number comes up!

#include <stdio.h>   // For printf
#include <stdlib.h>  // For rand

int main(void)
{
    int r;

    do {
        r = rand() % 100; // Get a random number between 0 and 99
        printf("%d\n", r);
    } while (r != 37);    // Repeat until 37 comes up
}

3.3.4 The `for` statement

Welcome to one of the most popular loops in the world! The for loop!

This is a great loop if you know the number of times you want to loop in advance.

You could do the same thing using just a while loop, but the for loop can help keep the code cleaner.

Here are two pieces of equivalent code—note how the for loop is just a more compact representation:

    // Print numbers between 0 and 9, inclusive...
    
    // Using a while statement:
    
    i = 0;
    while (i < 10) {
        printf("i is %d\n", i);
        i++;
    }
    
    // Do the exact same thing with a for-loop:
    
    for (i = 0; i < 10; i++) {
        printf("i is %d\n", i);
    }

That’s right, folks—they do exactly the same thing. But you can see how the for statement is a little more compact and easy on the eyes. (JavaScript users will fully appreciate its C origins at this point.)

It’s split into three parts, separated by semicolons. The first is the initialization, the second is the loop condition, and the third is what should happen at the end of the block if the loop condition is true. All three of these parts are optional.

    for (initialize things; loop if this is true; do this after each loop)

Note that the loop will not execute even a single time if the loop condition starts off false.

for-loop fun fact!

You can use the comma operator to do multiple things in each clause of the for loop!
for (i = 0, j = 999; i < 10; i++, j--) {
    printf("%d, %d\n", i, j);
}

An empty for will run forever:

    for(;;) {  // "forever"
        printf("I will print this again and again and again\n" );
        printf("for all eternity until the cold-death of the universe.\n");
    }

3.3.5 The `switch` Statement

Depending on what languages you’re coming from, you might or might not be familiar with switch, or C’s version might even be more restrictive than you’re used to. This is a statement that allows you to take a variety of actions depending on the value of an integer expression.

Basically, it evaluates an expression to an integer value, jumps to the case that corresponds to that value. Execution resumes from that point. If a break statement is encountered, then execution jumps out of the switch.

Let’s do an example where the user enters a number of goats and we print out a gut-feel of how many goats that is.

#include <stdio.h>

int main(void)
{
    int goat_count;

    printf("Enter a goat count: ");
    scanf("%d", &goat_count);       // Read an integer from the keyboard

    switch (goat_count) {
        case 0:
            printf("You have no goats.\n");
            break;

        case 1:
            printf("You have a singular goat.\n");
            break;

        case 2:
            printf("You have a brace of goats.\n");
            break;

        default:
            printf("You have a bona fide plethora of goats!\n");
            break;
    }
}

In that example, if the user enters, say, 2, the switch will jump to the case 2 and execute from there. When (if) it hits a break, it jumps out of the switch.

Also, you might see that default label there at the bottom. This is what happens when no cases match.

Every case, including default, is optional. And they can occur in any order, but it’s really typical for default, if any, to be listed last.

So the whole thing acts like an if-else cascade:

    if (goat_count == 0)
        printf("You have no goats.\n");
    else if (goat_count == 1)
        printf("You have a singular goat.\n");
    else if (goat_count == 2)
        printf("You have a brace of goats.\n");
    else:
        printf("You have a bona fide plethora of goats!\n");

With some key differences:

switch is often faster to jump to the correct code (though the spec makes no such guarantee).
if-else can do things like relational conditionals like < and >= and floating point and other types, while switch cannot.

There’s one more neat thing about switch that you sometimes see that is quite interesting: fall through.

Remember how break causes us to jump out of the switch?

Well, what happens if we don’t break?

Turns out we just keep on going into the next case! Demo!

    switch (x) {
        case 1:
            printf("1\n");
            // fall through!
        case 2:
            printf("2\n");
            break;
        case 3:
            printf("3\n");
            break;
    }

If x == 1, this switch will first hit case 1, it’ll print the 1, but then it just continues on to the next line of code… which prints 2!

And then, at last, we hit a break so we jump out of the switch.

if x == 2, then we just it the case 2, print 2, and break as normal.

Not having a break is called fall through.

ProTip: ALWAYS put a comment in the code where you intend to fall through, like I did above. It will save other programmers from wondering if you meant to do that.

In fact, this is one of the common places to introduce bugs in C programs: forgetting to put a break in your case. You gotta do it if you don’t want to just roll into the next case³¹.

4 Functions

Very much like other languages you’re used to, C has the concept of functions.

Functions can accept a variety of arguments and return a value. One important thing, though: the arguments and return value types are predeclared—because that’s how C likes it!

Let’s take a look at a function. This is a function that takes an int as an argument, and returns an int.

int plus_one(int n)  // The "definition"
{
    return n + 1;
}

The int before the plus_one indicates the return type.

The int n indicates that this function takes one int argument, stored in parameter n.

Continuing the program down into main(), we can see the call to the function, where we assign the return value into local variable j:

int main(void)
{
    int i = 10, j;
    
    j = plus_one(i);  // The "call"

    printf("i + 1 is %d\n", j);
}

Before I forget, notice that I defined the function before I used it. If hadn’t done that, the compiler wouldn’t know about it yet when it compiles main() and it would have given an unknown function call error. There is a more proper way to do the above code with function prototypes, but we’ll talk about that later.

Also notice that main() is a function!

It returns an int.

But what’s this void thing? This is a keyword that’s used to indicate that the function accepts no arguments.

You can also return void to indicate that you don’t return a value:

// This function takes no parameters and returns no value:

void hello(void)
{
    printf("Hello, world!\n");
}

int main(void)
{
    hello();  // Prints "Hello, world!"
}

4.1 Passing by Value

When you pass a value to a function, a copy of that value gets made in this magical mystery world known as the stack³². (The stack is just a hunk of memory somewhere that the program allocates memory on. Some of the stack is used to hold the copies of values that are passed to functions.)

For now, the important part is that a copy of the variable or value is being passed to the function. The practical upshot of this is that since the function is operating on a copy of the value, you can’t affect the value back in the calling function directly. Like if you wanted to increment a value by one, this would NOT work:

void increment(int a)
{
    a++;
}

int main(void)
{
    int i = 10;

    increment(i);
}

You might somewhat sensibly think that the value of i after the call would be 11, since that’s what the ++ does, right? This would be incorrect. What is really happening here?

Well, when you pass i to the increment() function, a copy gets made on the stack, right? It’s the copy that increment() works on, not the original; the original i is unaffected. We even gave the copy a name: a, right? It’s right there in the parameter list of the function definition. So we increment a, sure enough, but what good does that do us out in main() ? None! Ha!

That’s why in the previous example with the plus_one() function, we returned the locally modified value so that we could see it again in main().

Seems a little bit restrictive, huh? Like you can only get one piece of data back from a function, is what you’re thinking. There is, however, another way to get data back; C folks call it passing by reference. But no fancy-schmancy name will distract you from the fact that EVERYTHING you pass to a function WITHOUT EXCEPTION is copied onto the stack and the function operates on that local copy, NO MATTER WHAT. Remember that, even when we’re talking about this so-called passing by reference.

But that’s a story for another time.

4.2 Function Prototypes

So if you recall back in the ice age a few sections ago, I mentioned that you had to define the function before you used it, otherwise the compiler wouldn’t know about it ahead of time, and would bomb out with an error.

This isn’t quite strictly true. You can notify the compiler in advance that you’ll be using a function of a certain type that has a certain parameter list and that way the function can be defined anywhere at all, as long as the function prototype has been declared first.

Fortunately, the function prototype is really quite easy. It’s merely a copy of the first line of the function definition with a semicolon tacked on the end for good measure. For example, this code calls a function that is defined later, because a prototype has been declared first:

int foo(void);  // This is the prototype!

int main(void)
{
    int i;
    
    i = foo();
}

int foo(void)  // this is the definition, just like the prototype!
{
    return 3490;
}

You might notice something about the sample code we’ve been using…that is, we’ve been using the good old printf() function without defining it or declaring a prototype! How do we get away with this lawlessness? We don’t, actually. There is a prototype; it’s in that header file stdio.h that we included with #include, remember? So we’re still legit, officer!

4.3 Empty Parameter Lists

You might see these from time to time in older code, but you shouldn’t ever code one up in new code. Always use void to indicate that a function takes no parameters. There’s never³³ a reason to do this in modern code.

There are two contexts for this:

Omitting all parameters where the function is defined
Omitting all parameters in a prototype

Let’s look at a potential function definition first:

    void foo()  // Should really have a `void` in there
    {
        printf("Hello, world!\n");
    }

While the spec spells out that the behavior in this instance is as-if you’d indicated void (C11 §6.7.6.3¶14), the void type is there for a reason. Use it.

But in the case of a function prototype, there is a significant difference between using void and not:

    void foo();
    void foo(void);  // Not the same!

Leaving void out of the prototype indicates to the compiler that there is no additional information about the parameters to the function. It effectively turns off all that type checking.

With a prototype definitely use void when you have an empty parameter list.

5 Pointers—Cower In Fear!

Pointers are one of the most feared things in the C language. In fact, they are the one thing that makes this language challenging at all. But why?

Because they, quite honestly, can cause electric shocks to come up through the keyboard and physically weld your arms permanently in place, cursing you to a life at the keyboard in this language from the 70s!

Well, not really. But they can cause huge headaches if you don’t know what you’re doing when you try to mess with them.

Depending on what language you came from, you might already understand the concept of references, where a variable refers to an object of some type.

This is very much the same, except we have to be more explicit with C about when we’re talking about the reference or the thing it refers to.

5.1 Memory and Variables

Computer memory holds data of all kinds, right? It’ll hold floats, ints, or whatever you have. To make memory easy to cope with, each byte of memory is identified by an integer. These integers increase sequentially as you move up through memory. You can think of it as a bunch of numbered boxes, where each box holds a byte³⁴ of data. Or like a big array where each element holds a byte, if you come from a language with arrays. The number that represents each box is called its address.

Now, not all data types use just a byte. For instance, an int is often four bytes, as is a float, but it really depends on the system. You can use the sizeof operator to determine how many bytes of memory a certain type uses.

    // %zu is the format specifier for type size_t ("t" is for "type", but
    // it's pronounced "size tee"), which is what is returned by sizeof.
    // More on size_t later.
    
    printf("an int uses %zu bytes of memory\n", sizeof(int));
    
    // That prints "4" for me, but can vary by system.

When you have a data type that uses more than a byte of memory, the bytes that make up the data are always adjacent to one another in memory. Sometimes they’re in order, and sometimes they’re not³⁵, but that’s platform-dependent, and often taken care of for you without you needing to worry about pesky byte orderings.

So anyway, if we can get on with it and get a drum roll and some forboding music playing for the definition of a pointer, a pointer is a variable that holds an address. Imagine the classical score from 2001: A Space Odessey at this point. Ba bum ba bum ba bum BAAAAH!

Ok, so maybe a bit overwrought here, yes? There’s not a lot of mystery about pointers. They are the address of data. Just like an int variable can hold the value 12, a pointer variable can hold the address of data.

This means that all these things mean the same thing, i.e. a number that represents a point in memory:

Index into memory (if you’re thinking of memory like a big array)
Address
Location

I’m going to use these interchangeably. And yes, I just threw location in there because you can never have enough words that mean the same thing.

And a pointer variable holds that address number. Just like a float variable might hold 3.14159.

Often, we like to store the address of some data that we have stored in a variable, as opposed to any old random address out in memory wherever.

When we have that, we say we have a “pointer to” some data.

So if we have an int, say, and we want a pointer to it, what we want is some way to get the address of that int, right? After all, the pointer just holds the address of the data. What operator do you suppose we’d use to find the address of the int?

Well, by a shocking surprise that must come as something of a shock to you, gentle reader, we use the address-of operator (which happens to be an ampersand: “&”) to find the address of the data. Ampersand.

So for a quick example, we’ll introduce a new format specifier for printf() so you can print a pointer. You know already how %d prints a decimal integer, yes? Well, %p prints a pointer. Now, this pointer is going to look like a garbage number (and it might be printed in hexadecimal³⁶ instead of decimal), but it is merely the index into memory the data is stored in. (Or the index into memory that the first byte of data is stored in, if the data is multi-byte.) In virtually all circumstances, including this one, the actual value of the number printed is unimportant to you, and I show it here only for demonstration of the address-of operator.

#include <stdio.h>

int main(void)
{
    int i = 10;

    printf("The value of i is %d, and its address is %p\n", i, &i);
}

On my computer, this prints:

    The value of i is 10, and its address is 0x7ffda2546fc4

If you’re curious, that hexadecimal number is 140,727,326,896,068 in decimal (base 10 just like Grandma used to use). That’s the index into memory where the variable i’s data is stored. It’s the address of i. It’s the location of i. It’s a pointer to i.

It’s a pointer because it lets you know where i is in memory. Like a home address written on a scrap of paper tells you where you can find a particular house, this number indicates to us where in memory we can find the value of i. It points to i.

Again, we don’t really care what the address’s exact number is, generally. We just care that it’s a pointer to i.

5.2 Pointer Types

Well, this is all well and good. You can now successfully take the address of a variable and print it on the screen. There’s a little something for the ol’ resume, right? Here’s where you grab me by the scruff of the neck and ask politely what the frick pointers are good for.

Excellent question, and we’ll get to that right after these messages from our sponsor.

ACME ROBOTIC HOUSING UNIT CLEANING SERVICES. YOUR HOMESTEAD WILL BE DRAMATICALLY IMPROVED OR YOU WILL BE TERMINATED. MESSAGE ENDS.

Welcome back to another installment of Beej’s Guide to Whatever. When we met last we were talking about how to make use of pointers. Well, what we’re going to do is store a pointer off in a variable so that we can use it later. You can identify the pointer type because there’s an asterisk (*) before the variable name and after its type:

int main(void)
{
    int i;  // i's type is "int"
    int *p; // p's type is "pointer to an int", or "int-pointer"
}

Hey, so we have here a variable that is a pointer itself, and it can point to other ints. That it, is can hold the address of other ints. We know it points to ints, since it’s of type int* (read “int-pointer”).

When you do an assignment into a pointer variable, the type of the right hand side of the assignment has to be the same type as the pointer variable. Fortunately for us, when you take the address-of a variable, the resultant type is a pointer to that variable type, so assignments like the following are perfect:

    int i;
    int *p;  // p is a pointer, but is uninitialized and points to garbage
    
    p = &i;  // p is assigned the address of i--p now "points to" i

On the left of the assignment, we have a variable of type pointer-to-int (int*), and on the right side, we have expression of type pointer-to-int since i is an int (because address-of int gives you a pointer to int). The address of a thing can be stored in a pointer to that thing.

Get it? I know is still doesn’t quite make much sense since you haven’t seen an actual use for the pointer variable, but we’re taking small steps here so that no one gets lost. So now, let’s introduce you to the anti-address-of, operator. It’s kind of like what address-of would be like in Bizarro World.

5.3 Dereferencing

A pointer variable can be thought of as referring to another variable by pointing to it. It’s rare you’ll hear anyone in C land talking about “referring” or “references”, but I bring it up just so that the name of this operator will make a little more sense.

When you have a pointer to a variable (roughly “a reference to a variable”), you can use the original variable through the pointer by dereferencing the pointer. (You can think of this as “de-pointering” the pointer, but no one ever says “de-pointering”.)

What do I mean by “get access to the original variable”? Well, if you have a variable called i, and you have a pointer to i called p, you can use the dereferenced pointer p exactly as if it were the original variable i!

You almost have enough knowledge to handle an example. The last tidbit you need to know is actually this: what is the dereference operator? It is the asterisk, again: *. Now, don’t get this confused with the asterisk you used in the pointer declaration, earlier. They are the same character, but they have different meanings in different contexts³⁷.

Here’s a full-blown example:

#include <stdio.h>

int main(void)
{
    int i;
    int *p;  // this is NOT a dereference--this is a type "int*"

    p = &i;  // p now points to i, p holds address of i

    i = 10;  // i is now 10
    *p = 20; // i (yes i!) is now 20!!

    printf("i is %d\n", i);   // prints "20"
    printf("i is %d\n", *p);  // "20"! dereference-p is the same as i!
}

Remember that p holds the address of i, as you can see where we did the assignment to p on line 8. What the dereference operator does is tells the computer to use the object the pointer points to instead of using the pointer itself. In this way, we have turned *p into an alias of sorts for i.

Great, but why? Why do any of this?

5.4 Passing Pointers as Parameters

Right about now, you’re thinking that you have an awful lot of knowledge about pointers, but absolutely zero application, right? I mean, what use is *p if you could just simply say i instead?

Well, my feathered friend, the real power of pointers comes into play when you start passing them to functions. Why is this a big deal? You might recall from before that you could pass all kinds of parameters to functions and they’d be dutifully copied onto the stack, and then you could manipulate local copies of those variables from within the function, and then you could return a single value.

What if you wanted to bring back more than one single piece of data from the function? I mean, you can only return one thing, right? What if I answered that question with another question? …Er, two questions?

What happens when you pass a pointer as a parameter to a function? Does a copy of the pointer get put on the stack? You bet your sweet peas it does. Remember how earlier I rambled on and on about how EVERY SINGLE PARAMETER gets copied onto the stack and the function uses a copy of the parameter? Well, the same is true here. The function will get a copy of the pointer.

But, and this is the clever part: we will have set up the pointer in advance to point at a variable…and then the function can dereference its copy of the pointer to get back to the original variable! The function can’t see the variable itself, but it can certainly dereference a pointer to that variable!

This is analogous to writing a home address on a piece of paper, and then copying that onto another piece of paper. You now have two pointers to that house. In the case of a function call. one of the copies is stored in a pointer variable out in the calling scope, and the other is stored in a pointer variable that is the parameter of the function.

Example!

#include <stdio.h>

void increment(int *p)  // note that it accepts a pointer to an int
{
    *p = *p + 1;        // add one to the thing p points to
}

int main(void)
{
    int i = 10;
    int *j = &i;  // note the address-of; turns it into a pointer to i

    printf("i is %d\n", i);        // prints "10"
    printf("i is also %d\n", *j);  // prints "10"

    increment(j);                  // j is an int*--to i

    printf("i is %d\n", i);        // prints "11"!
}

Ok! There are a couple things to see here…not the least of which is that the increment() function takes an int* as a parameter. We pass it an int* in the call by changing the int variable i to an int* using the address-of operator. (Remember, a pointer holds an address, so we make pointers to variables by running them through the address-of operator.)

The increment() function gets a copy of the pointer on the stack. Both the original pointer j (in main()) and the copy of that pointer p (the parameter in increment()) point to the same address, namely the one holding the value i. (Again, by analogy, like two pieces of paper with the same home address written on them.) Dereferencing either will allow you to modify the original variable i! The function can modify a variable in another scope! Rock on!

The above example is often more concisely written in the call just by using address-of right in the argument list:

    printf("i is %d\n", i);  // prints "10"
    increment(&i);
    printf("i is %d\n", i);  // prints "11"!

Pointer enthusiasts will recall from early on in the guide, we used a function to read from the keyboard, scanf()…and, although you might not have recognized it at the time, we used the address-of to pass a pointer to a value to scanf(). We had to pass a pointer, see, because scanf() reads from the keyboard (typically) and stores the result in a variable. The only way it can see that variable that is local to that calling function is if we pass a pointer to that variable:

    int i = 0;
    
    scanf("%d", &i);         // pretend you typed "12"
    printf("i is %d\n", i);  // prints "i is 12"

See, scanf() dereferences the pointer we pass it in order to modify the variable it points to. And now you know why you have to put that pesky ampersand in there!

5.5 The `NULL` Pointer

Any pointer variable of any pointer type can be set to a special value called NULL. This indicates that this pointer doesn’t point to anything.

    int *p;
    
    p = NULL;

Since it doesn’t point to a value, dereferencing it is undefined behavior, and probably will result in a crash:

    int *p = NULL;
    
    *p = 12;  // CRASH or SOMETHING PROBABLY BAD. BEST AVOIDED.

Despite being called the billion dollar mistake by its creator ³⁸, the NULL pointer is a good sentinel value ³⁹ and general indicator that a pointer hasn’t yet been initialized.

(Of course, the pointer points to garbage unless you explicitly assign it to point to an address or NULL.)

5.6 A Note on Declaring Pointers

The syntax for declaring a pointer can get a little weird. Let’s look at this example:

    int a;
    int b;

We can condense that into a single line, right?

    int a, b;  // Same thing

So a and b are both ints. No problem.

But what about this?

    int a;
    int *p;

Can we make that into one line? We can. But where does the * go?

The rule is that the * goes in front of any variable that is a pointer type. That is. the * is not part of the int in this example. it’s a part of variable p.

With that in mind, we can write this:

    int a, *p;  // Same thing

It’s important to note that the following line does not declare two pointers:

    int *p, q;  // p is a pointer to an int; q is just an int.

This can be particularly insidious-looking if the programmer writes this following (valid) line of code which is functionally identical to the one above.

    int* p, q;  // p is a pointer to an int; q is just an int.

So take a look at this and determine which variables are pointers and which are not:

    int *a, b, c, *d, e, *f, g, h, *i;

I’ll drop the answer in a footnote⁴⁰.

5.7 `sizeof` and Pointers

Just a little bit of syntax here that might be confusing and you might see from time to time.

Recall that sizeof operates on the type of the expression.

    int *p;
    
    sizeof(int); // Returns size of an `int`
    sizeof p     // p is type int*, so returns size of `int*`
    sizeof *p    // *p is type int, so returns size of `int`

You might see code with that last sizeof in there. Just remember that sizeof is all about the type of the expression, not the variables in the expression themselves.

6 Arrays

Luckily, C has arrays. I mean, I know it’s considered a low-level language⁴¹ but it does at least have the concept of arrays built-in. And since a great many languages drew inspiration from C’s syntax, you’re probably already familiar with using [ and ] for declaring and using arrays in C.

But only barely! As we’ll find out later, arrays are just syntactic sugar in C—they’re actually all pointers and stuff deep down. Freak out! But for now, let’s just use them as arrays. Phew.

6.1 Easy Example

Let’s just crank out an example:

#include <stdio.h>

int main(void)
{
    int i;
    float f[4];  // Declare an array of 4 floats

    f[0] = 3.14159;  // Indexing starts at 0, of course.
    f[1] = 1.41421;
    f[2] = 1.61803;
    f[3] = 2.71828;

    // Print them all out:

    for (i = 0; i < 4; i++) {
        printf("%f\n", f[i]);
    }
}

When you declare an array, you have to give it a size. And the size has to be fixed⁴².

In the above example, we made an array of 4 floats. The value in the square brackets in the declaration lets us know that.

Later on in subsequent lines, we access the values in the array, setting them or getting them, again with square brackets.

Hopefully this looks familiar from languages you already know!

6.2 Getting the Length of an Array

You can’t…ish. C doesn’t record this information. You have to manage it separately in another variable.

When I say “can’t”, I actually mean there are some circumstances when you can. There is a trick to get the number of elements in an array in the scope in which an array is declared. But, generally speaking, this won’t work the way you want if you pass the array into a function.

Let’s take a look at this trick. The basic idea is that you take the sizeof the array, and then divide that by the size of each element to get the length. For example, if an int is 4 bytes, and the array is 32 bytes long, there must be room for 32/4 or 8 ints in there.

    int x[12];  // 12 ints
    
    printf("%zu\n", sizeof x);     // 48 total bytes
    printf("%zu\n", sizeof(int));  // 4 bytes per int
    
    printf("%zu\n", sizeof x / sizeof(int));  // 48/4 = 12 ints!

If it’s an array of chars, then sizeof the array is the number of elements, since sizeof(char) is defined to be 1. For anything else, you have to divide by the size of each element.

But this trick only works in the scope in which the array was defined. If you pass the array to a function, it doesn’t work. Even if you make it “big” in the function signature:

    void foo(int x[12])
    {
        printf("%zu\n", sizeof x);     // 8?! What happened to 48?
        printf("%zu\n", sizeof(int));  // 4 bytes per int
    
        printf("%zu\n", sizeof x / sizeof(int));  // 8/4 = 2 ints?? WRONG.
    }

This is because when you “pass” arrays to functions, you’re only passing a pointer to the first element, and that’s what sizeof measures. More on this in the Passing Single Dimensional Arrays to Functions section, below.

6.3 Array Initializers

You can initialize an array with constants ahead of time:

#include <stdio.h>

int main(void)
{
    int i;
    int a[5] = {22, 37, 3490, 18, 95};  // Initialize with these values

    for (i = 0; i < 5; i++) {
        printf("%d\n", a[i]);
    }
}

Catch: initializer values must be constant terms. Can’t throw variables in there. Sorry, Illinois!

You should never have more items in your initializer than there is room for in the array, or the compiler will get cranky:

    foo.c: In function ‘main’:
    foo.c:6:39: warning: excess elements in array initializer
        6 |     int a[5] = {22, 37, 3490, 18, 95, 999};
          |                                       ^~~
    foo.c:6:39: note: (near initialization for ‘a’)

But (fun fact!) you can have fewer items in your initializer than there is room for in the array. The remaining elements in the array will be automatically initialized with zero.

    int a[5] = {22, 37, 3490};
    
    // is the same as:
    
    int a[5] = {22, 37, 3490, 0, 0};

It’s a common shortcut to see this in an initializer when you want to set an entire array to zero:

    int a[100] = {0};

Which means, “Make the first element zero, and then automatically make the rest zero, as well.”

Lastly, you can also have C compute the size of the array from the initializer, just by leaving the size off:

    int a[3] = {22, 37, 3490};
    
    // is the same as:
    
    int a[] = {22, 37, 3490};  // Left the size off!

6.4 Out of Bounds!

C doesn’t stop you from accessing arrays out of bounds. It might not even warn you.

Let’s steal the example from above and keep printing off the end of the array. It only has 5 elements, but let’s try to print 10 and see what happens:

#include <stdio.h>

int main(void)
{
    int i;
    int a[5] = {22, 37, 3490, 18, 95};

    for (i = 0; i < 10; i++) {  // BAD NEWS: printing too many elements!
        printf("%d\n", a[i]);
    }
}

Running it on my computer prints:

Yikes! What’s that? Well, turns out printing off the end of an array results in what C developers call undefined behavior. We’ll talk more about this beast later, but for now it means, “You’ve done something bad, and anything could happen during your program run.”

And by anything, I mean typically things like finding zeroes, finding garbage numbers, or crashing. But really the C spec says in this circumstance the compiler is allowed to emit code that does anything⁴³.

Short version: don’t do anything that causes undefined behavior. Ever⁴⁴.

6.5 Multidimensional Arrays

You can add as many dimensions as you want to your arrays.

    int a[10];
    int b[2][7];
    int c[4][5][6];

These are stored in memory in row-major order ⁴⁵.

You an also use initializers on multidimensional arrays by nesting them:

#include <stdio.h>

int main(void)
{
    int row, col;

    int a[2][5] = {      // Initialize a 2D array
        {0, 1, 2, 3, 4},
        {5, 6, 7, 8, 9}
    };

    for (row = 0; row < 2; row++) {
        for (col = 0; col < 5; col++) {
            printf("(%d,%d) = %d\n", row, col, a[row][col]);
        }
    }
}

For output of:

    (0,0) = 0
    (0,1) = 1
    (0,2) = 2
    (0,3) = 3
    (0,4) = 4
    (1,0) = 5
    (1,1) = 6
    (1,2) = 7
    (1,3) = 8
    (1,4) = 9

6.6 Arrays and Pointers

[Casually] So… I kinda might have mentioned up there that arrays were pointers, deep down? We should take a shallow dive into that now so that things aren’t completely confusing. Later on, we’ll look at what the real relationship between arrays and pointers is, but for now I just want to look at passing arrays to functions.

6.6.1 Getting a Pointer to an Array

I want to tell you a secret. Generally speaking, when a C programmer talks about a pointer to an array, they’re talking about a pointer to the first element of the array⁴⁶.

So let’s get a pointer to the first element of an array.

#include <stdio.h>

int main(void)
{
    int a[5] = {11, 22, 33, 44, 55};
    int *p;

    p = &a[0];  // p points to the array
                // Well, to the first element, actually

    printf("%d\n", *p);  // Prints "11"
}

This is so common to do in C that the language allows us a shorthand:

    p = &a[0];  // p points to the array
    
    // is the same as:
    
    p = a;      // p points to the array, but much nicer-looking!

Just referring to the array name in isolation is the same as getting a pointer to the first element of the array! We’re going to use this extensively in the upcoming examples.

But hold on a second–isn’t p an int*? And *p gives us 11, same as a[0]? Yessss. You’re starting to get a glimpe of how arrays and pointers are related in C.

6.6.2 Passing Single Dimensional Arrays to Functions

Let’s do an example with a single dimensional array. I’m going to write a couple functions that we can pass the array to that do different things.

Prepare for some mind-blowing function signatures!

#include <stdio.h>

// Passing as a pointer to the first element
void times2(int *a, int len)
{
    for (int i = 0; i < len; i++)
        printf("%d\n", a[i] * 2);
}

// Same thing, but using array notation
void times3(int a[], int len)
{
    for (int i = 0; i < len; i++)
        printf("%d\n", a[i] * 3);
}

// Same thing, but using array notation with size
void times4(int a[5], int len)
{
    for (int i = 0; i < len; i++)
        printf("%d\n", a[i] * 4);
}

int main(void)
{
    int x[5] = {11, 22, 33, 44, 55};

    times2(x, 5);
    times3(x, 5);
    times4(x, 5);
}

All those methods of listing the array as a parameter in the function are identical.

    void times2(int *a, int len)
    void times3(int a[], int len)
    void times4(int a[5], int len)

In C, the first is the most common, by far.

And, in fact, in the latter situation, the compiler doesn’t even care what number you pass in (other than it has to be greater than zero⁴⁷). It doesn’t enforce anything at all.

Now that I’ve said that, the size of the array in the function declaration actually does matter when you’re passing multidimensional arrays into functions, but let’s come back to that.

6.6.3 Changing Arrays in Functions

We’ve said that arrays are just pointers in disguise. This means that if you pass an array to a function, you’re likely passing a pointer to the first element in the array.

But if the function has a pointer to the data, it is able to manipulate that data! So changes that a function makes to an array will be visible back out in the caller.

Here’s an example where we pass a pointer to an array into a function, the function manipulates the values in that array, and those changes are visible out in the caller.

#include <stdio.h>

void double_array(int *a, int len)
{
    // Multiple each element by 2
    //
    // This doubles the values in x in main() since x and a both point
    // to the same array in memory!

    for (int i = 0; i < len; i++)
        a[i] *= 2;
}

int main(void)
{
    int x[5] = {1, 2, 3, 4, 5};

    double_array(x, 5);

    for (int i = 0; i < 5; i++)
        printf("%d\n", x[i]);  // 2, 4, 6, 8, 10!
}

Later when we talk about the equivalence between arrays and pointers, we’ll see how this makes a lot more sense. For now, it’s enough to know that functions can make changes to arrays that are visible out in the caller.

6.6.4 Passing Multidimensional Arrays to Functions

The story changes a little when we’re talking about multidimensional arrays. C needs to know all the dimensions (except the first one) so it has enough information to know where in memory to look to find a value.

Here’s an example where we’re explicit with all the dimensions:

#include <stdio.h>

void print_2D_array(int a[2][3])
{
    for (int row = 0; row < 2; row++) {
        for (int col = 0; col < 3; col++)
            printf("%d ", a[row][col]);
        printf("\n");
    }
}

int main(void)
{
    int x[2][3] = {
        {1, 2, 3},
        {4, 5, 6}
    };

    print_2D_array(x);
}

But in this case, these two⁴⁸ are equivalent:

    void print_2D_array(int a[2][3])
    void print_2D_array(int a[][3])

The compiler really only needs the second dimension so it can figure out how far in memory to skip for each increment of the first dimension.

Also, the compiler does minimal compile-time bounds checking (if you’re lucky), and C does zero runtime checking of bounds. No seat belts! Don’t crash!

7 Strings

Finally! Strings! What could be simpler?

Well, turns out strings aren’t actually strings in C. That’s right! They’re pointers! Of course they are!

Much like arrays, strings in C barely exist.

But let’s check it out—it’s not really such a big deal.

7.1 Constant Strings

Before we start, let’s talk about constant strings in C. These are sequences of characters in double quotes ("). (Single quotes enclose characters, and are a different animal entirely.)

Examples:

    "Hello, world!\n"
    "This is a test."
    "When asked if this string had quotes in it, she replied, \"It does.\""

The first one has a newline at the end—quite a common thing to see.

The last one has quotes embedded within it, but you see each is preceded by (we say “escaped by”) a backslash (\) indicating that a literal quote belongs in the string at this point. This is how the C compiler can tell the difference between printing a double quote and the double quote at the end of the string.

7.2 String Variables

Now that we know how to make a constant string, let’s assign it to a variable so we can do something with it.

    char *s = "Hello, world!";

Check out that type: pointer to a char⁴⁹. The string variable s is actually a pointer to the first character in that string, namely the H.

And we can print it with the %s (for “string”) format specifier:

    char *s = "Hello, world!";
    
    printf("%s\n", s);  // "Hello, world!"

7.3 String Variables as Arrays

Another option is this, equivalent to the above char* usage:

    char s[14] = "Hello, world!";
    
    // or, if we were properly lazy:
    
    char s[] = "Hello, world!";

This means you can use array notation to access characters in a string. Let’s do exactly that to print all the characters in a string on the same line:

#include <stdio.h>

int main(void)
{
    char s[] = "Hello, world!";

    for (int i = 0; i < 13; i++)
        printf("%c\n", s[i]);
}

Note that we’re using the format specifier %c to print a single character.

Also, check this out. The program will still work fine if we change the definition of s to be a char* type:

#include <stdio.h>

int main(void)
{
    char *s = "Hello, world!";   // char* here

    for (int i = 0; i < 13; i++)
        printf("%c\n", s[i]);    // But still use arrays here...?
}

And we still can use array notation to get the job done when printing it out! This is surprising, but is still only because we haven’t talked about array/pointer equivalence yet. But this is yet another hint that arrays and pointers are the same thing, deep down.

7.4 String Initializers

We’ve already seen some examples with initializing string variables with constant strings:

    char *s = "Hello, world!";
    char t[] = "Hello, again!";

But these two are subtly different.

This one is a pointer to a constant string (i.e. a pointer to the first character in a constant string):

    char *s = "Hello, world!";

If you try to mutate that string with this:

    char *s = "Hello, world!";
    
    s[0] = 'z';  // BAD NEWS: tried to mutate a constant string!

The behavior is undefined. Probably, depending on your system, a crash will result.

But declaring it as an array is different. This one is a non-constant, mutable copy of the constant string that we can change at will

    char t[] = "Hello, again!";  // t is an array copy of the string 
    t[0] = 'z'; //  No problem
    
    printf("%s\n", t);  // "zello, again!"

So remember: if you have a pointer to a constant string, don’t try to change it!

7.5 Getting String Length

You can’t, since C doesn’t track it for you. And when I say “can’t”, I actually mean “can”⁵⁰. There’s a function in <string.h> called strlen() that can be used to compute the length of any string.

#include <stdio.h>
#include <string.h>

int main(void)
{
    char *s = "Hello, world!";

    printf("The string is %zu characters long.\n", strlen(s));
}

The strlen() function returns type size_t, which is an integer type so you can use it for integer math. We print size_t with %zu.

The above program prints:

    The string is 13 characters long.

Great! So it is possible to get the string length!

But… if C doesn’t track the length of the string anywhere, how does it know how long the string is?

7.6 String Termination

C does strings a little differently than many programming languages, and in fact differently than almost every modern programming language.

When you’re making a new language, you have basically two options for storing a string in memory:

Store the bytes of the string along with a number indicating the length of the string.
Store the bytes of the string, and mark the end of the string with a special byte called the terminator.

If you want strings longer than 255 characters, option 1 requires at least two bytes to store the length. Whereas option 2 only requires one byte to terminate the string. So a bit of savings there.

Of course, these days is seems ridiculous to worry about saving a byte (or 3—lots of languages will happily let you have strings that are 4 gigabytes in length). But back in the day, it was a bigger deal.

So C took approach #2. In C, a “string” is defined by two basic characteristics:

A pointer to the first character in the string.
A zero-valued byte (or NUL character⁵¹) somewhere in memory after the pointer that indicates the end of the string.

A NUL character can be written in C code as \0, though you don’t often have to do this.

When you include a constant string in your code, the NUL character is automatically, implicitly included.

    char *s = "Hello!";  // Actually "Hello!\0" behind the scenes

So with this in mind, let’s write our own strlen() function that counts characters in a string until it finds a NUL.

The procedure is to look down the string for a single NUL character, counting as we go⁵²:

    int my_strlen(char *s)
    {
        int count = 0;
    
        while (s[count] != '\0')  // Single quotes for single char
            count++;
    
        return count;
    }

And that’s basically how the built-in strlen() gets the job done.

7.7 Copying a String

You can’t copy a string through the assignment operator (=). All that does is make a copy of the pointer to the first character… so you end up with two pointers to the same string:

#include <stdio.h>

int main(void)
{
    char s[] = "Hello, world!";
    char *t;

    // This makes a copy of the pointer, not a copy of the string!
    t = s;

    // We modify t
    t[0] = 'z';

    // But printing s shows the modification!
    // Because t and s point to the same string!

    printf("%s\n", s);  // "zello, world!"
}

If you want to make a copy of a string, you have to copy it a byte at a time—but this is made easier with the strcpy() function⁵³.

Before you copy the string, make sure you have room to copy it into, i.e. the destination array that’s going to hold the characters needs to be at least as long as the string you’re copying.

#include <stdio.h>
#include <string.h>

int main(void)
{
    char s[] = "Hello, world!";
    char t[100];  // Each char is one byte, so plenty of room

    // This makes a copy of the string!
    strcpy(t, s);

    // We modify t
    t[0] = 'z';

    // And s remains unaffected because it's a different string
    printf("%s\n", s);  // "Hello, world!"

    // But t has been changed
    printf("%s\n", t);  // "zello, world!"
}

Notice with strcpy(), the destination pointer is the first argument, and the source pointer is the second. A mnemonic I use to remember this is that it’s the order you would have put t and s if an assignment = worked for strings.

8 Structs

In C, have something called a struct, which is a user-definable type that holds multiple pieces of data, potentially of different types.

It’s a convenient way to bundle multiple variables into a single one. This can be beneficial for passing variables to functions (so you just have to pass one instead of many), and useful for organizing data and making code more readable.

If you’ve come from another language, you might be familiar with the idea of classes and objects. These don’t exist in C, natively⁵⁴. You can think of a struct as a class with only data members, and no methods.

8.1 Declaring a Struct

You can declare a struct in your code like so:

    struct car {
        char *name;
        float price;
        int speed;
    };

This is often done at the global scope outside any functions so that the struct is globally available.

When you do this, you’re making a new type. The full type name is struct car. (Not just car—that won’t work.)

There aren’t any variables of that type yet, but we can declare some:

    struct car saturn;

And now we have an uninitialized variable saturn⁵⁵ of type struct car.

We should initialize it! But how do we set the values of those individual fields?

Like in many other languages that stole it from C, we’re going to use the dot operator (.) to access the individual fields.

    saturn.name = "Saturn SL/2";
    saturn.price = 15999.99;
    saturn.speed = 175;
    
    printf("Name:           %s\n", saturn.name);
    printf("Price (USD):    %f\n", saturn.price);
    printf("Top Speed (km): %d\n", saturn.speed);

8.2 Struct Initializers

That example in the previous section was a little unwieldy. There must be a better way to initialize that struct variable!

You can do it with an initializer by putting values in for the fields in the order they appear in the struct when you define the variable. (This won’t work after the variable has been defined—it has to happen in the definition).

    struct car {
        char *name;
        float price;
        int speed;
    };
    
    // Now with an initializer! Same field order as in the struct declaration:
    struct car saturn = {"Saturn SL/2", 16000.99, 175};
    
    printf("Name:      %s\n", saturn.name);
    printf("Price:     %f\n", saturn.price);
    printf("Top Speed: %d km\n", saturn.speed);

The fact that the fields in the initializer need to be in the same order is a little freaky. If someone changes the order in struct car, it could break all the other code!

We can be more specific with our initializers:

    struct car saturn = {.speed=172, .name="Saturn SL/2"};

Now it’s independent of the order in the struct declaration. Which is safer code, for sure.

Similar to array initializers, any missing field designators are initialized to zero (in this case, that would be .price, which I’ve omitted).

8.3 Passing Structs to Functions

You can do a couple things to pass a struct to a function.

Pass the struct.
Pass a pointer to the struct.

Recall that when you pass something to a function, a copy of that thing gets made for the function to operate on, whether it’s a copy of a pointer, an int, a struct, or anything.

There are basically two cases when you’d want to pass a pointer to the struct:

You need the function to be able to make changes to the struct that was passed in, and have those changes show in the caller.
The struct is somewhat large and it’s more expensive to copy that onto the stack than it is to just copy a pointer⁵⁶

For those two reasons, it’s far more common to pass a pointer to a struct to a function.

Let’s try that, making a function that will allow you to set the .price field of the struct car:

struct car {
    char *name;
    float price;
    int speed;
};

int main(void)
{
    struct car saturn = {.speed=175, .name="Saturn SL/2"};

    // Pass a pointer to this struct car, along with a new,
    // more realistic, price:
    set_price(&saturn, 800.00);

    // ... code continues ...

You should be able to come up with the function signature for set_price() just by looking at the types of the arguments we have there.

saturn is a struct car, so &saturn must be the address of the struct car, AKA a pointer to a struct car, namely a struct car*.

And 800.0 is a float.

So the function declaration must look like this:

    void set_price(struct car *c, float new_price)

We just need to write the body. One attempt might be:

    void set_price(struct car *c, float new_price) {
        c.price = new_price;  // ERROR!!
    }

That won’t work because the dot operator only works on structs… it doesn’t work on pointers to structs.

Ok, so we can dereference the struct to de-pointer it to get to the struct itself. Dereferencing a struct car* results in the struct car that the pointer points to, which we should be able to use the dot operator on:

    void set_price(struct car *c, float new_price) {
        (*c).price = new_price;  // Works, but non-idiomatic :(
    }

And that works! But it’s a little clunky to type all those parens and the asterisk. C has some syntactic sugar called the arrow operator that helps with that.

8.4 The Arrow Operator

    void set_price(struct car *c, float new_price) {
        // (*c).price = new_price;  // Works, but non-idiomatic :(
        //
        // The line above is 100% equivalent to the one below:
    
        c->price = new_price;  // That's the one!
    }

The arrow operator helps refer to fields in pointers to structs.

So when accessing fields. when do we use dot and when do we use arrow?

If you have a struct, use dot (.).
If you have a pointer to a struct, use arrow (->).

8.5 Copying and Returning `struct`s

Here’s an easy one for you!

Just assign from one to the other!

    struct a, b;
    
    b = a;  // Copy the struct

And returning a struct (as opposed to a pointer to one) from a function also makes a similar copy to the receiving variable.

This is not a “deep copy”. All fields are copied as-is, including pointers to things.

9 File Input/Output

We’ve already seen a couple examples of I/O with scanf() and printf() for doing I/O at the console (screen/keyboard).

But we’ll push those concepts a little farther this chapter.

9.1 The `FILE*` Data Type

When we do any kind of I/O in C, we do so though a piece of data that you get in the form of a FILE* type. This FILE* holds all the information needed to communicate with the I/O subsystem about which file you have open, where you are in the file, and so on.

The spec refers to these as streams, i.e. a stream of data from a file or from any source. I’m going to use “files” and “streams” interchangeably, but really you should think of a “file” as a special case of a “stream”. There are other ways to stream data into a program than just reading from a file.

We’ll see in a moment how to go from having a filename to getting an open FILE* for it, but first I want to mention three streams that are already open for you and ready for use.

`FILE*` name	Description
`stdin`	Standard Input, generally the keyboard by default
`stdout`	Standard Output, generally the screen by default
`stderr`	Standard Error, generally the screen by default, as well

We’ve actually been using these implicitly already, it turns out. For example, these two calls are the same:

    printf("Hello, world!\n");
    fprintf(stdout, "Hello, world!\n");  // printf to a file

But more on that later.

Also you’ll notice that both stdout and stderr go to the screen. While this seems at first either like an oversight or redundancy, it actually isn’t. Typical operating systems allow you to redirect the output of either of those into different files, and it can be convenient to be able to separate error messages from regular non-error output.

For example, in a POSIX shell (like sh, ksh, bash, zsh, etc.) on a Unix-like system, we could run a program and send just the non-error (stdout) output to one file, and all the error (stderr) output to another file.

    $ ./foo > output.txt 2> errors.txt   # This command is Unix-specific

For this reason, you should send serious error messages to stderr instead of stdout.

9.2 Reading Text Files

Streams are largely categorized two different ways: text and binary.

Text streams are allowed to do significant translation of the data, most notably translations of newlines to their different representations⁵⁷. Text files are logically a sequence of lines separated by newlines. To be portable, your input data should always end with a newline.

But the general rule is that if you’re able to edit the file in a regular text editor, it’s a text file. Otherwise, it’s binary. More on binary later.

So let’s get to work—how do we open a file for reading, and pull data out of it?

Let’s create a file called hello.txt that has just this in it:

    Hello, world!

And let’s write a program to open the file, read a character out of it, and then close the file when we’re done. That’s the game plan!

#include <stdio.h>

int main(void)
{
    FILE *fp;                      // Variable to represent open file

    fp = fopen("hello.txt", "r");  // Open file for reading

    char c = fgetc(fp);            // Read a single character
    printf("%c\n", c);             // Print char to stdout

    fclose(fp);                    // Close the file when done
}

See how when we opened the file with fopen(), it returned the FILE* to us so we could use it later.

(I’m leaving it out for brevity, but fopen() will return NULL if something goes wrong, so you should really error check it!)

Also notice the "r" that we passed in—this means “open a text stream for reading”. (There are various strings we can pass to fopen() with additional meaning, like writing, or appending, and so on.)

After that, we used the fgetc() function to get a character from the stream.

Finally, we close the stream when we’re done with it. All streams are automatically closed when the program exits, but it’s good form and good housekeeping to explicitly close any files yourself when done with them.

The FILE* keeps track of our position in the file. So subsequent calls to fgetc() would get the next character in the file, and then the next, until the end.

But that sounds like a pain. Let’s see if we can make it easier.

9.3 End of File: `EOF`

There is a special character defined as a macro: EOF. This is what fgetc() will return when the end of the file has been reached and you’ve attempted to read another character.

We can use this to read the whole file in a loop.

#include <stdio.h>

int main(void)
{
    FILE *fp;

    fp = fopen("hello.txt", "r");
    char c;

    while ((c = fgetc(fp)) != EOF)
        printf("%c", c);

    fclose(fp);
}

(If line 10 is too weird, just break it down starting with the innermost-nested parens. The first thing we do is assign the result of fgets() into c, and then we compare that against EOF. We’ve just crammed it into a single line. This might look hard to read, but study it—it’s idiomatic C.)

And running this, we see:

    Hello, world!

But still, we’re operating a character at a time, and lots of text files make more sense at the line level. Let’s switch to that.

9.3.1 Reading a Line at a Time

So how can we get an entire line at once? fgets() to the rescue! For arguments, it takes a pointer to a char buffer to hold bytes, a maximum number of bytes to read, and a FILE* to read from. It returns NULL on end-of-file or error. fgets() is even nice enough to NUL-terminate the string when its done⁵⁸.

Let’s do a similar loop as before, except let’s have a multiline file and read it in a line at a time.

Here’s a file quote.txt:

    A wise man can learn more from
    a foolish question than a fool
    can learn from a wise answer.
                      --Bruce Lee

And here’s some code that reads that file a line at a time and prints out a line number before each one:

#include <stdio.h>

int main(void)
{
    FILE *fp;
    char s[1024];  // Big enough for any line this program will encounter
    int linecount = 0;

    fp = fopen("quote.txt", "r");

    while (fgets(s, sizeof s, fp) != NULL) 
        printf("%d: %s", ++linecount, s);

    fclose(fp);
}

Which gives the output:

    1: A wise man can learn more from
    2: a foolish question than a fool
    3: can learn from a wise answer.
    4:                   --Bruce Lee

9.4 Formatted Input

You know how you can get formatted output with printf() (and, thus, fprintf() like we’ll see, below)?

You can do the same thing with fscanf().

Let’s have a file with a series of data records in it. In this case, whales, with name, length in meters, and weight in tonnes. whales.txt:

    blue 29.9 173
    right 20.7 135
    gray 14.9 41
    humpback 16.0 30

Yes, we could read these with fgets() and then parse the string with sscanf() (and in some ways that’s more resilient against corrupted files), but in this case, let’s just use fscanf() and pull it in directly.

The sscanf() function skips whitespace when reading, and returns EOF on end-of-file or error.

#include <stdio.h>

int main(void)
{
    FILE *fp;
    char name[1024];  // Big enough for any line this program will encounter
    float length;
    int mass;

    fp = fopen("whales.txt", "r");

    while (fscanf(fp, "%s %f %d", name, &length, &mass) != EOF)
        printf("%s whale, %d tonnes, %.1f meters\n", name, mass, length);

    fclose(fp);
}

Which gives the result:

    blue whale, 173 tonnes, 29.9 meters
    right whale, 135 tonnes, 20.7 meters
    gray whale, 41 tonnes, 14.9 meters
    humpback whale, 30 tonnes, 16.0 meters

9.5 Writing Text Files

In much the same way we can use fgetc(), fgets(), and fscanf() to read text streams, we can use fputc(), fputs(), and fprintf() to write text streams.

To do so, we have to fopen() the file in write mode by passing "w" as the second argument. Opening an existing file in "w" mode will instantly truncate that file to 0 bytes for a full overwrite.

We’ll put together a simple program that outputs a file output.txt using a variety of output functions.

#include <stdio.h>

int main(void)
{
    FILE *fp;
    int x = 32;

    fp = fopen("output.txt", "w");

    fputc('B', fp);
    fputc('\n', fp);
    fprintf(fp, "x = %d\n", x);
    fputs("Hello, world!\n", fp);

    fclose(fp);
}

And this produces a file, output.txt, with these contents:

    B
    x = 32
    Hello, world!

Fun fact: since stdout is a file, you could replace line 8 with:

    fp = stdout;

and the program would have outputted to the console instead of to a file. Try it!

9.6 Binary File I/O

So far we’ve just been talking text files. But there’s that other beast we mentioned early on called binary files, or binary streams.

These work very similarly to text files, except the I/O subsystem doesn’t perform any translations on the data like it might with a text file. With binary files, you get a raw stream of bytes, and that’s all.

The big difference in opening the file is that you have to add a "b" to the mode. That is, to read a binary file, open it in "rb" mode. To write a file, open it in "wb" mode.

Because it’s streams of bytes, and streams of bytes can contain NUL characters, and the NUL character is the end-of-string marker in C, it’s rare that people use the fprintf()-and-friends functions to operate on binary files.

Instead the most common functions are fread() and fwrite(). The functions read and write a specified number of bytes to the stream.

To demo, we’ll write a couple programs. One will write a sequence of byte values to disk all at once. And the second program will read a byte at a time and print them out⁵⁹.

#include <stdio.h>

int main(void)
{
    FILE *fp;
    unsigned char bytes[] = {5, 37, 0, 88, 255, 12};

    fp = fopen("output.bin", "wb");  // wb mode for "write binary"!

    // In the call to fwrite, the arguments are:
    //
    // * Pointer to data to write
    // * Size of each "piece" of data
    // * Count of each "piece" of data
    // * FILE*

    fwrite(bytes, sizeof(char), sizeof bytes, fp);

    fclose(fp);
}

Those two middle arguments to fwrite() are pretty odd. But basically what we want to tell the function is, “We have items that are this big, and we want to write that many of them.” This makes it convenient if you have a record of a fixed length, and you have a bunch of them in an array. You can just tell it the size of one record and how many to write.

In the example above, we tell it each record is the size of a char, and we have 6 of them as computed by sizeof bytes.

Running the program gives us a file output.bin, but opening it in a text editor doesn’t show anything friendly! It’s binary data—not text. And random binary data I just made up, at that!

If I run it through a hex dump ⁶⁰ program, we can see the output as bytes:

    05 25 00 58 ff 0c

And those values in hex do match up to the values (in decimal) that we wrote out.

But now let’s try to read them back in with a different program. This one will open the file for binary reading ("rb" mode) and will read the bytes one at a time in a loop.

fread() has the neat feature where it returns the number of bytes read, or 0 on EOF. So we can loop until we see that, printing numbers as we go.

#include <stdio.h>

int main(void)
{
    FILE *fp;
    unsigned char c;

    fp = fopen("output.bin", "rb"); // rb for "read binary"!

    while (fread(&c, sizeof(char), 1, fp) > 0)
        printf("%d\n", c);
}

And, running it, we see our original numbers!

Woo hoo!

9.6.1 `struct` and Number Caveats

As we saw in the structs section, the compiler is free to add padding to a struct as it sees fit. And different compilers might do this differently. And the same compiler on different architectures could do it differently. And the same compiler on the same architectures could do it differently.

What I’m getting at is, it’s not portable to just fwrite() an entire struct out to a file when you don’t know where the padding will end up.

How do we fix this? Hold that thought—we’ll look at some ways to do this after looking at another related problem.

Numbers!

Turns out all architectures don’t represent numbers in memory the same way.

Let’s look at a simple fwrite() of a 2-byte number. We’ll write it in hex so each byte is clear. The most significant byte will have the value 0x12 and the least significant will have the value 0x34.

    unsigned short v = 0x1234;  // Two bytes, 0x12 and 0x34
    
    fwrite(&v, sizeof v, 1, fp);

What ends up in the stream?

Well, it seems like it should be 0x12 followed by 0x34, right?

But if I run this on my machine and hex dump the result, I get:

    34 12

They’re reversed! What gives?

This has something to do with what’s called the endianess⁶¹ of the architecture. Some write the most significant bytes first, and some the least significant bytes first.

This means that if you write a multibyte number out straight from memory, you can’t do it in a portable way⁶².

A similar problem exists with floating point. Most systems use the same format for their floating point numbers, but some do not. No guarantees!

So… how can we fix all these problems with numbers and structs to get our data written in a portable way?

The summary is to serialize the data, which is a general term that means to take all the data and write it out in a format that you control, that is well-known, and programmable to work the same way on all platforms.

As you might imagine, this is a solved problem. There are a bunch of serialization libraries you can take advantage of, such as Google’s protocol buffers⁶³, out there and ready to use. They will take care of all the gritty details for you, and even will allow data from your C programs to interoperate with other languages that support the same serialization methods.

Do yourself and everyone a favor! Serialize your binary data when you write it to a stream!

10 `typedef`: Making New Types

Well, not so much making new types as getting new names for existing types. Sounds kinda pointless on the surface, but we can really use this to make our code cleaner.

10.1 `typedef` in Theory

Basically, you take an existing type and you make an alias for it with typedef.

Like this:

    typedef int antelope;  // Make "antelope" an alias for "int"
    
    antelope x = 10;       // Type "antelope" is the same as type "int"

You can take any existing type and do it. You can even make a number of types with a comma list:

    typedef int antelope, bagel, mushroom;  // These are all "int"

That’s really useful, right? That you can type mushroom instead of int? You must be super excited about this feature!

OK, Professor Sarcasm—we’ll get to some more common applications of this in a moment.

10.1.1 Scoping

typedef follows regular scoping rules.

For this reason, it’s quite common to find typedef at file scope (“global”) so that all functions can use the new types at will.

10.2 `typedef` in Practice

So renaming int to something else isn’t that exciting. Let’s see where typedef commonly makes an appearance.

10.2.1 `typedef` and `struct`s

Sometimes a struct will be typedef’d to a new name so you don’t have to type the word struct over and over.

    struct animal {
        char *name;
        int leg_count, speed;
    };
    
    //  original name      new name
    //            |         |
    //            v         v
    //      |-----------| |----|
    typedef struct animal animal;
    
    struct animal y;  // This works
    animal z;         // This also works because "animal" is an alias

Personally, I don’t care for this practice. I like the clarity the code has when you add the word struct to the type; programmers know what they’re getting. But it’s really common so I’m including it here.

Now I want to run the exact same example in a way that you might commonly see. We’re going to put the struct animal in the typedef. You can mash it all together like this:

    //  original name
    //            |
    //            v
    //      |-----------|
    typedef struct animal {
        char *name;
        int leg_count, speed;
    } animal;                         // <-- new name
    
    struct animal y;  // This works
    animal z;         // This also works because "animal" is an alias

That’s exactly the same as the previous example, just more concise.

But that’s not all! There’s another common shortcut that you might see in code using what are called anonymous structures⁶⁴. It turns out you don’t actually need to name the structure in a variety of places, and with typedef is one of them.

Let’s do the same example with an anonymous structure:

    //  anonymous struct!
    //         |
    //         v
    //      |----|
    typedef struct {
        char *name;
        int leg_count, speed;
    } animal;                         // <-- new name
    
    //struct animal y;  // ERROR: this no longer works
    animal z;           // This works because "animal" is an alias

As another example, we might find something like this:

    typedef struct {
        int x, y;
    } point;
    
    point p = {.x=20, .y=40};
    
    printf("%d, %d\n", p.x, p.y);  // 20, 40

10.2.2 `typedef` and Other Types

It’s not that using typedef with a simple type like int is completely useless… it helps you abstract the types to make it easier to change them later.

For example, if you have float all over your code in 100 zillion places, it’s going to be painful to change them all to double if you find you have to do that later for some reason.

But if you prepared a little with:

    typedef float app_float;
    
    // and
    
    app_float f1, f2, f3;

Then if later you want to change to another type, like long double, you just nee to change the typedef:

    //        voila!
    //      |---------|
    typedef long double app_float;
    
    // and
    
    app_float f1, f2, f3;  // Now these are all long doubles

10.2.3 `typedef` and Pointers

You can make a type that is a pointer.

    typedef int *intptr;
    
    int a = 10;
    intptr x = &a;  // "intptr" is type "int*"

I really don’t like this practice. It hides the fact that x is a pointer type because you don’t see a * in the declaration.

IMHO, it’s better to explicitly show that you’re declaring a pointer type so that other devs can clearly see it and don’t mistake x for having a non-pointer type.

10.2.4 `typedef` and Capitalization

I’ve seen all kinds of capitalization on typedef.

    typedef struct {
        int x, y;
    } my_point;          // lower snake case
    
    typedef struct {
        int x, y;
    } MyPoint;          // CamelCase
    
    typedef struct {
        int x, y;
    } Mypoint;          // Leading uppercase
    
    typedef struct {
        int x, y;
    } MY_POINT;          // UPPER SNAKE CASE

The C11 specification doesn’t dictate one way or another, and shows examples in all uppercase and all lowercase.

K&R2 uses leading uppercase predominantly, but show some examples in uppercase and snake case (with _t).

If you have a style guide in use, stick with it. If you don’t, grab one and stick with it.

10.3 Arrays and `typedef`

The syntax is a little weird, and this is rarely seen in my experience, but you can typedef an array of some number of items.

    // Make type five_ints an array of 5 ints
    typedef int five_ints[5];
    
    five_ints x = {11, 22, 33, 44, 55};

I don’t like it because it hides the array nature of the variable, but it’s possible to do.

11 Pointers II: Arithmetic

Time to get more into it with a number of new pointer topics! If you’re not up to speed with pointers, check out the first section in the guide on the matter.

11.1 Pointer Arithmetic

Turns out you can do math on pointers, notably addition and subtraction.

But what does it mean when you do that?

In short, if you have a pointer to a type, adding one to the pointer moves to the next item of that type directly after it in memory.

It’s important to remember that as we move pointers around and look at different places in memory, we need to make sure that we’re always pointing to a valid place in memory before we dereference. If we’re off in the weeds and we try to see what’s there, the behavior is undefined and a crash is a common result.

This is a little chicken-and-eggy with Array/Pointer Equivalence, below, but we’re going to give it a shot, anyway.

11.1.1 Adding to Pointers

First, let’s take an array of numbers.

    int a[5] = {11, 22, 33, 44, 55};

Then let’s get a pointer to the first element in that array:

    int a[5] = {11, 22, 33, 44, 55};
    
    int *p = &a[0];  // Or "int *p = a;" works just as well

The let’s print the value there by dereferencing the pointer:

    printf("%d\n", *p);  // Prints 11

Now let’s use pointer arithmetic to print the next element in the array, the one at index 1:

    printf("%d\n", *(p + 1));  // Prints 22!!

What happened there? C knows that p is a pointer to an int. So it knows the sizeof an int⁶⁵ and it knows to skip that many bytes to get to the next int after the first one!

In fact, the prior example could be written these two equivalent ways:

    printf("%d\n", *p);        // Prints 11
    printf("%d\n", *(p + 0));  // Prints 11

because adding 0 to a pointer results in the same pointer.

Let’s think of the upshot here. We can iterate over elements of an array this way instead of using an array:

    int a[5] = {11, 22, 33, 44, 55};
    
    int *p = &a[0];  // Or "int *p = a;" works just as well
    
    for (int i = 0; i < 5; i++) {
        printf("%d\n", *(p + i));  // Same as p[i]!
    }

And that works the same as if we used array notation! Oooo! Getting closer to that array/pointer equivalence thing! More on this later in this chapter.

But what’s actually happening, here? How do it work?

Remember from early on that memory is like a big array, where a byte is stored at each array index.

And the array index into memory has a few names:

Index into memory
Location
Address
Pointer!

So a point is an index into memory, somewhere.

For a random example, say that a number 3490 was stored at address (“index”) 23,237,489,202. If we have an int pointer to that 3490, that value of that pointer is 23,237,489,202… because the pointer is the memory address. Different words for the same thing.

And now let’s say we have another number, 4096, stored right after the 3490 at address 23,237,489,210 (8 higher than the 3490 because each int in this example is 8 bytes long).

If we add 1 to that pointer, it actually jumps ahead sizeof(int) bytes to the next int. It knows to jump that far ahead because it’s an int pointer. If it were a float pointer, it’d jump sizeof(float) bytes ahead to get to the next float!

So you can look at the next int, by adding 1 to the pointer, the one after that by adding 2 to the pointer, and so on.

11.1.2 Changing Pointers

We saw how we could add an integer to a pointer in the previous section. This time, let’s modify the pointer, itself.

You can just add (or subtract) integer values directly to (or from) any pointer!

Let’s do that example again, except with a couple changes. First, I’m going to add a 999 to the end of our numbers to act as a sentinel value. This will let us know where the end of the data is.

    int a[] = {11, 22, 33, 44, 55, 999};  // Add 999 here as a sentinel
    
    int *p = &a[0];  // p points to the 11

And we also have p pointing to the element at index 0 of a, namely 11, just like before.

Now—let’s starting incrementing p so that it points at subsequent elements of the array. We’ll do this until p points to the 999; that is, we’ll do it until *p == 999:

    while (*p != 999) {       // While the thing p points to isn't 999
        printf("%d\n", *p);   // Print it
        p++;                  // Move p to point to the next int!
    }

Pretty crazy, right?

When we give it a run, first p points to 11. Then we increment p, and it points to 22, and then again, it points to 33. And so on, until it points to 999 and we quit.

11.1.3 Subtracting Pointers

You can subtract a value from a pointer to get to earlier address, as well, just like we were adding to them before.

But we can also subtract two pointers to find the difference between them, e.g. we can calculate how many ints there are between two int*s. The catch is that this only works within a single array⁶⁶—if the pointers point to anything else, you get undefined behavior.

Remember how strings are char*s in C? Let’s see if we can use this to write another variant of strlen() to compute the length of a string that utilizes pointer subtraction.

The idea is that if we have a pointer to the beginning of the string, we can find a pointer to the end of the string by scanning ahead for the NUL character.

And if we have a pointer to the beginning of the string, and we computed the pointer to the end of the string, we can just subtract the two pointers to come up with the length!

#include <stdio.h>

int my_strlen(char *s)
{
    // Start scanning from the beginning of the string
    char *p = s;

    // Scan until we find the NUL character
    while (*p != '\0')
        p++;

    // Return the difference in pointers
    return p - s;
}

int main(void)
{
    printf("%d\n", my_strlen("Hello, world!"));  // Prints "13"
}

Remember that you can only use pointer subtraction between two pointers that point to the same array!

11.2 Array/Pointer Equivalence

We’re finally ready to talk about this! We’ve seen plenty of examples of places where we’ve intermixed array notation, but let’s give out the fundamental formula of array/pointer equivalence:

    a[b] == *(a + b)

Study that! Those are equivalent and can be used interchangeably!

I’ve oversimplified a bit, because in my above example a and b can both be expressions, and we might want a few more parentheses to force order of operations in case the expressions are complex.

The spec is specific, as always, declaring (in C11 §6.5.2.1¶2):

E1[E2] is identical to (*((E1)+(E2)))

but that’s a little harder to grok. Just make sure you include parentheses if the expressions are complicated so all your math happens in the right order.

This means we can decide if we’re going to use array or pointer notation for any array or pointer (assuming it points to an element of an array).

Let’s use an array and pointer with both array and pointer notation:

#include <stdio.h>

int main(void)
{
    int a[] = {11, 22, 33, 44, 55};  // Add 999 here as a sentinel

    int *p = a;  // p points to the first element of a, 11

    // Print all elements of the array a variety of ways:

    for (int i = 0; i < 5; i++)
        printf("%d\n", a[i]);      // Array notation with a

    for (int i = 0; i < 5; i++)
        printf("%d\n", p[i]);      // Array notation with p

    for (int i = 0; i < 5; i++)
        printf("%d\n", *(a + i));  // Pointer notation with a

    for (int i = 0; i < 5; i++)
        printf("%d\n", *(p + i));  // Pointer notation with p

    for (int i = 0; i < 5; i++)
        printf("%d\n", *(p++));    // Moving pointer p
        //printf("%d\n", *(a++));    // Moving array variable a--ERROR!
}

So you can see that in general, if you have an array variable, you can use pointer or array notion to access elements. Same with a pointer variable.

The one big difference is that you can modify a pointer to point to a different address, but you can’t do that with an array variable.

11.2.1 Array/Pointer Equivalence in Function Calls

This is where you’ll encounter this concept the most, for sure.

If you have a function that takes a pointer argument, e.g.:

    int my_strlen(char *s)

this means you can pass either an array or a pointer to this function and have it work!

    char s[] = "Antelopes";
    char *t = "Wombats";
    
    printf("%d\n", my_strlen(s));  // Works!
    printf("%d\n", my_strlen(t));  // Works, too!

And it’s also why these two function signatures are equivalent:

    int my_strlen(char *s)    // Works!
    int my_strlen(char s[])   // Works, too!

11.3 `void` Pointers

You’ve already seen the void keyword used with functions, but this is an entirely separate, unrelated animal.

Sometimes it’s useful to have a pointer to a thing that you don’t know the type of.

I know. Bear with me just a second.

There are basically two use cases for this.

A function is going to operate on something byte-by-byte. For example, memcpy() copies bytes of memory from one pointer to another, but those pointers can point to any type. memcpy() takes advantage of the fact that if you iterate through char*s, you’re iterating through the bytes of an object no matter what type the object is. More on this in the Multibyte Values subsection.
Another function is calling a function you passed to it (a callback), and it’s passing you data. You know the type of the data, but the function calling you doesn’t. So it passes you void*s—’cause it doesn’t know the type—and you convert those to the type you need. The built-in qsort() and bsearch() use this technique.

Let’s look at an example, the built-in memcpy() function:

    void *memcpy(void *s1, void *s2, size_t n);

This function copies n bytes of memory starting from address s1 into the memory starting at address s2.

But look! s1 and s2 are void*s! Why? What does it mean? Let’s run more examples to see.

For instance, we could copy a string with memcpy() (though strcpy() is more appropriate for strings):

#include <stdio.h>
#include <string.h>

int main(void)
{
    char s[] = "Goats!";
    char t[100];

    memcpy(t, s, 7);  // Copy 7 bytes--including the NUL terminator!

    printf("%s\n", t);  // "Goats!"
}

Or we can copy some ints:

#include <stdio.h>
#include <string.h>

int main(void)
{
    int a[] = {11, 22, 33};
    int b[3];

    memcpy(b, a, 3 * sizeof(int));  // Copy 3 ints of data

    printf("%d\n", b[1]);  // 22
}

That one’s a little wild—you see what we did there with memcpy()? We copied the data from a to b, but we had to specify how many bytes to copy, and an int is more than one byte.

OK, then—how many bytes does an int take? Answer: depends on the system. But we can tell how many bytes any type takes with the sizeof operator.

So there’s the answer: an int takes sizeof(int) bytes of memory to store.

And if we have 3 of them in our array, like we did in that example, the entire space used for the 3 ints must be 3 * sizeof(int).

(In the string example, earlier, it would have been more technically accurate to copy 7 * sizeof(char) bytes. But chars are always one byte large, by definition, so that just devolves into 7 * 1.)

We could even copy a float or a struct with memcpy()! (Though this is abusive—we should just use = for that):

    struct antelope my_antelope;
    struct antelopy my_clone_antelope;
    
    // ...
    
    memcpy(&my_clone, &my_antelope, sizeof my_antelope);

Look at how versatile memcpy() is! If you have a pointer to a source and a pointer to a destination, and you have the number of bytes you want to copy, you can copy any type of data.

Imagine if we didn’t have void*. We’d have to write specialized memcpy() functions for each type:

    memcpy_int(int *a, int *b, int count);
    memcpy_float(float *a, float *b, int count);
    memcpy_double(double *a, double *b, int count);
    memcpy_char(char *a, char *b, int count);
    memcpy_unsigned_char(unsigned char *a, unsigned char *b, int count);
    
    // etc... blech!

Much better to just use void* and have one function that can do it all.

That’s the power of void*. You can write functions that don’t care about the type and is still able to do things with it.

But with great power comes great responsibility. Maybe not that great in this case, but there are some limits.

You cannot do pointer arithmetic on a void*.
You cannot dereference a void*.
You cannot use the arrow operator on a void*, since it’s also a dereference.
You cannot use array notation on a void*, since it’s also a dereference, as well⁶⁷.

And if you think about it, these rules make sense. All those operations rely on knowing the sizeof the type of data pointed to, and with void*, we don’t know the size of the data being pointed to—it could be anything!

But wait—if you can’t dereference a void* what good can it ever do you?

Like with memcpy(), it helps you write generic functions that can handle multiple types of data. But the secret is that, deep down, you convert the void* to another type before you use it!

And conversion is easy: you can just assign into a variable of the desired type⁶⁸.

    char a = 'X';  // A single char
    
    void *p = &a;  // p points to the 'X'
    char *q = p;   // q also points to the 'X'
    
    printf("%c\n", *p);  // ERROR--cannot dereference void*!
    printf("%c\n", *q);  // Prints "X"

Let’s write our own memcpy() to try this out. We can copy bytes (chars), and we know the number of bytes because it’s passed in.

    void *my_memcpy(void *dest, void *src, int byte_count)
    {
        // Convert void*s to char*s
        char *s = src, *d = dest;
    
        // Now that we have char*s, we can dereference and copy them
        while (byte_count--) {
            *d++ = *s++;
        }
    
        // Most of these functions return the destination, just in case
        // that's useful to the caller.
        return dest;
    }

Right there at the beginning, we copy the void*s into char*s so that we can use them as char*s. It’s as easy as that.

Then some fun in a while loop, where we decrement byte_count until it becomes false (0). Remember that with post-decrement, the value of the expression is computed (for while to use) and then the variable is decremented.

And some fun in the copy, where we assign *d = *s to copy the byte, but we do it with post-increment so that both d and s move to the next byte after the assignment is made.

Lastly, most memory and string functions return a copy of a pointer to the destination string just in case the caller wants to use it.

Now that we’ve done that, I just want to quickly point out that we can use this technique to iterate over the bytes of any object in C, floats, structs, or anything!

Let’s run one more real-world example with the built-in qsort() routine that can sort anything thanks to the magic of void*s.

(In the following example, you can ignore the word const, which we haven’t covered yet.)

#include <stdio.h>
#include <stdlib.h>

// The type of structure we're going to sort
struct animal {
    char *name;
    int leg_count;
};

// This is a comparison function called by qsort() to help it determine
// what exactly to sort by. We'll use it to sort an array of struct
// animals by leg_count.
int compar(const void *elem1, const void *elem2)
{
    // We know we're sorting struct animals, so let's make both
    // arguments pointers to struct animals
    const struct animal *animal1 = elem1;
    const struct animal *animal2 = elem2;

    // Return <0 =0 or >0 depending on whatever we want to sort by.

    // Let's sort ascending by leg_count, so we'll return the difference
    // in the leg_counts
    return animal1->leg_count - animal2->leg_count;
}

int main(void)
{
    // Let's build an array of 4 struct animals with different
    // characteristics. This array is out of order by leg_count, but
    // we'll sort it in a second.
    struct animal a[4] = {
        {.name="Dog", .leg_count=4},
        {.name="Monkey", .leg_count=2},
        {.name="Antelope", .leg_count=4},
        {.name="Snake", .leg_count=0}
    };

    // Call qsort() to sort the array. qsort() needs to be told exactly
    // what to sort this data by, and we'll do that inside the compar()
    // function.
    //
    // This call is saying: qsort array a, which has 4 elements, and
    // each element is sizeof(struct animal) bytes big, and this is the
    // function that will compare any two elements.
    qsort(a, 4, sizeof(struct animal), compar);

    // Print them all out
    for (int i = 0; i < 4; i++) {
        printf("%d: %s\n", a[i].leg_count, a[i].name);
    }
}

As long as you give qsort() a function that can compare two items that you have in your array to be sorted, it can sort anything. And it does this without needing to have the types of the items hardcoded in there anywhere. qsort() just rearranges blocks of bytes based on the results of the compar() function you passed in.

12 Manual Memory Allocation

This is one of the big areas where C likely diverges from languages you already know: manual memory management.

Other languages uses reference counting, garbage collection, or other means to determine when to allocate new memory for some data—and when to deallocate it when no variables refer to it.

And that’s nice. It’s nice to be able to not worry about it, to just drop all the references to an item and trust that at some point the memory associated with it will be freed.

But C’s not like that, entirely.

Of course, in C, some variables are automatically allocated and deallocated when they come into scope and leave scope. We call these automatic variables. They’re your average run-of-the-mill block scope “local” variables. No problem.

But what if you want something to persist longer than a particular block? This is where manual memory management comes into play.

You can tell C explicitly to allocate for you a certain number of bytes that you can use as you please. And these bytes will remain allocated until you explicitly free that memory⁶⁹.

It’s important to free the memory you’re done with! If you don’t, we call that a memory leak and your process will continue to reserve that memory until it exits.

If you manually allocated it, you have to manually free it when you’re done with it.

So how do we do this? We’re going to learn a couple new functions, and make use of the sizeof operator to help us learn how many bytes to allocate.

In common C parlance, devs say that automatic local variables are allocated “on the stack”, and manually-allocated memory is “on the heap”. The spec doesn’t talk about either of those things, but all C devs will know what you’re talking about if you bring them up.

All functions we’re going to learn in this chapter can be found in <stdlib.h>.

12.1 Allocating and Deallocating, `malloc()` and `free()`

The malloc() function accepts a number of bytes to allocate, and returns a void pointer to that block of newly-allocated memory.

Since it’s a void*, you can assign it into whatever pointer type you want… normally this will correspond in some way to the number of bytes you’re allocating.

So… how many bytes should I allocate? We can use sizeof to help with that. If we want to allocate enough room for a single int, we can use sizeof(int) and pass that to malloc().

After we’re done with some allocated memory, we can call free() to indicate we’re done with that memory and it can be used for something else. As an argument, you pass the same pointer you got from malloc() (or a copy of it). It’s undefined behavior to use a memory region after you free() it.

Let’s try. We’ll allocate enough memory for an int, and then store something there, and the print it.

    // Allocate space for a single int (sizeof(int) bytes-worth):
    
    int *p = malloc(sizeof(int));
    
    *p = 12;  // Store something there
    
    printf("%d\n", *p);  // Print it: 12
    
    free(p);  // All done with that memory
    
    //*p = 3490;  // ERROR: undefined behavior! Use after free()!

Now, in that contrived example, there’s really no benefit to it. We could have just used an automatic int and it would have worked. But we’ll see how the ability to allocate memory this way has its advantages, especially with more complex data structures.

One more thing you’ll commonly see takes advantage of the fact that sizeof can give you the size of the result type of any constant expression. So you could put a variable name in there, too, and use that. Here’s an example of that, just like the previous one:

    int *p = malloc(sizeof *p);  // *p is an int, so same as sizeof(int)

12.2 Error Checking

All the allocation functions return a pointer to the newly-allocated stretch of memory, or NULL if the memory cannot be allocated for some reason.

Some OSes like Linux can be configured in such a way that malloc() never returns NULL, even if you’re out of memory. But despite this, you should always code it up with protections in mind.

    int *x;
    
    x = malloc(sizeof(int) * 10);
    
    if (x == NULL) {
        printf("Error allocating 10 ints\n");
        // do something here to handle it
    }

Here’s a common pattern that you’ll see, where we do the assignment and the condition on the same line:

    int *x;
    
    if ((x = malloc(sizeof(int) * 10)) == NULL)
        printf("Error allocating 10 ints\n");
        // do something here to handle it
    }

12.3 Allocating Space for an Array

We’ve seen how to allocate space for a single thing; now what about for a bunch of them in an array?

In C, an array is a bunch of the same thing back-to-back in a contiguous stretch of memory.

We can allocate a contiguous stretch of memory—we’ve seen how to do that. If we wanted 3490 bytes of memory, we could just ask for it:

    char *p = malloc(3490);  // Voila

And—indeed!—that’s an array of 3490 chars (AKA a string!) since each char is 1 byte. In other words, sizeof(char) is 1.

Note: there’s no initialization done on the newly-allocated memory—it’s full of garbage. Clear it with memset() if you want to, or see calloc(), below.

But we can just multiply the size of the thing we want by the number of elements we want, and then access them using either pointer or array notation. Example!

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    // Allocate space for 10 ints
    int *p = malloc(sizeof(int) * 10);

    // Assign them values 0-45:
    for (int i = 0; i < 10; i++)
        p[i] = i * 5;

    // Print all values 0, 5, 10, 15, ..., 40, 45
    for (int i = 0; i < 10; i++)
        printf("%d\n", p[i]);

    // Free the space
    free(p);
}

The key’s in that malloc() line. If we know each int takes sizeof(int) bytes to hold it, and we know we want 10 of them, we can just allocate exactly that many bytes with:

    sizeof(int) * 10

And this trick works for every type. Just pass it to sizeof and multiply by the size of the array.

12.4 An Alternative: `calloc()`

This is another allocation function that works similarly to malloc(), with two key differences:

Instead of a single argument, you pass the size of one element, and the number of elements you wish to allocate. It’s like it’s made for allocating arrays.
It clears the memory to zero.

You still use free() to deallocate memory obtained through calloc().

Here’s a comparison of calloc() and malloc().

    // Allocate space for 10 ints with calloc(), initialized to 0:
    int *p = calloc(sizeof(int), 10);
    
    // Allocate space for 10 ints with malloc(), initialized to 0:
    int *q = malloc(sizeof(int) * 10);
    memset(q, 0, sizeof(int) * 10);   // set to 0

Again, the result is the same for both except malloc() doesn’t zero the memory by default.

12.5 Changing Allocated Size with `realloc()`

If you’ve already allocated 10 ints, but later you decide you need 20, what can you do?

One option is to allocate some new space, and then memcpy() the memory over… but it turns out that sometimes you don’t need to move anything. And there’s one function that’s just smart enough to do the right thing in all the right circumstances: realloc().

It takes a pointer to some previously-allocted memory (by malloc() or calloc()) and a new size for the memory region to be.

It then grows or shrinks that memory, and returns a pointer to it. Sometimes it might return the same pointer (if the data didn’t have to be copied elsewhere), or it might return a different one (if the data did have to be copied).

Be sure when you call realloc(), you specify the number of bytes to allocate, and not just the number of array elements! That is:

    num_floats *= 2;
    
    np = realloc(p, num_floats);  // WRONG: need bytes, not number of elements!
    
    np = realloc(p, num_floats * sizeof(float));  // Better!

Let’s allocate an array of 20 floats, and then change our mind and make it an array of 40.

We’re going to assign the return value of realloc() into another pointer just to make sure it’s not NULL. If it’s not, then we can reassign it into our original pointer. (If we just assigned the return value directly into the original pointer, we’d lose that pointer if the function returned NULL and we’d have no way to get it back.)

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    // Allocate space for 20 floats
    float *p = malloc(sizeof *p * 20);  // sizeof *p same as sizeof(float)

    // Assign them fractional values 0.0-1.0:
    for (int i = 0; i < 20; i++)
        p[i] = i / 20.0;

    // But wait! Let's actually make this an array of 40 elements
    float *new_p = realloc(p, sizeof *p * 40);

    // Check to see if we successfully reallocated
    if (new_p == NULL) {
        printf("Error reallocing\n");
        return 1;
    }

    // If we did, we can just reassign p
    p = new_p;

    // And assign the new elements values in the range 1.0-2.0
    for (int i = 20; i < 40; i++)
        p[i] = 1.0 + (i - 20) / 20.0;

    // Print all values 0.0-2.0 in the 40 elements:
    for (int i = 0; i < 40; i++)
        printf("%f\n", p[i]);

    // Free the space
    free(p);
}

Notice in there how we took the return value from realloc() and reassigned it into the same pointer variable p that we passed in. That’s pretty common to do.

Also if line 7 is looking weird, with that sizeof *p in there, remember that sizeof works on the size of the type of the expression. And the type of *p is float, so that line is equivalent to sizeof(float).

12.5.1 Reading in Lines of Arbitrary Length

I want to demonstrate two things with this full-blown example.

Use of realloc() to grow a buffer as we read in more data.
Use of realloc() to shrink the buffer down to the perfect size after we’ve completed the read.

What we see here is a loop that calls fgetc() over and over to append to a buffer until we see that the last character is a newline.

Once it finds the newline, it shrinks the buffer to just the right size and returns it.

#include <stdio.h>
#include <stdlib.h>

// Read a line of arbitrary size from a file
//
// Returns a pointer to the line.
// Returns NULL on EOF or error.
//
// It's up to the caller to free() this pointer when done with it.
//
// Note that this strips the newline from the result. If you need
// it in there, probably best to switch this to a do-while.

char *readline(FILE *fp)
{
    int offset = 0;   // Index next char goes in the buffer
    int bufsize = 4;  // Preferably power of 2 initial size
    char *buf;        // The buffer
    int c;            // The character we've read in

    buf = malloc(bufsize);  // Allocate initial buffer

    if (buf == NULL)   // Error check
        return NULL;

    // Main loop--read until newline or EOF
    while (c = fgetc(fp), c != '\n' && c != EOF) {

        // Check if we're out of room in the buffer accounting
        // for the extra byte for the NUL terminator
        if (offset == bufsize - 1) {  // -1 for the NUL terminator
            bufsize *= 2;  // 2x the space

            char *new_buf = realloc(buf, bufsize);

            if (new_buf == NULL) {
                free(buf);   // On error, free and bail
                return NULL;
            }

            buf = new_buf;  // Successful realloc
        }

        buf[offset++] = c;  // Add the byte onto the buffer
    }

    // We hit newline or EOF...

    // If at EOF and we read no bytes, free the buffer and
    // return NULL to indicate we're at EOF:
    if (c == EOF && offset == 0) {
        free(buf);
        return NULL;
    }

    // Shrink to fit
    if (offset < bufsize - 1) {  // If we're short of the end
        char *new_buf = realloc(buf, offset + 1); // +1 for NUL terminator

        // If successful, point buf to new_buf;
        // otherwise we'll just leave buf where it is
        if (new_buf != NULL)
            buf = new_buf;
    }

    // Add the NUL terminator
    buf[offset] = '\0';

    return buf;
}

int main(void)
{
    FILE *fp = fopen("foo.txt", "r");

    char *line;

    while ((line = readline(fp)) != NULL) {
        printf("%s\n", line);
        free(line);
    }

    fclose(fp);
}

When growing memory like this, it’s common (though hardly a law) to double the space needed each step just to minimize the number of realloc()s that occur.

Finally you might note that readline() returns a pointer to a malloc()d buffer. As such, it’s up to the caller to explicitly free() that memory when it’s done with it.

12.5.2 `realloc()` with `NULL`

Trivia time! These two lines are equivalent:

    char *p = malloc(3490);
    char *p = realloc(NULL, 3490);

That could be convenient if you have some kind of allocation loop and you don’t want to special-case the first malloc().

    int *p = NULL;
    int length = 0;
    
    while (!done) {
        // Allocate 10 more ints:
        length += 10;
        p = realloc(p, sizeof *p * length);
    
        // Do amazing things
        // ...
    }

In that example, we didn’t need an initial malloc() since p was NULL to start.

12.6 Aligned Allocations

You probably aren’t going to need to use this.

And I don’t want to get too far off in the weeds talking about it right now, but there’s this thing called memory alignment, which has to do with the memory address (pointer value) being a multiple of a certain number.

For example, a system might require that 16-bit values begin on memory addresses that are multiples of 2. Or that 64-bit values begin on memory addresses that are multiples of 2, 4, or 8, for example. It depends on the CPU.

Some systems require this kind of alignment for fast memory access, or some even for memory access at all.

Now, if you use malloc(), calloc(), or realloc(), C will give you a chunk of memory that’s well-aligned for any value at all, even structs. Works in all cases.

But there might be times that you know that some data can be aligned at a smaller boundary, or must be aligned at a larger one for some reason. I imagine this is more common with embedded systems programming.

In those cases, you can specify an alignment with aligned_alloc().

The alignment is an integer power of two greater than zero, so 2, 4, 8, 16, etc. and you give that to aligned_alloc() before the number of bytes you’re interested in.

The other restriction is that the number of bytes you allocate needs to be a multiple of the alignment. But this might be changing. See C Defect Report 460 ⁷⁰

Let’s do an example, allocating on a 64-byte boundary:

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

int main(void)
{
    // Allocate 256 bytes aligned on a 64-byte boundary
    char *p = aligned_alloc(64, 256);  // 256 == 64 * 4

    // Copy a string in there and print it
    strcpy(p, "Hello, world!");
    printf("%s\n", p);

    // Free the space
    free(p);
}

I want to throw a note here about realloc() and aligned_alloc(). realloc() doesn’t have any alignment guarantees, so if you need to get some aligned reallocated space, you’ll have to do it the hard way with memcpy().

Here’s a non-standard aligned_realloc() function, if you need it:

    void *aligned_realloc(void *ptr, size_t old_size, size_t alignment, size_t size)
    {
        char *new_ptr = aligned_alloc(alignment, size);
    
        if (new_ptr == NULL)
            return NULL;
    
        size_t copy_size = old_size < size? old_size: size;  // get min
    
        if (ptr != NULL)
            memcpy(new_ptr, ptr, copy_size);
    
        free(ptr);
    
        return new_ptr;
    }

Note that it always copies data, taking time, while real realloc() will avoid that if it can. So this is hardly efficient. Avoid needing to reallocate custom-aligned data.

13 Scope

Scope is all about what variables are visible in what contexts.

13.1 Block Scope

This is the scope of almost all the variables devs define. It includes what other languages might call “function scope”, i.e. variables that are declared inside functions.

The basic rule is that if you’ve declared a variable in a block delimited by squirrelly braces, the scope of that variable is that block.

If there’s a block inside a block, then variables declared in the inner block are local to that block, and cannot be seen in the outer scope.

Once a variable’s scope ends, that variable can no longer be referenced, and you can consider its value to be gone into the great bit bucket ⁷¹ in the sky.

An example with nested scope:

int main(void)
{
    int a = 12;         // Local to outer block, but visible in inner block

    if  (a == 12) {
        int b = 99;     // Local to inner block, not visible in outer block

        printf("%d %d\n", a, b);  // OK: "12 99"
    }

    printf("%d\n", a);  // OK, we're still in a's scope

    printf("%d\n", b);  // ILLEGAL, out of b's scope
}

13.1.1 Where To Define Variables

Another fun fact is that you can define variables anywhere in the block, within reason—they have the scope of that block, but cannot be used before they are defined.

#include <stdio.h>

int main(void)
{
    int i = 0;

    printf("%d\n", i);     // OK: "0"

    //printf("%d\n", j);   // ILLEGAL--can't use j before it's defined

    int j = 5;

    printf("%d %d\n", i, j);   // OK: "0 5"
}

Historically, C required all the variables be defined before any code in the block, but this is no longer the case in the C99 standard.

13.1.2 Variable Hiding

If you have a variable named the same thing at an inner scope as one at an outer scope, the one at the inner scope takes precedence at long as you’re running in the inner scope. That is, it hides the one at outer scope for the duration of its lifetime.

#include <stdio.h>

int main(void)
{
    int i = 10;

    {
        int i = 20;

        printf("%d\n", i);  // Inner scope i, 20 (outer i is hidden)
    }

    printf("%d\n", i);  // Outer scope i, 10
}

You might have noticed in that example that I just threw a block in there at line 7, not so much as a for or if statement to kick it off! This is perfectly legal. Sometimes a dev will want to group a bunch of local variables together for a quick computation and will do this, but it’s rare to see.

13.2 File Scope

If you define a variable outside of a block, that variable has file scope. It’s visible in all functions in the file that come after it, and shared between them. (An exception is if a block defines a variable of the same name, it would hide the one at file scope.)

This is closest to what you would consider to be “global” scope in another language.

For example:

#include <stdio.h>

int shared = 10;    // File scope! Visible to the whole file after this!

void func1(void)
{
    shared += 100;  // Now shared holds 110
}

void func2(void)
{
    printf("%d\n", shared);  // Prints "110"
}

int main(void)
{
    func1();
    func2();
}

Note that if shared were declared at the bottom of the file, it wouldn’t compile. It has to be declared before any functions use it.

There are ways to further modify items at file scope, namely with (static)[#static] and (extern)[#extern], but we’ll talk more about those later.

13.3 `for`-loop Scope

I really don’t know what to call this, as C11 §6.8.5.3¶1 doesn’t give it a proper name. We’ve done it already a few times in this guide, as well. It’s when you declare a variable inside the first clause of a for-loop:

    for (int i = 0; i < 10; i++)
        printf("%d\n", i);
    
    printf("%d\n", i);  // ILLEGAL--i is only in scope for the for-loop

In that example, i’s lifetime begins the moment it is defined, and continues for the duration of the loop.

If the loop body is enclosed in a block, the variables defined in the for-loop are visible from that inner scope.

Unless, of course, that inner scope hides them. This crazy example prints 999 five times:

#include <stdio.h>

int main(void)
{
    for (int i = 0; i < 5; i++) {
        int i = 999;  // Hides the i in the for-loop scope
        printf("%d\n", i);
    }
}

13.4 A Note on Function Scope

The C spec does refer to function scope, but it’s used exclusively with labels, something we haven’t discussed yet. More on that another day.

14 Types II: Way More Types!

We’re used to char, int, and float types, but it’s now time to take that stuff to the next level and see what else we have out there in the types department!

14.1 Signed and Unsigned Integers

So far we’ve used int as a signed type, that is, a value that can be either negative or positive. But C also has specific unsigned integer types that can only hold positive numbers.

These types are prefaced by the keyword unsigned.

    int a;           // signed
    signed int a;    // signed
    signed a;        // signed, "shorthand" for "int" or "signed int", rare
    unsigned int b;  // unsigned
    unsigned c;      // unsigned, shorthand for "unsigned int"

Why? Why would you decide you only wanted to hold positive numbers?

Answer: you can get larger numbers in an unsigned variable than you can in a signed ones.

But why is that?

You can think of integers being represented by a certain number of bits⁷². On my computer, an int is represented by 64 bits.

And each permutation of bits that are either 1 or 0 represents a number. We can decide how to divvy up these numbers.

With signed numbers, we use (roughly) half the permutations to represent negative numbers, and the other half to represent positive numbers.

With unsigned, we use all the permutations to represent positive numbers.

On my computer with 64-bit ints using two’s complement ⁷³ to represent unsigned numbers, I have the following limits on integer range:

Type	Minimum	Maximum
`int`	`-9,223,372,036,854,775,808`	`9,223,372,036,854,775,807`
`unsigned int`	`0`	`18,446,744,073,709,551,615`

Notice that the largest positive unsigned int is approximately twice as large as the largest positive int. So you can get some flexibility there.

14.2 Character Types

Remember char? The type we can use to hold a single character?

    char c = 'B';
    
    printf("%c\n", c);  // "B"

I have a shocker for you: it’s actually an integer.

    char c = 'B';
    
    // Change this from %c to %d:
    printf("%d\n", c);  // 66 (!!)

Deep down, char is just a small int, namely an integer that uses just a single byte of space, limiting its range to…

Here the C spec gets just a little funky. It assures us that a char is a single byte, i.e. sizeof(char) == 1. But then in C11 §3.6¶3 it goes out of its way to say:

A byte is composed of a contiguous sequence of bits, the number of which is implementation-defined.

Wait—what? Some of you might be used to the notion that a byte is 8 bits, right? I mean, that’s what it is, right? And the answer is, “Almost certainly.”⁷⁴ But C is an old language, and machines back in the day had, shall we say, a more relaxed opinion over how many bits were in a byte. And through the years, C has retained this flexibility.

But assuming your bytes in C are 8 bits, like they are for virtually all machines in the world that you’ll ever see, the range of a char is…

—So before I can tell you, it turns out that chars might be signed or unsigned depending on your compiler. Unless you explicitly specify.

In many cases, just having char is fine because you don’t care about the sign of the data. But if you need signed or unsigned chars, you must be specific:

    char a;           // Could be signed or unsigned
    signed char b;    // Definitely signed
    unsigned char c;  // Definitely unsigned

OK, now, finally, we can figure out the range of numbers if we assume that a char is 8 bits and your system uses the virtually universal two’s complement representation for signed and unsigned⁷⁵.

So, assuming those constraints, we can finally figure our ranges:

`char` type	Minimum	Maximum
`signed char`	`-128`	`127`
`unsigned char`	`0`	`255`

And the ranges for char are implementation-defined.

Let me get this straight. char is actually a number, so can we do math on it?

Yup! Just remember to keep things in the range of a char!

#include <stdio.h>

int main(void)
{
    char a = 10, b = 20;

    printf("%d\n", a + b);  // 30!
}

What about those constant characters in single quotes, like 'B'? How does that have a numeric value?

The spec is also hand-wavey here, since C isn’t designed to run on a single type of underlying system.

But let’s just assume for the moment that your character set is based on ASCII ⁷⁶ for at least the first 128 characters. In that case, the character constant will be converted to a char whose value is the same as the ASCII value of the character.

That was a mouthful. Let’s just have an example:

#include <stdio.h>

int main(void)
{
    char a = 10;
    char b = 'B';  // ASCII value 66

    printf("%d\n", a + b);  // 76!
}

This depends on your execution environment and the character set used ⁷⁷. One of the most popular character sets today is Unicode ⁷⁸ (which is a superset of ASCII), so for your basic 0-9, A-Z, a-z and punctuation, you’ll almost certainly get the ASCII values out of them.

14.3 More Integer Types: `short`, `long`, `long long`

So far we’ve just generally been using two integer types:

char
int

and we recently learned about the unsigned variants of the integer types. And we learned that char was secretly a small int in disguise. So we know the ints can come in multiple bit sizes.

But there are a couple more integer types we should look at, and the minimum minimum and maximum values they can hold.

Yes, I said “minimum” twice. The spec says that these types will hold numbers of at least these sizes, so your implementation might be different. The header file <limits.h> defines macros that hold the minimum and maximum integer values; rely on that to be sure, and never hardcode or assume these values.

These additional types are short int, long int, and long long int. Commonly, when using these types, C developers leave the int part off (e.g. long long), and the compiler is perfectly happy.

    // These two lines are equivalent:
    long long int x;
    long long x;
    
    // And so are these:
    short int x;
    short x;

Let’s take a look at the integer data types and sizes in ascending order, grouped by signedness.

Type	Minimum Bytes	Minimum Value	Maximum Value
`char`	1	-127 or 0	127 or 255⁷⁹
`signed char`	1	-127	127
`short`	2	-32767	32767
`int`	2	-32767	32767
`long`	4	-2147483647	2147483647
`long long`	8	-9223372036854775807	9223372036854775807
`unsigned char`	1	0	255
`unsigned short`	2	0	65535
`unsigned int`	2	0	65535
`unsigned long`	4	0	44294967295
`unsigned long long`	8	0	9223372036854775807

There is no long long long type. You can’t just keep adding longs like that. Don’t be silly.

Two’s complement fans might have noticed something funny about those numbers. Why does, for example, the signed char stop at -127 instead of -128? Remember: these are only the minimums required by the spec. Some number representations (like sign and magnitude ⁸⁰) top off at ±127.

Let’s run the same table on my 64-bit, two’s complement system and see what comes out:

Type	My Bytes	Minimum Value	Maximum Value
`char`	1	-128	127⁸¹
`signed char`	1	-128	127
`short`	2	-32768	32767
`int`	4	-2147483648	2147483647
`long`	8	-9223372036854775808	9223372036854775807
`long long`	8	-9223372036854775808	9223372036854775807
`unsigned char`	1	0	255
`unsigned short`	2	0	65535
`unsigned int`	4	0	4294967295
`unsigned long`	8	0	18446744073709551615
`unsigned long long`	8	0	18446744073709551615

That’s a little more sensible, but we can see how my system has larger limits than the minimums in the specification.

So what are the macros in <limits.h>?

Type	Min Macro	Max Macro
`char`	`CHAR_MIN`	`CHAR_MAX`
`signed char`	`SCHAR_MIN`	`SCHAR_MAX`
`short`	`SHRT_MIN`	`SHRT_MAX`
`int`	`INT_MIN`	`INT_MAX`
`long`	`LONG_MIN`	`LONG_MAX`
`long long`	`LLONG_MIN`	`LLONG_MAX`
`unsigned char`	`0`	`UCHAR_MAX`
`unsigned short`	`0`	`USHRT_MAX`
`unsigned int`	`0`	`UINT_MAX`
`unsigned long`	`0`	`ULONG_MAX`
`unsigned long long`	`0`	`ULLONG_MAX`

Notice there’s a way hidden in there to determine if a system uses signed or unsigned chars. If CHAR_MAX == UCHAR_MAX, it must be unsigned.

Also notice there’s no minimum macro for the unsigned variants—they’re just 0.

14.4 More Float: `double` and `long double`

Let’s see what the spec has to say about floating point numbers in §5.2.4.2.2¶1-2:

The following parameters are used to define the model for each floating-point type:

Parameter Definition

$s$ sign ($\pm1$)

$b$ base or radix of exponent representation (an integer $> 1$)

$e$ exponent (an integer between a minimum $e_{min}$ and a maximum $e_{max}$)

$p$ precision (the number of base-$b$ digits in the significand)

$f_k$ nonnegative integers less than $b$ (the significand digits)

A floating-point number ($x$) is defined by the following model:

$x=sb^e\sum\limits_{k=1}^p f_kb^{-k},$ $e_{min}\le e\le e_{max}$

Parameter	Definition
\(s\)	sign (\(\pm1\))
\(b\)	base or radix of exponent representation (an integer \(> 1\))
\(e\)	exponent (an integer between a minimum \(e_{min}\) and a maximum \(e_{max}\))
\(p\)	precision (the number of base-\(b\) digits in the significand)
\(f_k\)	nonnegative integers less than \(b\) (the significand digits)

I hope that cleared it right up for you.

Okay, fine. Let’s step back a bit and see what’s practical.

Note: we refer to a bunch of macros in this section. They can be found in the header <float.h>.

Floating point number are encoded in a specific sequence of bits (IEEE-754 format ⁸² is tremendously popular) in bytes.

Diving in a bit more, the number is basically represented as the significand (which is the number part—the significant digits themselves, also sometimes referred to as the mantissa) and the exponent, which is what power to raise the digits to. Recall that a negative exponent can make a number smaller.

Imagine we’re using $10$ as a number to raise by an exponent. We could represent the following numbers by using a significand of $12345$, and exponents of $-3$, $4$, and $0$ to encode the following floating point values:

$12345\times10^{-3}=12.345$

$12345\times10^4=123450000$

$12345\times10^0=12345$

For all those numbers, the significand stays the same. The only difference is the exponent.

On your machine, the base for the exponent is probably $2$, not $10$, since computers like binary. You can check it by printing the FLT_RADIX macro.

So we have a number that’s represented by a number of bytes, encoded in some way. Because there are a limited number of bit patterns, a limited number of floating point numbers can be represented.

But more particularly, only a certain number of significant decimal digits can be represented accurately.

How can you get more? You can use larger data types!

And we have a couple of them. We know about float already, but for more precision we have double. And for even more precision, we have long double (unrelated to long int except by name).

The spec doesn’t go into how many bytes of storage each type should take, but on my system, we can see the relative size increases:

Type	`sizeof`
`float`	4
`double`	8
`long double`	16

So each of the types (on my system) uses those additional bits for more precision.

But how much precision are we talking, here? How many decimal numbers can be represented by these values?

Well, C provides us with a bunch of macros in <float.h> to help us figure that out.

It gets a little wonky if you are using a base-2 (binary) system for storing the numbers (which is virtually everyone on the planet, probably including you), but bear with me while we figure it out.

14.4.1 How Many Decimal Digits?

The million dollar question is, “How many significant decimal digits can I store in a given floating point type so that I get out the same decimal number when I print it?”

The number of decimal digits you can store in a floating point type and surely get the same number back out when you print it is given by these macros:

Type	Decimal Digits You Can Store	Minimum
`float`	`FLT_DIG`	6
`double`	`DBL_DIG`	10
`long double`	`LDBL_DIG`	10

On my system, FLT_DIG is 6, so I can be sure that if I print out a 6 digit float, I’ll get the same thing back. (It could be more digits—some numbers will come back correctly with more digits. But 6 is definitely coming back.)

For example, printing out floats following this pattern of increasing digits, we apparently make it to 8 digits before something goes wrong, but after that we’re back to 7 correct digits.

    0.12345
    0.123456
    0.1234567
    0.12345678
    0.123456791  <-- Things start going wrong
    0.1234567910

Let’s do another demo. In this code we’ll have two floats that both hold numbers that have FLT_DIG significant decimal digits⁸³. Then we add those together, for what should be 12 significant decimal digits. But that’s more than we can store in a float and correctly recover as a string—so we see when we print it out, things start going wrong after the 7th significant digit.

#include <stdio.h>
#include <float.h>

int main(void)
{
    // Both these numbers have 6 significant digits, so they can be
    // stored accurately in a float:

    float f = 3.14159f;
    float g = 0.00000265358f;

    printf("%.5f\n", f);   // 3.14159       -- correct!
    printf("%.11f\n", g);  // 0.00000265358 -- correct!

    // Now add them up
    f += g;                // 3.14159265358 is what f _should_ be

    printf("%.11f\n", f);  // 3.14159274101 -- wrong!
}

(The above code has an f after the numeric constants—this indicates that the constant is type float, as opposed to the default of double. More on this later.)

Remember that FLT_DIG is the safe number of digits you can store in a float and retrieve correctly.

Sometimes you might get one or two more out of it. But sometimes you’ll only get FLT_DIG digits back. The sure thing: if you store any number of digits up to and including FLT_DIG in a float, you’re sure to get them back correctly.

So that’s the story. FLT_DIG. The End.

…Or is it?

14.4.2 Converting to Decimal and Back

But storing a base 10 number in a floating point number and getting it back out is only half the story.

Turns out floating point numbers can encode numbers that require more decimal places to print out completely. It’s just that your big decimal number might not map to one of those numbers.

That is, when you look at floating point numbers from one to the next, there’s a gap. If you try to encode a decimal number in that gap, it’ll use the closest floating point number. That’s why you can only encode FLT_DIG for a float.

But what about those floating point numbers that aren’t in the gap? How many places do you need to print those out accurately?

Another way to phrase this question is for any given floating point number, how many decimal digits do I have to preserve if I want to convert the decimal number back into an identical floating point number? That is, how many digits do I have to print in base 10 to recover all the digits in base 2 in the original number?

Sometimes it might only be a few. But to be sure, you’ll want to convert to decimal with a certain safe number of decimal places. That number is encoded in the following macros:

Macro	Description
`FLT_DECIMAL_DIG`	Number of decimal digits encoded in a `float`.
`DBL_DECIMAL_DIG`	Number of decimal digits encoded in a `double`.
`LDBL_DECIMAL_DIG`	Number of decimal digits encoded in a `long double`.
`DECIMAL_DIG`	Same as the widest encoding, `LDBL_DECIMAL_DIG`.

Let’s see an example where DBL_DIG is 15 (so that’s all we can have in a constant), but DBL_DECIMAL_DIG is 17 (so we have to convert to 17 decimal numbers to preserve all the bits of the original double).

Let’s assign the 15 significant digit number 0.123456789012345 to x, and let’s assign the 1 significant digit number 0.0000000000000006 to y.

    x is exact: 0.12345678901234500    Printed to 17 decimal places
    y is exact: 0.00000000000000060

But let’s add them together. This should give 0.1234567890123456, but that’s more than DBL_DIG, so strange things might happen… let’s look:

    x + y not quite right: 0.12345678901234559    Should end in 4560!

That’s what we get for printing more than DBL_DIG, right? But check this out… that number, above, is exactly representable as it is!

If we assign 0.12345678901234559 (17 digits) to z and print it, we get:

    z is exact: 0.12345678901234559   17 digits correct! More than DBL_DIG!

If we’d truncated z down to 15 digits, it wouldn’t have been the same number. That’s why to preserve all the bits of a double, we need DBL_DECIMAL_DIG and not just the lesser DBL_DIG.

All that being said, it’s clear that when we’re messing with decimal numbers in general, it’s not safe to print more than FLT_DIG, DBL_DIG, or LDBL_DIG digits to be sensible in relation to the original base 10 numbers and any subsequent math.

But when converting from float to a decimal representation and back to float, definitely use FLT_DECIMAL_DIG to do that so that all the bits are preserved exactly.

14.5 Constant Numeric Types

When you write down a constant number, like 1234, it has a type. But what type is it? Let’s look at the how C decides what type the constant is, and how to force it to choose a specific type.

14.5.1 Hexadecimal and Octal

In addition to good ol’ decimal like Grandma used to bake, C also supports constants of different bases.

If you lead a number with 0x, it is read as a hex number:

    int a = 0x1A2B;   // Hexadecimal
    int b = 0x1a2b;   // Case doesn't matter for hex digits
    
    printf("%x", a);  // Print a hex number, "1a2b"

If you lead a number with a 0, it is read as an octal number:

    int a = 012;
    
    printf("%o\n", a);  // Print an octal number, "12"

This is particularly problematic for beginner programmers who try to pad decimal numbers on the left with 0 to line things up nice and pretty, inadvertently changing the base of the number:

    int x = 11111;  // Decimal 11111
    int y = 00111;  // Decimal 73 (Octal 111)
    int z = 01111;  // Decimal 585 (Octal 1111)

14.5.1.1 A Note on Binary

An unofficial extension⁸⁴ in many C compilers allows you to represent a binary number with a 0b prefix:

    int x = 0b101010;    // Binary 101010
    
    printf("%d\n", x);   // Prints 42 decimal

There’s no printf() format specifier for printing a binary number. You have to do it a character at a time with bitwise operators.

14.5.2 Integer Constants

You can force a constant integer to be a certain type by appending a suffix to it that indicates the type.

We’ll do some assignments to demo, but most often devs leave off the suffixes unless needed to be precise. The compiler is pretty good at making sure the types are compatible.

    int           x = 1234;
    long int      x = 1234L;
    long long int x = 1234LL
    
    unsigned int           x = 1234U;
    unsigned long int      x = 1234UL;
    unsigned long long int x = 1234ULL;

The suffix can be uppercase or lowercase. And the U and L or LL can appear either one first.

Type	Suffix
`int`	None
`long int`	`L`
`long long int`	`LL`
`unsigned int`	`U`
`unsigned long int`	`UL`
`unsigned long long int`	`ULL`

I mentioned in the table that “no suffix” means int… but it’s actually more complex than that.

So what happens when you have an unsuffixed number like:

    int x = 1234;

What type is it?

What C will generally do is choose the smallest type from int up that can hold the value.

But specifically, that depends on the number’s base (decimal, hex, or octal), as well.

The spec has a great table indicating which type gets used for what unsuffixed value. In fact, I’m just going to copy it wholesale right here.

C11 §6.4.4.1¶5 reads, “The type of an integer constant is the first of the first of the corresponding list in which its value can be represented.”

And then goes on to show this table:

Suffix	Decimal Constant	Octal or Hexadecimal Constant
none	`int` `long int`	`int` `unsigned int` `long int` `unsigned long int` `long long int` `unsigned long long int`
`u` or `U`	`unsigned int` `unsigned long int` `unsigned long long int`	`unsigned int` `unsigned long int` `unsigned long long int`
`l` or `L`	`long int` `long long int`	`long int` `unsigned long int` `long long int` `unsigned long long int`
Both `u` or `U` and `l` or `L`	`unsigned long int` `unsigned long long int`	`unsigned long int` `unsigned long long int`
`ll` or `LL`	`long long int`	`long long int` `unsigned long long int`
Both `u` or `U` and `ll` or `LL`	`unsigned long long int`	`unsigned long long int`

What that’s saying is that, for example, if you specify a number like 123456789U, first C will see if it can be unsigned int. If it doesn’t fit there, it’ll try unsigned long int. And then unsigned long long int. It’ll use the smallest type that can hold the number.

14.5.3 Floating Point Constants

You’d think that a floating point constant like 1.23 would have a default type of float, right?

Surprise! Turns out unsuffiexed floating point numbers are type double! Happy belated birthday!

You can force it to be of type float by appending an f (or F—it’s case-insensitive). You can force it to be of type long double by appending l (or L).

Type	Suffix
`float`	`F`
`double`	None
`long double`	`L`

For example:

    float x       = 3.14f;
    double x      = 3.14;
    long double x = 3.14L;

This whole time, though, we’ve just been doing this, right?

    float x = 3.14;

Isn’t the left a float and the right a double? Yes! But C’s pretty good with automatic numeric conversions, so it’s more common to have an unsuffixed floating point constant than not. More on that later.

14.5.3.1 Scientific Notation

Remember earlier when we talked about how a floating point number can be represented by a significand, base, and exponent?

Well, there’s a common way of writing such a number, shown here followed by it’s more recognizable equivalent which is what you get when you actually run the math:

$1.2345\times10^3 = 1234.5$

Writing numbers in the form $s\times b^e$ is called scientific notation⁸⁵. In C, these are written using “E notation”, so these are equivalent:

Scientific Notation	E notation
$1.2345\times10^{-3}=12.345$	`1.2345e-3`
$1.2345\times10^4=123450000$	`1.2345e+4`

You can print a number in this notation with %e:

    printf("%e\n", 123456.0);  // Prints 1.234560e+05

A couple little fun facts about scientific notation:

You don’t have to write them with a single leading digit before the decimal point. Any number of numbers can go in front.
```
    double x = 123.456e+3;  // 123456
```
However, when you print it, it will change the exponent so there is only one digit in front of the decimal point.
The plus can be left off the exponent, as it’s default, but this is uncommon in practice from what I’ve seen.
```
    1.2345e10 == 1.2345e+10
```
You can apply the F or L suffixes to E-notation constants:
```
    1.2345e10F
    1.2345e10L
```

14.5.3.2 Hexadecimal Floating Point Constants

But wait, there’s more floating to be done!

Turns out there are hexadecimal floating point constants, as well!

These work similar to decimal floating point numbers, but they begin with a 0x just like integer numbers.

The catch is that you must specify an exponent, and this exponent produces a power of 2. That is: $2^x$.

And then you use a p instead of an e when writing the number:

So 0xa.1p3 is $10.0625\times2^3 == 80.5$.

When using floating point hex constants, We can print hex scientific notation with %a:

    double x = 0xa.1p3;
    
    printf("%a\n", x);  // 0x1.42p+6
    printf("%f\n", x);  // 80.500000

15 Types III: Conversions

In this chapter, we want to talk all about converting from one type to another. C has a variety of ways of doing this, and some might be a little different that you’re used to in other languages.

Before we talk about how to make conversions happen, let’s talk about how they work when they do happen.

15.1 String Conversions

Unlike many languages, C doesn’t do string-to-number (and vice-versa) conversions in quite as streamlined a manner as it does numeric conversions.

For these, we’ll have to call functions to do the dirty work.

15.1.1 Numeric Value to String

When we want to convert a number to a string, we can use either sprintf() (pronounced SPRINT-f) or snprintf() (s-n-print-f)⁸⁶

These basically work like printf(), except they output to a string instead, and you can print that string later, or whatever.

For example, turning part of the value π into a string:

#include <stdio.h>

int main(void)
{
    char s[10];
    float f = 3.14159;

    // Convert "f" to string, storing in "s", writing at most 10 characters
    // including the NUL terminator

    snprintf(s, 10, "%f", f);

    printf("String value: %s\n", s);  // String value: 3.141590
}

If we wanted to convert a double, we’d use %lf. Or a long double, %Lf.

15.1.2 String to Numeric Value

There are a couple families of functions to do this in C. We’ll call these the atoi (pronounced a-to-i) family and the strtol (stir-to-long) family.

For basic conversion from a string to a number, try the atoi functions from <stdlib.h>. These have bad error-handling characteristics (including undefined behavior if you pass in a bad string), so use them carefully.

Function	Description
`atoi`	String to `int`
`atof`	String to `float`
`atol`	String to `long int`
`atoll`	String to `long long int`

Though the spec doesn’t cop to it, the a at the beginning of the function stands for ASCII ⁸⁷, so really atoi() is “ASCII-to-integer”, but saying so today is a bit ASCII-centric.

Here’s an example converting a string to a float:

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *pi = "3.14159";
    float f;

    f = atof(pi);

    printf("%f\n", f);
}

But, like I said, we get undefined behavior from weird things like this:

    int x = atoi("what");  // "What" ain't no number I ever heard of

(When I run that, I get 0 back, but you really shouldn’t count on that in any way. You could get something completely different.)

For better error handling characteristics, let’s check out all those strtol functions, also in <stdlib.h>. Not only that, but they convert to more types and more bases, too!

Function	Description
`strtol`	String to `long int`
`strtoll`	String to `long long int`
`strtoul`	String to `unsigned long int`
`strtoull`	String to `unsigned long long int`
`strtof`	String to `float`
`strtod`	String to `double`
`strtold`	String to `long double`

These functions all follow a similar pattern of use, and are a lot of people’s first experience with pointers to pointers! But never fret—it’s easier than it looks.

Let’s do an example where we convert a string to an unsigned long, discarding error information (i.e. information about bad characters in the input string):

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *s = "3490";

    // Convert string s, a number in base 10, to an unsigned long int.
    // NULL means we don't care to learn about any error information.

    unsigned long int x = strtoul(s, NULL, 10);

    printf("%lu\n", x);  // 3490
}

Notice a couple things there. Even though we didn’t deign to capture any information about error characters in the string, strtoul() won’t give us undefined behavior; it will just return 0.

Also, we specified that this was a decimal (base 10) number.

Does this mean we can convert numbers of different bases? Sure! Let’s do binary!

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *s = "101010";  // What's the meaning of this number?

    // Convert string s, a number in base 2, to an unsigned long int.

    unsigned long int x = strtoul(s, NULL, 2);

    printf("%lu\n", x);  // 42
}

OK, that’s all fun and games, but what’s with that NULL in there? What’s that for?

That helps us figure out if an error occurred in the processing of the string. It’s a pointer to a pointer to a char, which sounds scary, but isn’t once you wrap your head around it.

Let’s do an example where we feed in a deliberately bad number, and we’ll see how strtol() lets us know where the first invalid digit is.

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *s = "34x90";  // "x" is not a valid digit in base 10!
    char *badchar;

    // Convert string s, a number in base 10, to an unsigned long int.

    unsigned long int x = strtoul(s, &badchar, 10);

    // It tries to convert as much as possible, so gets this far:

    printf("%lu\n", x);  // 34

    // But we can see the offending bad character because badchar
    // points to it!

    printf("Invalid character: %c\n", *badchar);  // "x"
}

So there we have strtoul() modifying what badchar points to in order to show us where things went wrong⁸⁸.

But what if nothing goes wrong? In that case, badchar will point to the NUL terminator at the end of the string. So we can test for it:

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *s = "3490";  // "x" is not a valid digit in base 10!
    char *badchar;

    // Convert string s, a number in base 10, to an unsigned long int.

    unsigned long int x = strtoul(s, &badchar, 10);

    // Check if things went well

    if (*badchar == '\0') {
        printf("Success! %lu\n", x);
    } else  {
        printf("Partial conversion: %lu\n", x);
        printf("Invalid character: %c\n", *badchar);
    }
}

So there you have it. The atoi()-style functions are good in a controlled pinch, but the strtol()-style functions give you far more control over error handling and the base of the input.

15.2 Numeric Conversions

15.2.1 Boolean

If you convert a zero to bool, the result is 0. Otherwise it’s 1.

15.2.2 Integer to Integer Conversions

If an integer type is converted to unsigned and doesn’t fit in it, the unsigned result wraps around odometer-style until it fits in the unsigned⁸⁹.

If an integer type is converted to a signed number and doesn’t fit, the result is implementation-defined! Something documented will happen, but you’ll have to look it up⁹⁰

15.2.3 Integer and Floating Point Conversions

If a floating point type is converted to an integer type, the fractional part is discarded with prejudice⁹¹.

But—and here’s the catch—if the number is too large to fit in the integer, you get undefined behavior. So don’t do that.

Going From integer or floating point to floating point, C makes a best effort to find the closest floating point number to the integer that it can.

Again, though, if the original value can’t be represented, it’s undefined behavior.

15.3 Implicit Conversions

These are conversions the compiler does automatically for you when you mix and match types.

15.3.1 The Integer Promotions

In a number of places, if a int can be used to represent a value from char or short (signed or unsigned), that value is promoted up to int. If it doesn’t fit in an int, it’s promoted to unsigned int.

This is how we can do something like this:

    char x = 10, y = 20;
    int i = x + y;

In that case, x and y get promoted to int by C before the math takes place.

The integer promotions take place during The Usual Arithmetic Conversions, with variadic functions⁹², unary + and - operators, or when passing values to functions without prototypes⁹³.

15.3.2 The Usual Arithmetic Conversions

These are automatic conversions that C does around numeric operations that you ask for. (That’s actually what they’re called, by the way, by C11 §6.3.1.8.) Note that for this section, we’re just talking about numeric types—strings will come later.

These conversions answer questions about what happens when you mix types, like this:

    int x = 3 + 1.2;   // Mixing int and double
                       // 4.2 is converted to int
                       // 4 is stored in x
    
    float y = 12 * 2;  // Mixing float and int
                       // 24 is converted to float
                       // 24.0 is stored in y

Do they become ints? Do they become floats? How does it work?

Here are the steps, paraphrased for easy consumption.

If one thing in the expression is a floating type, convert the other things to that floating type.
Otherwise, if both types are integer types, perform the integer promotions on each, then make the operand types as big as they need to be hold the common largest value. Sometimes this involves changing signed to unsigned.

If you want to know the gritty details, check out C11 §6.3.1.8. But you probably don’t.

Just generally remember that int types become float types if there’s a floating point type anywhere in there, and the compiler makes an effort to make sure mixed integer types don’t overflow.

Finally, if you convert from one floating point type to another, the compiler will try to make an exact conversion. If it can’t, it’ll do the best approximation it can. If the number is too large to fit in the type you’re converting into, boom: undefined behavior!

15.3.3 `void*`

The void* type is interesting because it can be converted from or to any pointer type.

    int x = 10;
    
    void *p = &x;  // &x is type int*, but we store it in a void*
    
    int *q = p;    // p is void*, but we store it in an int*

15.4 Explicit Conversions

These are conversions from type to type that you have to ask for; the compiler won’t do it for you.

You can convert from one type to another by assigning one type to another with an =.

You can also convert explicitly with a cast.

15.4.1 Casting

You can explicitly change the type of an expression by putting a new type in parentheses in front of it. Some C devs frown on the practice unless absolutely necessary, but it’s likely you’ll come across some C code with these in it.

Let’s do an example where we want to convert an int into a long so that we can store it in a long.

Note: this example is contrived and the cast in this case is completely unnecessary because the x + 12 expression would automatically be changed to long int to match the wider type of y.

    int x = 10;
    long int y = (long int)x + 12;

In that example, even those x was type int before, the expression (long int)x has type long int. We say, “We cast x to long int.”

More commonly, you might see a cast being used to convert a void* into a specific pointer type so it can be dereferenced.

A callback from the built-in qsort() function might display this behavior since it has void*s passed into it:

    int compar(const void *elem1, const void *elem2)
    {
        return *((const int*)elem2) - *((const int*)elem1);
    }

But you could also clearly write it with an assignment:

    int compar(const void *elem1, const void *elem2)
    {
        const int *e1 = elem1;
        const int *e2 = elem2;
    
        return *e2 - *e1;
    }

One place you’ll see casts more commonly is to avoid a warning when printing pointer values with the rarely-used %p which gets picky with anything other than a void*:

    int x = 3490;
    int *p = &x;
    
    printf("%p\n", p);

generates this warning:

    warning: format ‘%p’ expects argument of type ‘void *’, but argument
             2 has type ‘int *’

You can fix it with a cast:

    printf("%p\n", (void *)p);

Another place is with explicit pointer changes, if you don’t want to use an intervening void*, but these are also pretty uncommon:

    long x = 3490;
    long *p = &x;
    unsigned char *c = (unsigned char *)p;

Again, casting is rarely needed in practice. If you find yourself casting, there might be another way to do the same thing, or maybe you’re casting unnecessarily.

Or maybe it is necessary. Personally, I try to avoid it, but am not afraid to use it if I have to.

16 Types IV: Qualifiers and Specifiers

Now that we have some more types under our belts, turns out we can give these types some additional attributes that control their behavior. These are the type qualifiers and storage-class specifiers.

16.1 Type Qualifiers

These are going to allow you to declare constant values, and also to give the compiler optimization hints that it can use.

16.1.1 `const`

This is the most common type qualifier you’ll see. It means the variable is constant, and any attempt to modify it will result in a very angry compiler.

    const int x = 2;
    
    x = 4;  // COMPILER PUKING SOUNDS, can't assign to a constant

You can’t change a const value.

Often you see const in parameter lists for functions:

    void foo(const int x)
    {
        printf("%d\n", x + 30);  // OK, doesn't modify "x"
    }

16.1.1.1 `const` and Pointers

This one gets a little funky, because there are two usages that have two meanings when it comes to pointers.

For one thing, we can make it so you can’t change the thing the pointer points to. You do this by putting the const up front with the type name (before the asterisk) in the type declaration.

    int x[] = {10, 20};
    const int *p = x; 
    
    p++;  // We can modify p, no problem
    
    *p = 30; // Compiler error! Can't change what it points to

Somewhat confusingly, these two things are equivalent:

    const int *p;  // Can't modify what p points to
    int const *p;  // Can't modify what p points to, just like the previous line

Great, so we can’t change the thing the pointer points to, but we can change the pointer itself. What if we want the other way around? We want to be able to change what the pointer points to, but not the pointer itself?

Just move the const after the asterisk in the declaration:

    int *const p;   // We can't modify "p" with pointer arithmetic
    
    p++;  // Compiler error!

But we can modify what they point to:

    int x = 10;
    int *const p = &x;
    
    *p = 20;   // Set "x" to 20, no problem

You can also do make both things const:

    const int *const p;  // Can't modify p or *p!

Finally, if you have multiple levels of indirection, you should const the appropriate levels. Just because a pointer is const, doesn’t mean the pointer it points to must also be. You can explicitly set them like in the following examples:

    char **p;
    p++;     // OK!
    (*p)++;  // OK!
    
    char **const p;
    p++;     // Error!
    (*p)++;  // OK!
    
    char *const *p;
    p++;     // OK!
    (*p)++;  // Error!
    
    char *const *const p;
    p++;     // Error!
    (*p)++;  // Error!

16.1.1.2 `const` Correctness

One more thing I have to mention is that the compiler will warn on something like this:

    const int x = 20;
    int *p = &x;

saying something to the effect of:

    initialization discards 'const' qualifier from pointer type target

What’s happening there?

Well, we need to look at the types on either side of the assignment:

        const int x = 20;
        int *p = &x;
    //    ^       ^
    //    |       |
    //  int*    const int*

The compiler is warning us that the value on the right side of the assignment is const, but the one of the left is not. And the compiler is letting us know that it is discarding the “const-ness” of the expression on the right.

That is, we can still try to do the following, but it’s just wrong. The compiler will warn, and it’s undefined behavior:

    const int x = 20;
    int *p = &x;
    
    *p = 40;  // Undefined behavior--maybe it modifies "x", maybe not!
    
    printf("%d\n", x);  // 40, if you're lucky

16.1.2 `restrict`

TLDR: you never have to use this and you can ignore it every time you see it.

restrict is a hint to the compiler that a particular piece of memory will only be accessed by one pointer and never another. If a developer declares a pointer to be restrict and then accesses the object it points to in another way, the behavior is undefined.

Basically you’re telling C, “Hey—I guarantee that this one single pointer is the only way I access this memory, and if I’m lying, you can pull undefined behavior on me.”

And C uses that information to perform certain optimizations.

For example, let’s write a function to swap two variables, and we’ll use the restrict keyword to assure C that we’ll never pass in pointers to the same thing. And then let’s blow it an try passing in pointers to the same thing.

void swap(int *restrict a, int *restrict b)
{
    int t;

    t = *a;
    *a = *b;
    *b = t;
}

int main(void)
{
    int x = 10, y = 20;

    swap(&x, &y);  // OK! "a" and "b", above, point to different things

    swap(&x, &x);  // Undefined behavior! "a" and "b" point to the same thing
}

If we were to take out the restrict keywords, above, that would allow both calls to work safely. But then the compiler might not be able to optimize.

restrict has block scope, that is, the restriction only lasts for the scope its used. If it’s in a parameter list for a function, it’s in the block scope of that function.

If the restricted pointer points to an array, the restriction covers the entire array.

If it’s outside any function in file scope, the restriction covers the entire program.

You’re likely to see this in library functions like printf():

    int printf(const char * restrict format, ...);

Again, that’s just telling the compiler that inside the printf() function, there will be only one pointer that refers to any part of that format string.

16.1.3 `volatile`

You’re unlikely to see or need this unless you’re dealing with hardware directly.

volatile tells the compiler that a value might change behind its back and should be looked up every time.

An example might be where the compiler is looking in memory at an address that continuously updates behind the scenes, e.g. some kind of hardware timer.

If the compiler decides to optimize that and store the value in a register for a protracted time, the value in memory will update and won’t be reflected in the register.

By declaring something volatile, you’re telling the compiler, “Hey, the thing this points at might change at any time for reasons outside this program code.”

    volatile int *p;

16.1.4 `_Atomic`

This is an optional C feature that we’ll talk about another time.

16.2 Storage-Class Specifiers

Storage-class specifiers are similar to type quantifiers. They give the compiler more information about the type of a variable.

16.2.1 `auto`

You barely ever see this keyword, since auto is the default for block scope variables. It’s implied.

These are the same:

    {
        int a;         // auto is the default...
        auto int a;    // So this is redundant
    }

The auto keyword indicates that this object has automatic storage duration. That is, it exists in the scope in which it is defined, and is automatically deallocated when the scope is exited.

One gotcha about automatic variables is that their value is indeterminate until you explicitly initialize them. We say they’re full of “random” or “garbage” data, though neither of those really makes me happy. In any case, you won’t know what’s in it unless you initialize it.

Always initialize all automatic variables before use!

16.2.2 `static`

This keyword has two meanings, depending on if the variable is file scope or block scope.

Let’s start with block scope.

16.2.2.1 `static` in Block Scope

In this case, we’re basically saying, “I just want a single instance of this variable to exist, shared between calls.”

That is, its value will persist between calls.

static in block scope with an initializer will only be initialized one time on program startup, not each time the function is called.

Let’s do an example:

#include <stdio.h>

void counter(void)
{
    static int count = 1;  // This is initialized one time

    printf("This has been called %d time(s)\n", count);

    count++;
}

int main(void)
{
    counter();  // "This has been called 1 time(s)"
    counter();  // "This has been called 2 time(s)"
    counter();  // "This has been called 3 time(s)"
    counter();  // "This has been called 4 time(s)"
}

See how the value of count persists between calls?

One thing of note is that static block scope variables are initialized to 0 by default.

    static int foo;      // Default starting value is `0`...
    static int foo = 0;  // So the `0` assignment is redundant

Finally, be advised that if you’re writing multithreaded programs, you have to be sure you don’t let multiple threads trample the same variable.

16.2.2.2 `static` in File Scope

When you get out to file scope, outside any blocks, the meaning rather changes.

Variables at file scope already persist between function calls, so that behavior is already there.

Instead what static means in this context is that this variable isn’t visible outside of this particular source file. Kinda like “global”, but only in this file.

More on that in the section about building with multiple source files.

16.2.3 `extern`

The extern storage-class specifier gives us a way to refer to objects in other source files.

Let’s say, for example, the file bar.c had the following as its entirety:

// bar.c

int a = 37;

Just that. Declaring a new int a in file scope.

But what if we had another source file, foo.c, and we wanted to refer to the a that’s in bar.c?

It’s easy with the extern keyword:

// foo.c

extern int a;

int main(void)
{
    printf("%d\n", a);  // 37, from bar.c!

    a = 99;

    printf("%d\n", a);  // Same "a" from bar.c, but it's now 99
}

We could have also made the extern int a in block scope, and it still would have referred to the a in bar.c:

// foo.c

int main(void)
{
    extern int a;

    printf("%d\n", a);  // 37, from bar.c!

    a = 99;

    printf("%d\n", a);  // Same "a" from bar.c, but it's now 99
}

Now, if a in bar.c had been marked static. this wouldn’t have worked. static variables at file scope are not visible outside that file.

A final note about extern on functions. For functions, extern is the default, so it’s redundant. You can declare a function static if you only want it visible in a single source file.

16.2.4 `register`

This is a keyword to hint to the compiler that this variable is frequently-used, and should be made as fast as possible to access. The compiler is under no obligation to agree to it.

Now, modern C compiler optimizers are pretty effective at figuring this out themselves, so it’s rare to see these days.

But if you must:

#include <stdio.h>

int main(void)
{
    register int a;   // Make "a" as fast to use as possible.

    for (a = 0; a < 10; a++)
        printf("%d\n", a);
}

It does come at a price, however. You can’t take the address of a register:

    register int a;
    int *p = &a;    // COMPILER ERROR! Can't take address of a register

The same applies to any part of an array:

    register int a[] = {11, 22, 33, 44, 55};
    int p = a;  // COMPILER ERROR! Can't take address of a[0]

Or dereferencing part of an array:

    register int a[] = {11, 22, 33, 44, 55};
    
    int a = *(a + 2);  // COMPILER ERROR! Address of a[0] taken

Interestingly, for the equivalent with array notation, gcc only warns:

    register int a[] = {11, 22, 33, 44, 55};
    
    int a = a[2];  // COMPILER WARNING!

with:

    warning: ISO C forbids subscripting ‘register’ array

The fact that you can’t take the address of a register variable frees the compiler up to make optimizations around that assumption if it hasn’t figured them out already. Also adding register to a const variable prevents one from accidentally passing its pointer to another function that willfully ignore its constness⁹⁴.

A bit of historic backstory, here: deep inside the CPU are little dedicated “variables” called registers⁹⁵. They are super fast to access compared to RAM, so using them gets you a speed boost. But they’re not in RAM, so they don’t have an associated memory address (which is why you can’t take the address-of or get a pointer to them).

But, like I said, modern compilers are really good at producing optimal code, using registers whenever possible regardless of whether or not you specified the register keyword. Not only that, but the spec allows them to just treat it as if you’d typed auto, if they want. So no guarantees.

16.2.5 `_Thread_local`

When you’re using multiple threads and you have some variables in either global or static block scope, this is a way to make sure that each thread gets its own copy of the variable. This’ll help you avoid race conditions and threads stepping on each other’s toes.

If you’re in block scope, you have to use this along with either extern or static.

Also, if you include <threads.h>, you can use the rather more palatable thread_local as an alias for the uglier _Thread_local.

More information can be found in the Threads section.

17 Multifile Projects

So far we’ve been looking at toy programs that for the most part fit in a single file. But complex C programs are made up of many files that are all compiled and linked together into a single executable.

In this chapter we’ll check out some of the common patterns and practices for putting together larger projects.

17.1 Includes and Function Prototypes

A really common situation is that some of your functions are defined in one file, and you want to call them from another.

This actually works out of the box with a warning… let’s first try it and then look at the right way to fix the warning.

For these examples, we’ll put the filename as the first comment in the source.

To compile them, you’ll need to specify all the sources on the command line:

    # output file   source files
    #     v            v
    #   |----| |---------|
    gcc -o foo foo.c bar.c

In that examples, foo.c and bar.c get built into the executable named foo.

So let’s take a look at the source file bar.c:

// File bar.c

int add(int x, int y)
{
    return x + y;
}

And the file foo.c with main in it:

// File foo.c

#include <stdio.h>

int main(void)
{
    printf("%d\n", add(2, 3));  // 5!
}

See how from main() we call add()—but add() is in a completely different source file! It’s in bar.c, while the call to it is in foo.c!

If we build this with:

    gcc -o foo foo.c bar.c

we get this warning:

    warning: implicit declaration of function ‘add’

But if we ignore that (which really we should never do—always get your code to build with zero warnings!) and try to run it:

    ./foo
    5

Indeed, we get the result of $2+3$! Yay!

So… about that warning. Let’s fix it.

What implicit declaration means is that we’re using a function, namely add() in this case, without letting C know anything about it ahead of time. C wants to know what it returns, what types it takes as arguments, and things such as that.

We saw how to fix that earlier with a function prototype. Indeed, if we add one of those to foo.c before we make the call, everything works well:

// File foo.c

#include <stdio.h>

int add(int, int);  // Add the prototype

int main(void)
{
    printf("%d\n", add(2, 3));  // 5!
}

No more warning!

But that’s a pain—needing to type in the prototype every time you want to use a function. I mean, we used printf() right there and didn’t need to type in a prototype; what gives?

If you remember from what back with hello.c at the beginning of the book, we actually did include the prototype for printf()! It’s in the file stdio.h! And we included that with #include!

Can we do the same with our add() function? Make a prototype for it and put it in a header file?

Sure!

Header files in C have a .h extension by default. And they often, but not always, have the same name as their corresponding .c file. So let’s make a bar.h file for our bar.c file, and we’ll stick the prototype in it:

// File bar.h

int add(int, int);

And now let’s modify foo.c to include that file. Assuming it’s in the same directory, we include it inside double quotes (as opposed to angle brackets):

// File foo.c

#include <stdio.h>

#include "bar.h"  // Include from current directory

int main(void)
{
    printf("%d\n", add(2, 3));  // 5!
}

Notice how we don’t have the prototype in foo.c anymore—we included it from bar.h. Now any file that wants that add() functionality can just #include "bar.h" to get it, and you don’t need to worry about typing in the function prototype.

As you might have guessed, #include literally includes the named file right there in your source code, just as if you’d typed it in.

We’re almost there! There’s just one more piece of boilerplate we have to add.

17.2 Dealing with Repeated Includes

It’s not uncommon that a header file will itself #include other headers needed for the functionality of its corresponding C files. I mean, why not?

And it could be that you have a header #included multiple times from different places. Maybe that’s no problem, but maybe it would cause compiler errors. And we can’t control how many places #include it!

Even, worse we might get into a crazy situation where header a.h includes header b.h, and b.h includes a.h! It’s an #include infinite cycle!

Trying to build such a thing gives an error:

    error: #include nested depth 200 exceeds maximum of 200

What we need to do is make it so that if a file gets included once, subsequent #includes for that file are ignored.

The stuff that we’re about to do is so common that you should just automatically do it every time you make a header file!

And the common way to do this is with a preprocessor variable that we set the first time we #include the file. And then for subsequent #includes, we first check to make sure that the variable isn’t defined.

For that variable name, it’s super common to take the name of the header file, like bar.h, make it uppercase, and replace the period with an underscore: BAR_H.

So put a check at the very, very top of the file where you see if it’s already been included, and effectively comment the whole thing out if it has.

(Don’t put a leading underscore (because a leading underscore followed by a capital letter is reserved) or a double leading underscore (because that’s also reserved.))

#ifndef BAR_H   // If BAR_H isn't defined...
#define BAR_H   // Define it (with no particular value)

// File bar.h

int add(int, int);

#endif          // End of the #ifndef BAR_H

This will effectively cause the header file to be included only a single time, no matter how many places try to #include it.

17.3 `static` and `extern`

When it comes to multifile projects, you can make sure file-scope variables and functions are not visible from other source files with the static keyword.

And you can refer to objects in other files with extern.

For more info, check out the sections in the book on the static and extern storage-class specifiers.

17.4 Compiling with Object Files

This isn’t part of the spec, but it’s 99.999% common in the C world.

You can compile C files into an intermediate representation called object files. These are compiled machine code that hasn’t been put into an executable yet.

Object files in Windows have a .OBJ extension; in Unix-likes, they’re .o.

In gcc, we can build some like this, with the -c (compile only!) flag:

    gcc -c foo.c     # produces foo.o
    gcc -c bar.c     # produces bar.o

And then we can link those together into a single executable:

    gcc -o foo foo.o bar.o

Voila, we’ve produced an executable foo from the two object files.

But you’re thinking, why bother? Can’t we just:

    gcc -o foo foo.c bar.c

and kill two boids ⁹⁶ with one stone?

For little programs, that’s fine. I do it all the time.

But for larger programs, we can take advantage of the fact that compiling from source to object files is relatively slow, and linking together a bunch of object files is relatively fast.

This really shows with the make utility that only rebuilds sources that are newer than their outputs.

Let’s say you had a thousand C files. You could compile them all to object files to start (slowly) and then combine all those object files into an executable (fast).

Now say you modified just one of those C source files—here’s the magic: you only have to rebuild that one object file for that source file! And then you rebuild the executable (fast). All the other C files don’t have to be touched.

In other words, by only rebuilding the object files we need to, we cut down on compilation times radically. (Unless of course you’re doing a “clean” build, in which case all the object files have to be created.)

18 The Outside Environment

When you run a program, it’s actually you talking to the shell, saying, “Hey, please run this thing.” And the shell says, “Sure,” and then tells the operating system, “Hey, could you please make a new process and run this thing?” And if all goes well, the OS complies and your program runs.

But there’s a whole world outside your program in the shell that can be interacted with from within C. We’ll look at a few of those in this chapter.

18.1 Command Line Arguments

Many command line utilities accept command line arguments. For example, if we want to see all files that end in .txt, we can type something like this on a Unix-like system:

    ls *.txt

(or dir instead of ls on a Windows system).

In this case, the command is ls, but it arguments are all all files that end with .txt⁹⁷.

So how can we see what is passed into program from the command line?

Say we have a program called add that adds all numbers passed on the command line and prints the result:

    ./add 10 30 5
    45

That’s gonna pay the bills for sure!

But seriously, this is a great tool for seeing how to get those arguments from the command line and break them down.

First, let’s see how to get them at all. For this, we’re going to need a new main()!

Here’s a program that prints out all the command line arguments. For example, if we name the executable foo, we can run it like this:

    ./foo i like turtles

and we’ll see this output:

    arg 0: ./foo
    arg 1: i
    arg 2: like
    arg 3: turtles

It’s a little weird, because the zeroth argument is the name of the executable, itself. But that’s just something to get used to. The arguments themselves follow directly.

Source:

#include <stdio.h>

int main(int argc, char *argv[])
{
    for (int i = 0; i < argc; i++) {
        printf("arg %d: %s\n", i, argv[i]);
    }
}

Whoa! What’s going on with the main() function signature? What’s argc and argv⁹⁸ (pronounced arg-c and arg-v)?

Let’s start with the easy one first: argc. This is the argument count, including the program name, itself. If you think of all the arguments as an array of strings, which is exactly what they are, then you can think of argc as the length of that array, which is exactly what it is.

And so what we’re doing in that loop is going through all the argvs and printing them out one at a time, so for a given input:

    ./foo i like turtles

we get a corresponding output:

    arg 0: ./foo
    arg 1: i
    arg 2: like
    arg 3: turtles

With that in mind, we should be good to go with our adder program.

Our plan:

Look at all the command line arguments (past argv[0], the program name)
Convert them to integers
Add them to a running total
Print the result

Let’s get to it!

#include <stdio.h>
#include <stdlib.h>

int main(int argc, char **argv)
{
    int total = 0;

    for (int i = 1; i < argc; i++) {  // Start at 1, the first argument
        int value = atoi(argv[i]);    // Use strtol() for better error handling

        total += value;
    }

    printf("%d\n", total);
}

Sample runs:

    $ ./add
    0
    $ ./add 1
    1
    $ ./add 1 2
    3
    $ ./add 1 2 3
    6
    $ ./add 1 2 3 4
    10

Of course, it might puke if you pass in a non-integer, but hardening against that is left as an exercise to the reader.

18.1.1 The Last `argv` is `NULL`

One bit of fun trivia about argv is that after the last string is a pointer to NULL.

That is:

    argv[argc] == NULL

is always true!

This might seem pointless, but it turns out to be useful in a couple places; we’ll take a look at one of those right now.

18.1.2 The Alternate: `char **argv`

Remember that when you call a function, C doesn’t differentiate between array notation and pointer notation in the function signature.

That is, these are the same:

    void foo(char a[])
    void foo(char *a)

Now, it’s been convenient to think of argv as an array of strings, i.e. an array of char*s, so this made sense:

    int main(int argc, char *argv[])

but because of the equivalence, you could also write:

    int main(int argc, char **argv)

Yeah, that’s a pointer to a pointer, all right! If it makes it easier, think of it as a pointer to a string. But really, it’s a pointer to a value that points to a char.

Also recall that these are equivalent:

    argv[i]
    *(argv + i)

which means you can do pointer arithmetic on argv.

So an alternate way to consume the command line arguments might be to just walk along the argv array by bumping up a pointer until we hit that NULL at the end.

Let’s modify our adder to do that:

#include <stdio.h>
#include <stdlib.h>

int main(int argc, char **argv)
{
    int total = 0;
    
    // Cute trick to get the compiler to stop warning about the
    // unused variable argc:
    (void)argc;

    for (char **p = argv; *p != NULL; p++) {
        int value = atoi(*p);  // Use strtol() for better error handling

        total += value;
    }

    printf("%d\n", total);
}

Personally, I use array notation to access argv, but have seen this style floating around, as well.

18.1.3 Fun Facts

Just a few more things about argc and argv.

Some environments might not set argv[0] to the program name. If it’s not available, argv[0] will be an empty string. I’ve never seen this happen.
The spec is actually pretty liberal with what an implementation can do with argv and where those values come from. But every system I’ve been on works the same way, as we’ve discussed in this section.
You can modify argc, argv, or any of the strings that argv points to. (Just don’t make those strings longer than they already are!)
On some Unix-like systems, modifying the string argv[0] results in the output of ps changing⁹⁹.

Normally, if you have a program called foo that you’ve run with ./foo, you might see this in the output of ps:
```
     4078 tty1     S      0:00 ./foo
```
But if you modify argv[0] like so, being careful that the new string "Hi! " is the same length as the old one "./foo":
```
    strcpy(argv[0], "Hi!  ");
```
and then run ps while the program ./foo is still executing, we’ll see this instead:
```
     4079 tty1     S      0:00 Hi!  
```
This behavior is not in the spec and is highly system-dependent.

18.2 Exit Status

Did you notice that the function signatures for main() have it returning type int? What’s that all about? It has to do with a thing called the exit status, which is an integer that can be returned to the program that launched yours to let it know how things went.

Now, there are a number of ways a program can exit in C, including returning from main(), or calling one of the exit() variants.

All of these methods accept an int as an argument.

Side note: did you see that in basically all my examples, even though main() is supposed to return an int, I don’t actually return anything? In any other function, this would be illegal, but there’s a special case in C: if execution reaches the end of main() without finding a return, it automatically does a return 0.

But what does the 0 mean? What other numbers can we put there? And how are they used?

The spec is both clear and vague on the matter, as is common. Clear because it spells out what you can do, but vague in that it doesn’t particularly limit it, either.

Nothing for it but to forge ahead and figure it out!

Let’s get Inception ¹⁰⁰ for a second: turns out that when you run your program, you’re running it from another program.

Usually this other program is some kind of shell ¹⁰¹ that doesn’t do much on its own except launch other programs.

But this is a multi-phase process, especially visible in command-line shells:

The shell launches your program
The shell typically goes to sleep (for command-line shells)
Your program runs
Your program terminates
The shell wakes up and waits for another command

Now, there’s a little piece of communication that takes place between steps 4 and 5: the program can return a status value that the shell can interrogate. Typically, this value is used to indicate the success or failure of your program, and, if a failure, what type of failure.

This value is what we’ve been returning from main(). That’s the status.

Now, the C spec allows for two different status values, which have macro names defined in <stdlib.h>:

Status	Description
`EXIT_SUCCESS` or `0`	Program terminated successfully.
`EXIT_FAILURE`	Program terminated with an error.

Let’s write a short program that multiplies two numbers from the command line. We’ll require that you specify exactly two values. If you don’t, we’ll print an error message, and exit with an error status.

#include <stdio.h>
#include <stdlib.h>

int main(int argc, char **argv)
{
    if (argc != 3) {
        printf("usage: mult x y\n");
        return EXIT_FAILURE;   // Indicate to shell that it didn't work
    }

    printf("%d\n", atoi(argv[1]) * atoi(argv[2]));

    return 0;  // same as EXIT_SUCCESS, everything was good.
}

Now if we try to run this, we get the expected effect until we specify exactly the right number of command-line arguments:

    $ ./mult
    usage: mult x y
    
    $ ./mult 3 4 5
    usage: mult x y
    
    $ ./mult 3 4
    12

But that doesn’t really show the exit status that we returned, does it? We can get the shell to print it out, though. Assuming you’re running Bash or another POSIX shell, you can use echo $? to see it¹⁰².

Let’s try:

    $ ./mult
    usage: mult x y
    $ echo $?
    1
    
    $ ./mult 3 4 5
    usage: mult x y
    $ echo $?
    1
    
    $ ./mult 3 4
    12
    $ echo $?
    0

Interesting! We see that on my system, EXIT_FAILURE is 1. The spec doesn’t spell this out, so it could be any number. But try it; it’s probably 1 on your system, too.

18.2.1 Other Exit Status Values

The status 0 most definitely means success, but what about all the other integers, even negative ones?

Here we’re going off the C spec and into Unix land. In general, while 0 means success, a positive non-zero number means failure. So you can only have one type of success, and multiple types of failure. Bash says the exit code should be between 0 and 255, though a number of codes are reserved.

In short, if you want to indicate different error exit statuses in a Unix environment, you can start with 1 and work your way up.

On Linux, if you try any code outside the range 0-255, it will bitwise AND the code with 0xff, effectively clamping it to that range.

You can script the shell to later use these status codes to make decisions about what to do next.

18.3 Environment Variables

Before I get into this, I need to warn you that C doesn’t specify what an environment variable is. So I’m going to describe the environment variable system that works on every major platform I’m aware of.

Basically, the environment is the program that’s going to run your program, e.g. the bash shell. And it might have some bash variables defined. In case you didn’t know, the shell can make its own variables. Each shell is different, but in bash you can just type set and it’ll show you all of them.

Here’s an excerpt from the 61 variables that are defined in my bash shell:

    HISTFILE=/home/beej/.bash_history
    HISTFILESIZE=500
    HISTSIZE=500
    HOME=/home/beej
    HOSTNAME=FBILAPTOP
    HOSTTYPE=x86_64
    IFS=$' \t\n'

Notice they are in the form of key/value pairs. For example, one key is HOSTTYPE and its value is x86_64. From a C perspective, all values are strings, even if they’re numbers¹⁰³.

So, anyway! Long story short, it’s possible to get these values from inside your C program.

Let’s write a program that uses the standard getenv() function to look up a value that you set in the shell.

getenv() will return a pointer to the value string, or else NULL if the environment variable doesn’t exist.

#include <stdio.h>
#include <stdlib.h>

int main(void)
{
    char *val = getenv("FROTZ");  // Try to get the value

    // Check to make sure it exists
    if (val == NULL) {
        printf("Cannot find the FROTZ environment variable\n");
        return EXIT_FAILURE;
    }

    printf("Value: %s\n", val);
}

If I run this directly, I get this:

    $ ./foo
    Cannot find the FROTZ environment variable

which makes since, since I haven’t set it yet.

In bash, I can set it to something with¹⁰⁴:

    $ export FROTZ="C is awesome!"

Then if I run it, I get:

    $ ./foo
    Value: C is awesome!

In this way, you can set up data in environment variables, and you can get it in your C code and modify your behavior accordingly.

18.3.1 Setting Environment Variables

This isn’t standard, but a lot of systems provide ways to set environment variables.

If on a Unix-like, look up the documentation for putenv(), setenv(), and unsetenv(). On Windows, see _putenv().

18.3.2 Unix-like Alternative Environment Variables

If you’re on a Unix-like system, odds are you have another couple ways of getting access to environment variables. Note that although the spec points this out as a common extension, it’s not truly part of the C standard. It is, however, part of the POSIX standard.

One of these is a variable called environ that must be declared like so:

    extern char **environ;

It’s an array of strings terminated with a NULL pointer.

You should declare it yourself before you use it, or you might find it in the non-standard <unistd.h> header file.

Each string is in the form "key=value" so you’ll have to split it and parse it yourself if you want to get the keys and values out.

Here’s an example of looping through and printing out the environment variables a couple different ways:

#include <stdio.h>

extern char **environ;  // MUST be extern AND named "environ"

int main(void)
{
    for (char **p = environ; *p != NULL; p++) {
        printf("%s\n", *p);
    }

    // Or you could do this:
    for (int i = 0; environ[i] != NULL; i++) {
        printf("%s\n", environ[i]);
    }
}

For a bunch of output that looks like this:

    SHELL=/bin/bash
    COLORTERM=truecolor
    TERM_PROGRAM_VERSION=1.53.2
    LOGNAME=beej
    VSCODE_GIT_ASKPASS_NODE=/home/beej/.vscode-server/bin/ea3859d4ba2f3e577a159bc91e3074c5d85c0523/node
    HOME=/home/beej
    ... etc ...

Use getenv() if at all possible because it’s more portable. But if you have to iterate over environment variables, using environ might be the way to go.

Another non-standard way to get the environment variables is as a parameter to main(). It works much the same way, but you avoid needing to add your extern environ variable. Not even the POSIX spec supports this ¹⁰⁵ as far as I can tell, but it’s common in Unix land.

#include <stdio.h>

int main(int argc, char **argv, char **env)  // <-- env!
{
    (void)argc; (void)argv;  // Suppress unused warnings

    for (char **p = env; *p != NULL; p++) {
        printf("%s\n", *p);
    }

    // Or you could do this:
    for (int i = 0; env[i] != NULL; i++) {
        printf("%s\n", env[i]);
    }
}

Just like using environ but even less portable. It’s good to have goals.

19 The C Preprocessor

Before your program gets compiled, it actually runs through a phase called preprocessing. It’s almost like there’s a language on top of the C language that runs first. And it outputs the C code, which then gets compiled.

We’ve already seen this to an extent with #include! That’s the C Preprocessor! Where it sees that directive, it includes the named file right there, just as if you’d typed it in there. And then the compiler builds the whole thing.

But it turns out it’s a lot more powerful than just being able to include things. You can define macros that are substituted… and even macros that take arguments!

19.1 `#include`

Let’s start with the one we’ve already seen a bunch. This is, of course, a way to include other sources in your source. Very commonly used with header files.

While the spec allows for all kinds of behavior with #include, we’re going to take a more pragmatic approach and talk about the way it works on every system I’ve ever seen.

We can split header files into two categories: system and local. Things that are built-in, like stdio.h, stdlib.h, math.h, and so on, you can include with angle brackets:

    #include <stdio.h>
    #include <stdlib.h>

The angle brackets tell C, “Hey, don’t look in the current directory for this header file—look in the system-wide include directory instead.”

Which, of course, implies that there must be a way to include local files from the current directory. And there is: with double quotes:

    #include "myheader.h"

Or you can very probably look in relative directories using forward slashes and dots, like this:

    #include "mydir/myheader.h"
    #include "../someheader.py"

Don’t use a backslash (\) for your path separators in your #include! It’s undefined behavior! Use forward slash (/) only, even on Windows.

In summary, used angle brackets (< and >) for the system includes, and use double quotes (") for your personal includes.

19.2 Simple Macros

A macro is an identifier that gets expanded to another piece of code before the compiler even sees it. Think of it like a placeholder—when the preprocessor sees one of those identifiers, it replaces it with another value that you’ve defined.

We do this with #define (often read “pound define”). Here’s an example:

#include <stdio.h>

#define HELLO "Hello, world"
#define PI 3.14159

int main(void)
{
    printf("%s, %f\n", HELLO, PI);
}

On lines 3 and 4 we defined a couple macros. Wherever these appear elsewhere in the code (line 8), they’ll be substituted with the defined values.

From the C compiler’s perspective, it’s exactly as if we’d written this, instead:

#include <stdio.h>

int main(void)
{
    printf("%s, %f\n", "Hello, world", 3.14159);
}

See how HELLO was replaced with "Hello, world" and PI was replaced with 3.14159? From the compiler’s perspective, it’s just like those values had appeared right there in the code.

Note that the macros don’t have a specific type, per se. Really all that happens is they get replaced wholesale with whatever they’re #defined as. If the resulting C code is invalid, the compiler will puke.

You can also define a macro with no value:

    #define EXTRA_HAPPY

in that case, the macro exists and is defined, but is defined to be nothing. So anyplace it occurs in the text will just be replaced with nothing. We’ll see a use for this later.

It’s conventional to write macro names in ALL_CAPS even though that’s not technically required.

Overall, this gives you a way to define constant values that are effectively global and can be used any place. Even in those places where a const variable won’t work, e.g. in switch cases and fixed array lengths.

That said, the debate rages online whether a typed const variable is better than #define macro in the general case.

It can also be used to replace or modify keywords, a concept completely foreign to const, though this practice should be used sparingly.

19.3 Conditional Compilation

It’s possible to get the preprocessor to decide whether or not to present certain blocks of code to the compiler, or just remove them entirely before compilation.

We do that by basically wrapping up the code in conditional blocks, similar to if-else statements.

19.3.1 If Defined, `#ifdef` and `#endif`

First of all, let’s try to compile specific code depending on whether or not a macro is even defined.

#include <stdio.h>

#define EXTRA_HAPPY

int main(void)
{

#ifdef EXTRA_HAPPY
    printf("I'm extra happy!\n");
#endif

    printf("OK!\n");
}

In that example, we define EXTRA_HAPPY (to be nothing, but it is defined), then on line 8 we check to see if it is defined with an #ifdef directive. If it is defined, the subsequent code will be included up until the #endif.

So because it is defined, the code will be included for compilation and the output will be:

    I'm extra happy!
    OK!

If we were to comment out the #define, like so:

    //#define EXTRA_HAPPY

then it wouldn’t be defined, and the code wouldn’t be included in compilation. And the output would just be:

OK!

It’s important to remember that these decisions happen at compile time! The code actually get compiled or removed depending on the condition. This is in contrast to a standard if statement that gets evaluated while the program is running.

19.3.2 If Not Defined, `#ifndef`

There’s also the negative sense of “if defined”: “if not defined”, or #ifndef. We could change the previous example to read to output different things based on whether or not something was defined:

#ifdef EXTRA_HAPPY
    printf("I'm extra happy!\n");
#endif

#ifndef EXTRA_HAPPY
    printf("I'm just regular\n");
#endif

We’ll see a cleaner way to do that in the next section.

Tying it all back in to header files, we’ve seen how we can cause header files to only be included one time by wrapping them in preprocessor directives like this:

    #ifndef MYHEADER_H  // First line of myheader.h
    #define MYHEADER_H
    
    int x = 12;
    
    #endif  // Last line of myheader.h

This demonstrates how a macro persists across files and multiple #includes. If it’s not yet defined, let’s define it and compile the whole header file.

But the next time it’s included, we see that MYHEADER_H is defined, so we don’t send the header file to the compiler—it gets effectively removed.

19.3.3 `#else`

But that’s not all we can do! There’s also an #else that we can throw in the mix.

Let’s mod the previous example:

#ifdef EXTRA_HAPPY
    printf("I'm extra happy!\n");
#else
    printf("I'm just regular\n");
#endif

Now if EXTRA_HAPPY is not defined, it’ll hit the #else clause and print:

    I'm just regular

19.3.4 General Conditional: `#if`, `#elif`

This works very much like the #ifdef and #ifndef directives in that you can also have an #else and the whole thing wraps up with #endif.

The only difference is that the constant expression after the #if must evaluate to true (non-zero) for the code in the #if to be compiled. So instead of whether or not something is defined, we want an expression that evaluates to true.

#include <stdio.h>

#define HAPPY_FACTOR 1

int main(void)
{

#if HAPPY_FACTOR == 0
    printf("I'm not happy!\n");
#elif HAPPY_FACTOR == 1
    printf("I'm just regular\n");
#else
    printf("I'm extra happy!\n");
#endif

    printf("OK!\n");
}

Again, for the unmatched #if clauses, the compiler won’t even see those lines. For the above code, after the preprocessor gets finished with it, all the compiler sees is:

#include <stdio.h>

int main(void)
{

    printf("I'm just regular\n");

    printf("OK!\n");
}

One hackish thing this is used for is to comment out large numbers of lines quickly¹⁰⁶.

If you put an #if 0 (“if false”) at the front of the block to be commented out and an #endif at the end, you can get this effect:

    #if 0
        printf("All this code"); /* is effectively */
        printf("commented out"); // by the #if 0
    #endif

You might have noticed that there’s no #elifdef or #elifndef directives. How can we get the same effect with #if? That is, what if I wanted this:

    #ifdef FOO
        x = 2;
    #elifdef BAR  // ERROR: Not supported by standard C
        x = 3;
    #endif

How could I do it?

Turns out there’s a preprocessor operator called defined that we can use with an #if statement.

These are equivalent:

    #ifdef FOO
    #if defined FOO
    #if defined(FOO)   // Parentheses optional

As are these:

    #ifndef FOO
    #if !defined FOO
    #if !defined(FOO)   // Parentheses optional

Notice how we can use the standard logical NOT operator (!) for “not defined”.

So now we’re back in #if land and we can use #elif with impunity!

This broken code:

    #ifdef FOO
        x = 2;
    #elifdef BAR  // ERROR: Not supported by standard C
        x = 3;
    #endif

can be replaced with:

    #if defined FOO
        x = 2;
    #elif defined BAR
        x = 3;
    #endif

19.3.5 Losing a Macro: `#undef`

If you’ve defined something but you don’t need it any longer, you can undefine it with #undef.

#include <stdio.h>

int main(void)
{
#define GOATS

#ifdef GOATS
    printf("Goats detected!\n");  // prints
#endif

#undef GOATS  // Make GOATS no longer defined

#ifdef GOATS
    printf("Goats detected, again!\n"); // doesn't print
#endif
}

19.4 Built-in Macros

The standard defines a lot of built-in macros that you can test and use for conditional compilation. Let’s look at those here.

19.4.1 Mandatory Macros

These are all defined:

Macro	Description
`__DATE__`	The date of compilation—like when you’re compiling this file—in `Mmm dd yyyy` format
`__TIME__`	The time of compilation in `hh:mm:ss` format
`__FILE__`	A string containing this file’s name
`__LINE__`	The line number of the file this macro appears on
`__func__`	The name of the function this appears in, as a string¹⁰⁷
`__STDC__`	Defined with `1` if this is a standard C compiler
`__STDC_HOSTED__`	This will be `1` if the compiler is a hosted implementation¹⁰⁸, otherwise `0`
`__STDC_VERSION__`	This version of C, a constant `long int` in the form `yyyymmL`, e.g. `201710L`

Let’s put these together.

#include <stdio.h>

int main(void)
{
    printf("This function: %s\n", __func__);
    printf("This file: %s\n", __FILE__);
    printf("This line: %d\n", __LINE__);
    printf("Compiled on: %s %s\n", __DATE__, __TIME__);
    printf("C Version: %ld\n", __STDC_VERSION__);
}

The output on my system is:

    This function: main
    This file: foo.c
    This line: 7
    Compiled on: Nov 23 2020 17:16:27
    C Version: 201710

__FILE__, __func__ and __LINE__ are particularly useful to report error conditions in messages to developers. The assert() macro in <assert.h> uses these to call out where in the code the assertion failed.

19.4.2 Optional Macros

Your implementation might define these, as well. Or it might not.

Macro	Description
`__STDC_ISO_10646__`	If defined, `wchar_t` holds Unicode values, otherwise something else
`__STDC_MB_MIGHT_NEQ_WC__`	A `1` indicates that the values in multibyte characters might not map equally to values in wide characters
`__STDC_UTF_16__`	A `1` indicates that the system uses UTF-16 encoding in type `char16_t`
`__STDC_UTF_32__`	A `1` indicates that the system uses UTF-32 encoding in type `char32_t`
`__STDC_ANALYZABLE__`	A `1` indicates the code is analyzable¹⁰⁹
`__STDC_IEC_559__`	`1` if IEEE-754 (aka IEC 60559) floating point is supported
`__STDC_IEC_559_COMPLEX__`	`1` if IEC 60559 complex floating point is supported
`__STDC_LIB_EXT1__`	`1` if this implementation supports a variety of “safe” alternate standard library functions (they have `_s` suffixes on the name)
`__STDC_NO_ATOMICS__`	`1` if this implementation does not support `_Atomic` or `<stdatomic.h>`
`__STDC_NO_COMPLEX__`	`1` if this implementation does not support complex types or `<complex.h>`
`__STDC_NO_THREADS__`	`1` if this implementation does not support `<threads.h>`
`__STDC_NO_VLA__`	`1` if this implementation does not support variable-length arrays

19.5 Macros with Arguments

Macros are more powerful than simple substitution, though. You can set them up to take arguments that are substituted in, as well.

A question often arises for when to use parameterized macros versus functions. Short answer: use functions. But you’ll see lots of macros in the wild and in the standard library. People tend to use them for short, mathy things, and also for features that might change from platform to platform. You can define different keywords for one platform or another.

19.5.1 Macros with One Argument

Let’s start with a simple one that squares a number:

#include <stdio.h>

#define SQR(x) x * x  // Not quite right, but bear with me

int main(void)
{
    printf("%d\n", SQR(12));  // 144
}

What that’s saying is “everywhere you see SQR with some value, replace it with that value times itself”.

So line 7 will be changed to:

    printf("%d\n", 12 * 12);  // 144

which C comfortably converts to 144.

But we’ve made an elementary error in that macro, one that we need to avoid.

Let’s check it out. What if we wanted to compute SQR(3 + 4)? Well, $3+4=7$, so we must want to compute $7^2=49$. That’s it; 49—final answer.

Let’s drop it in our code and see that we get… 19?

    printf("%d\n", SQR(3 + 4));  // 19!!??

What happened?

If we follow the macro expansion, we get

    printf("%d\n", 3 + 4 * 3 + 4);  // 19!

Oops! Since multiplication takes precedence, we do the $4\times3=12$ first, and get $3+12+4=19$. Not what we were after.

So we have to fix this to make it right.

This is so common that you should automatically do it every time you make a parameterized math macro!

The fix is easy: just add some parentheses!

#define SQR(x) (x) * (x)   // Better... but still not quite good enough!

And now our macro expands to:

    printf("%d\n", (3 + 4) * (3 + 4));  // 49! Woo hoo!

But we actually still have the same problem which might manifest if we have a higher-precedence operator than multiply (*) nearby.

So the safe, proper way to put the macro together is to wrap the whole thing in additional parentheses, like so:

#define SQR(x) ((x) * (x))   // Good!

Just make it a habit to do that when you make a math macro and you can’t go wrong.

19.5.2 Macros with More than One Argument

You can stack these things up as much as you want:

    #define TRIANGLE_AREA(w, h) (0.5 * (w) * (h))

Let’s do some macros that solve for $x$ using the quadratic formula. Just in case you don’t have it on the top of your head, it says for equations of the form:

$ax^2+bx+c=0$

you can solve for $x$ with the quadratic formula:

$x=\cfrac{-b\pm\sqrt{b^2-4ac}}{2a}$

Which is crazy. Also notice the plus-or-minus ($\pm$) in there, indicating that there are actually two solutions.

So let’s make macros for both:

    #define QUADP(a, b, c) ((-(b) + sqrt((b) * (b) - 4 * (a) * (c))) / (2 * (a)))
    #define QUADM(a, b, c) ((-(b) - sqrt((b) * (b) - 4 * (a) * (c))) / (2 * (a)))

So that gets us some math. But let’s define one more that we can use as arguments to printf() to print both answers.

    //          macro              replacement
    //      |-----------| |----------------------------|
    #define QUAD(a, b, c) QUADP(a, b, c), QUADM(a, b, c)

That’s just a couple values separated by a comma—and we can use that as a “combined” argument of sorts to printf() like this:

    printf("x = %f or x = %f\n", QUAD(2, 10, 5));

Let’s put it together into some code:

#include <stdio.h>
#include <math.h>  // For sqrt()

#define QUADP(a, b, c) ((-(b) + sqrt((b) * (b) - 4 * (a) * (c))) / (2 * (a)))
#define QUADM(a, b, c) ((-(b) - sqrt((b) * (b) - 4 * (a) * (c))) / (2 * (a)))
#define QUAD(a, b, c) QUADP(a, b, c), QUADM(a, b, c)

int main(void)
{
    printf("2*x^2 + 10*x + 5 = 0\n");
    printf("x = %f or x = %f\n", QUAD(2, 10, 5));
}

And this gives us the output:

    2*x^2 + 10*x + 5 = 0
    x = -0.563508 or x = -4.436492

Plugging in either of those values gives us roughly zero (a bit off because the numbers aren’t exact):

$2\times-0.563508^2+10\times-0.563508+5\approx0.000003$

19.5.3 Macros with Variable Arguments

There’s also a way to have a variable number of arguments passed to a macro, using ellipses (...) after the known, named arguments. When the macro is expanded, all of the extra arguments will be in a comma-separated list in the __VA_ARGS__ macro, and can be replaced from there:

#include <stdio.h>

// Combine the first two arguments to a single number,
// then have a commalist of the rest of them:

#define X(a, b, ...) (10*(a) + 20*(b)), __VA_ARGS__

int main(void)
{
    printf("%d %f %s %d\n", X(5, 4, 3.14, "Hi!", 12));
}

The substitution that takes place on line 10 would be:

    printf("%d %f %s %d\n", (10*(5) + 20*(4)), 3.14, "Hi!", 12);

for output:

    130 3.140000 Hi! 12

You can also “stringify” __VA_ARGS__ by putting a # in front of it:

    #define X(...) #__VA_ARGS__
    
    printf("%s\n", X(1,2,3));  // Prints "1, 2, 3"

19.5.4 Stringification

Already mentioned, just above, you can turn any argument into a string by preceding it with a # in the replacement text.

For example, we could print anything as a string with this macro and printf():

    #define STR(x) #x
    
    printf("%s\n", STR(3.14159));

In that case, the substitution leads to:

    printf("%s\n", "3.14159");

Let’s see if we can use this to greater effect so that we can pass any int variable name into a macro, and have it print out it’s name and value.

#include <stdio.h>

#define PRINT_INT_VAL(x) printf("%s = %d\n", #x, x)

int main(void)
{
    int a = 5;

    PRINT_INT_VAL(a);  // prints "a = 5"
}

On line 9, we get the following macro replacement:

    printf("%s = %d\n", "a", 5);

19.5.5 Concatenation

We can concatenate two arguments together with ##, as well. Fun times!

    #define CAT(a, b) a ## b
    
    printf("%f\n", CAT(3.14, 1592));   // 3.141592

19.6 Multiline Macros

It’s possible to continue a macro to multiple lines if you escape the newline with a backslash (\).

Let’s write a multiline macro that prints numbers from 0 to the product of the two arguments passed in.

#include <stdio.h>

#define PRINT_NUMS_TO_PRODUCT(a, b) { \
    int product = (a) * (b); \
    for (int i = 0; i < product; i++) { \
        printf("%d\n", i); \
    } \
}

int main(void)
{
    PRINT_NUMS_TO_PRODUCT(2, 4);  // Outputs numbers from 0 to 7
}

A couple things to note there:

Escapes at the end of every line except the last one to indicate that the macro continues.
Though not strictly necessary, I wrapped the whole thing in curly braces. This did two things:
1. Made it look nice.
2. Made a new block scope for my product variable so it wouldn’t conflict with any other existing variables at the outer block scope.

19.7 Example: An Assert Macro

Adding asserts to your code is a good way to catch conditions that you think shouldn’t happen. C provides assert() functionality. It checks a condition, and if it’s false, the program bombs out telling you the file and line number on which the assertion failed.

But this is wanting.

First of all, you can’t specify an additional message with the assert.
Secondly, there’s no easy on-off switch for all the asserts.

We can address the first with macros.

Basically, when I have this code:

    ASSERT(x < 20, "x must be under 20");

I want something like this to happen (assuming the ASSERT() is on line 220 of foo.c):

    if (x < 20) {
        fprintf(stderr, "foo.c:220: assertion x < 20 failed: ");
        fprintf(stderr, "x must be under 20\n");
        exit(1)
    }

We can get the filename out of the __FILE__ macro, and the line number from __LINE__. The message is already a string, but x < 20 is not, so we’ll have to stringify it with #. We can make a multiline macro by using backslash escapes at the end of the line.

    #define ASSERT(c, m) \
    { \
        if (!(c)) { \
            fprintf(stderr, __FILE__ ":%d: assertion %s failed: %s\n", \
                            __LINE__, #c, m); \
            exit(1); \
        } \
    }

(It looks a little weird with __FILE__ out front like that, but remember it is a string literal, and string literals next to each other are automagically concatenated. __LINE__ on the other hand, it’s just an int.)

And that works! If I run this:

    int x = 30;
    
    ASSERT(x < 20, "x must be under 20");

I get this output:

    foo.c:23: assertion x < 20 failed: x must be under 20

Very nice!

The only thing left is a way to turn it on and off, and we could do that with conditional compilation.

Here’s the complete example:

#include <stdio.h>
#include <stdlib.h>

#define ASSERT_ENABLED 1

#if ASSERT_ENABLED
#define ASSERT(c, m) \
{ \
    if (!(c)) { \
        fprintf(stderr, __FILE__ ":%d: assertion %s failed: %s\n", \
                        __LINE__, #c, m); \
        exit(1); \
    } \
}
#else
#define ASSERT(c, m)  // Empty macro if not enabled
#endif

int main(void)
{
    int x = 30;

    ASSERT(x < 20, "x must be under 20");
}

This has the output:

    foo.c:23: assertion x < 20 failed: x must be under 20

19.8 The `#error` Directive

This directive causes the compiler to error out as soon as it sees it.

Commonly, this is used inside a conditional to prevent compilation unless some prerequisites are met:

    #ifndef __STDC_IEC_559__
        #error I really need IEEE-754 floating point to compile. Sorry!
    #endif

Some compilers have a non-standard complementary #warning directive that will output a warning but not stop compilation, but this is not in the C11 spec.

19.9 The `#pragma` Directive

This is one funky directive, short for “pragmatic”. You can use it to do… well, anything your compiler supports you doing with it.

Basically the only time you’re going to add this to your code is if some documentation tells you to do so.

19.9.1 Non-Standard Pragmas

Here’s one non-standard example of using #pragma to cause the compiler to execute a for loop in parallel with multiple threads (if the compiler supports the OpenMP ¹¹⁰ extension):

    #pragma omp parallel for
    for (int i = 0; i < 10; i++) { ... }

There are all kinds of #pragma directives documented across all four corners of the globe.

All unrecognized #pragmas are ignored by the compiler.

19.9.2 Standard Pragmas

There are also a few standard ones, and these start with STDC, and follow the same form:

    #pragma STDC pragma_name on-off

The on-off portion can be either ON, OFF, or DEFAULT.

And the pragma_name can be one of these:

Pragma Name	Description
`FP_CONTRACT`	Allow floating point expressions to be contracted into a single operation to avoid rounding errors that might occur from multiple operations.
`FENV_ACCESS`	Set to `ON` if you plan to access the floating point status flags. If `OFF`, the compiler might perform optimizations that cause the values in the flags to be inconsistent or invalid.
`CX_LIMITED_RANGE`	Set to `ON` to allow the compiler to skip overflow checks when performing complex arithmetic. Defaults to `OFF`.

For example:

    #pragma STDC FP_CONTRACT OFF
    #pragma STDC CX_LIMITED_RANGE ON

As for CX_LIMITED_RANGE, the spec points out:

The purpose of the pragma is to allow the implementation to use the formulas:

$(x+iy)\times(u+iv) = (xu-yv)+i(yu+xv)$

$(x+iy)/(u+iv) = [(xu+yv)+i(yu-xv)]/(u^2+v^2)$

$|x+iy|=\sqrt{x^2+y^2}$

where the programmer can determine they are safe.

19.9.3 `_Pragma` Operator

This is another way to declare a pragma that you could use in a macro.

These are equivalent:

    #pragma "Unnecessary" quotes
    _Pragma("\"Unnecessary\" quotes")

This can be used in a macro, if need be:

    #define PRAGMA(x) _Pragma(#x)

19.10 The `#line` Directive

This allows you to override the values for __LINE__ and __FILE__. If you want.

I’ve never wanted to do this, but in K&R2, they write:

For the benefit of other preprocessors that generate C programs […]

So maybe there’s that.

To override the line number to, say 300:

    #line 300

and __LINE__ will keep counting up from there.

To override the line number and the filename:

    #line 300 "newfilename"

19.11 The Null Directive

A # on a line by itself is ignored by the preprocessor. Now, to be entirely honest, I don’t know what the use case is for this.

I’ve seen examples like this:

    #ifdef FOO
        #
    #else
        printf("Something");
    #endif

which is just cosmetic; the line with the solitary # can be deleted with no ill effect.

Or maybe for cosmetic consistency, like this:

    #
    #ifdef FOO
        x = 2;
    #endif
    #
    #if BAR == 17
        x = 12;
    #endif
    #

But, with respect to cosmetics, that’s just ugly.

Another post mentions elimination of comments—that in GCC, a comment after a # will not be seen by the compiler. Which I don’t doubt, but the specification doesn’t seem to say this is standard behavior.

My searches for rationale aren’t bearing much fruit. So I’m going to just say this is some good ol’ fashioned C esoterica.

20 `struct`s II: More Fun with `struct`s

Turns out there’s a lot more you can do with structs than we’ve talked about, but it’s just a big pile of miscellaneous things. So we’ll throw them in this chapter.

If you’re good with struct basics, you can round out your knowledge here.

20.1 Anonymous `struct`s

These are “the struct with no name”. We also mention these in the typedef section, but we’ll refresh here.

Here’s a regular struct:

    struct animal {
        char *name;
        int leg_count, speed;
    };

And here’s the anonymous equivalent:

    struct {              // <-- No name!
        char *name;
        int leg_count, speed;
    };

Okaaaaay. So we have a struct, but it has no name, so we have no way of using it later? Seems pretty pointless.

Admittedly, in that example, it is. But we can still make use of it a couple ways.

One is rare, but since the anonymous struct represents a type, we can just put some variable names after it and use them.

    struct {                   // <-- No name!
        char *name;
        int leg_count, speed;
    } a, b, c;                 // 3 variables of this struct type
    
    a.name = "antelope";
    c.leg_count = 4;           // for example

But that’s still not that useful.

Far more common is use of anonymous structs with a typedef so that we can use it later (e.g. to pass variables to functions).

    typedef struct {                   // <-- No name!
        char *name;
        int leg_count, speed;
    } animal;                          // New type: animal
    
    animal a, b, c;
    
    a.name = "antelope";
    c.leg_count = 4;           // for example

Personally, I don’t use many anonymous structs. I think it’s more pleasant to see the entire struct animal before the variable name in a declaration.

But that’s just, like, my opinion, man.

20.2 Self-Referential `struct`s

For any graph-like data structure, it’s useful to be able to have pointers to the connected nodes/vertices. But this means that in the definition of a node, you need to have a pointer to a node. It’s chicken and eggy!

But it turns out you can do this in C with no problem whatsoever.

For example, here’s a linked list node:

    struct node {
        int data;
        struct node *next;
    };

It’s important to note that next is a pointer. This is what allows the whole thing to even build. Even though the compiler doesn’t know what the entire struct node looks like yet, all pointers are the same size.

Here’s a cheesy linked list program to test it out:

#include <stdio.h>
#include <stdlib.h>

struct node {
    int data;
    struct node *next;
};

int main(void)
{
    struct node *head;

    // Hackishly set up a linked list (11)->(22)->(33)
    head = malloc(sizeof(struct node));
    head->data = 11;
    head->next = malloc(sizeof(struct node));
    head->next->data = 22;
    head->next->next = malloc(sizeof(struct node));
    head->next->next->data = 33;
    head->next->next->next = NULL;

    // Traverse it
    for (struct node *cur = head; cur != NULL; cur = cur->next) {
        printf("%d\n", cur->data);
    }
}

Running that prints:

    11
    22
    33

20.3 Flexible Array Members

Back in the good old days, when people carved C code out of wood, some folks thought would be neat if they could allocate structs that had variable length arrays at the end of them.

I want to be clear that the first part of the section is the old way of doing things, and we’re going to do things the new way after that.

For example, maybe you could define a struct for holding strings and the length of that string. It would have a length and an array to hold the data. Maybe something like this:

    struct len_string {
        int length;
        char data[8];
    };

But that has 8 hardcoded as the maximum length of a string, and that’s not much. What if we did something clever and just malloc()d some extra space at the end after the struct, and then let the data overflow into that space?

Let’s do that, and then allocate another 40 bytes on top of it:

    struct len_string *s = malloc(sizeof *s + 40);

Because data is the last field of the struct, if we overflow that field, it runs out into space that we already allocated! For this reason, this trick only works if the short array is the last field in the struct.

    // Copy more than 8 bytes!
    
    strcpy(s->data, "Hello, world!");  // Won't crash. Probably.

In fact, there was a common compiler workaround for doing this, where you’d allocate a zero length array at the end:

    struct len_string {
        int length;
        char data[0];
    };

And then every extra byte you allocated was ready for use in that string.

Because data is the last field of the struct, if we overflow that field, it runs out into space that we already allocated!

    // Copy more than 8 bytes!
    
    strcpy(s->data, "Hello, world!");  // Won't crash. Probably.

But, of course, actually accessing the data beyond the end of that array is undefined behavior! In these modern times, we no longer deign to resort to such savagery.

Luckily for us, we can still get the same effect with C99 and later, but now it’s legal.

Let’s just change our above definition to have no size for the array¹¹¹:

    struct len_string {
        int length;
        char data[];
    };

Again, this only works if the flexible array member is the last field in the struct.

And then we can allocate all the space we want for those strings by malloc()ing larger than the struct len_string, as we do in this example that makes a new struct len_string from a C string:

    struct len_string *len_string_from_c_string(char *s)
    {
        int len = strlen(s);
    
        // Allocate "len" more bytes than we'd normally need
        struct len_string *ls = malloc(sizeof *ls + len);
    
        ls->length = len;
    
        // Copy the string into those extra bytes
        memcpy(ls->data, s, len);
    
        return ls;
    }

20.4 Padding Bytes

Beware that C is allowed to add padding bytes within or after a struct as it sees fit. You can’t trust that they will be directly adjacent in memory¹¹².

Let’s take a look at this program. We output two numbers. One is the sum of the sizeofs the individual field types. The other is the sizeof the entire struct.

One would expect them to be the same. The size of the total is the size of the sum of its parts, right?

#include <stdio.h>

struct foo {
    int a;
    char b;
    int c;
    char d;
};

int main(void)
{
    printf("%zu\n", sizeof(int) + sizeof(char) + sizeof(int) + sizeof(char));
    printf("%zu\n", sizeof(struct foo));
}

But on my system, this outputs:

    10
    16

They’re not the same! The compiler has added 6 bytes of padding to help it be more performant. Maybe you got different output with your compiler, but unless you’re forcing it, you can’t be sure there’s no padding.

20.5 `offsetof`

In the previous section, we saw that the compiler could inject padding bytes at will inside a structure.

What if we needed to know where those were? We can measure it with offsetof, defined in <stddef.h>.

Let’s modify the code from above to print the offsets of the individual fields in the struct:

#include <stdio.h>
#include <stddef.h>

struct foo {
    int a;
    char b;
    int c;
    char d;
};

int main(void)
{
    printf("%zu\n", offsetof(struct foo, a));
    printf("%zu\n", offsetof(struct foo, b));
    printf("%zu\n", offsetof(struct foo, c));
    printf("%zu\n", offsetof(struct foo, d));
}

For me, this outputs:

indicating that we’re using 4 bytes for each of the fields. It’s a little weird, because char is only 1 byte, right? The compiler is putting 3 padding bytes after each char so that all the fields are 4 bytes long. Presumably this will run faster on my CPU.

20.6 Fake OOP

There’s a slightly abusive thing that’s sort of OOP-like that you can do with structs.

Since the pointer to the struct is the same as a pointer to the first element of the struct, you can freely cast a pointer to the struct to a pointer to the first element.

What this means is that we can set up a situation like this:

    struct parent {
        int a, b;
    };
    
    struct child {
        struct parent super;  // MUST be first
        int c, d;
    };

Then we are able to pass a pointer to a struct child to a function that expects either that or a pointer to a struct parent!

Because struct parent super is the first item in the struct child, a pointer to any struct child is the same as a pointer to that super field¹¹³.

Let’s set up an example here. We’ll make structs as above, but then we’ll pass a pointer to a struct child into a function that needs a pointer to a struct parent… and it’ll still work.

#include <stdio.h>

struct parent {
    int a, b;
};

struct child {
    struct parent super;  // MUST be first
    int c, d;
};

// Making the argument `void*` so we can pass any type into it
// (namely a struct parent or struct child)
void print_parent(void *p)
{
    // Expects a struct parent--but a struct child will also work
    // because the pointer points to the struct parent in the first
    // field:
    struct parent *self = p;

    printf("Parent: %d, %d\n", self->a, self->b);
}

void print_child(struct child *self)
{
    printf("Child: %d, %d\n", self->c, self->d);
}

int main(void)
{
    struct child c = {.super.a=1, .super.b=2, .c=3, .d=4};

    print_child(&c);
    print_parent(&c);  // Also works even though it's a struct child!
}

See what we did on the last line of main()? We called print_parent() but passed a struct child* as the argument! Even though print_parent() needs the argument to point to a struct parent, we’re getting away with it because the first field in the struct child is a struct parent.

Again, this works because a pointer to a struct has the same value as a pointer to the first field in that struct.

20.7 Bit-Fields

In my experience, these are rarely used, but you might see them out there from time to time, especially in lower-level applications that pack bits together into larger spaces.

Let’s take a look at some code to demonstrate a use case:

#include <stdio.h>

struct foo {
    unsigned int a;
    unsigned int b;
    unsigned int c;
    unsigned int d;
};

int main(void)
{
    printf("%zu\n", sizeof(struct foo));
}

For me, this prints 16. Which makes sense, since unsigneds are 4 bytes on my system.

But what if we knew that all the values that were going to be stored in a and b could be stored in 5 bits, and the values in c, and d could be stored in 3 bits? That’s only a total 16 bits. Why have 128 bits reserved for them if we’re only going to use 16?

Well, we can tell C to pretty-please try to pack these values in. We can specify the maximum number of bits that values can take (from 1 up the size of the containing type).

We do this by putting a colon after the field name, followed by the field width in bits.

struct foo {
    unsigned int a:5;
    unsigned int b:5;
    unsigned int c:3;
    unsigned int d:3;
};

Now when I ask C how big my struct foo is, it tells me 4! It was 16 bytes, but now it’s only 4. It has “packed” those 4 values down into 4 bytes, which is a four-fold memory savings.

The tradeoff is, of course, that the 5-bit fields can only hold values from 0-31 and the 3-bit fields can only hold values from 0-7. But life’s all about compromise, after all.

20.7.1 Non-Adjacent Bit-Fields

A gotcha: C will only combine adjacent bit-fields. If they’re interrupted by non-bit-fields, you get no savings:

    struct foo {            // sizeof(struct foo) == 16 (for me)
        unsigned int a:1;   // since a is not adjacent to c.
        unsigned int b;
        unsigned int c:1;
        unsigned int d;
    };

In that example, since a is not adjacent to c, they are both “packed” in their own ints.

So we have one int each for a, b, c, and d. Since my ints are 4 bytes, that’s a grand total of 16 bytes.

A quick rearrangement yields some space savings from 16 bytes down to 12 bytes (on my system):

    struct foo {            // sizeof(struct foo) == 12 (for me)
        unsigned int a:1;
        unsigned int c:1;
        unsigned int b;
        unsigned int d;
    };

And now, since a is next to c, the compiler puts them together into a single int.

So we have one int for a combined a and c, and one int each for b and d. For a grand total of 3 ints, or 12 bytes.

Put all your bitfields together to get the compiler to combine them.

20.7.2 Signed or Unsigned `int`s

If you just declare a bit-field to be int, the different compilers will treat it as signed or unsigned. Just like the situation with char.

Be specific about the signedness when using bit-fields.

20.7.3 Unnamed Bit-Fields

In some specific circumstances, you might need to reserve some bits for hardware reasons, but not need to use them in code.

For example, let’s say you have a byte where the top 2 bits have a meaning, the bottom 1 bit has a meaning, but the middle 5 bits do not get used by you¹¹⁴.

We could do something like this:

    struct foo {
        unsigned char a:2;
        unsigned char dummy:5;
        unsigned char b:1;
    };

And that works—in our code we use a and b, but never dummy. It’s just there to eat up 5 bits to make sure a and b are in the “required” (by this contrived example) positions within the byte.

C allows us a way to clean this up: unnamed bit-fields. You can just leave the name (dummy) out in this case, and C is perfectly happy for the same effect:

    struct foo {
        unsigned char a:2;
        unsigned char :5;   // <-- unnamed bit-field!
        unsigned char b:1;
    };

20.7.4 Zero-Width Unnamed Bit-Fields

Some more esoterica out here… Let’s say you were packing bits into an unsigned int, and you needed some adjacent bit-fields to pack into the next unsigned int.

That is, if you do this:

    struct foo {
        unsigned int a:1;
        unsigned int b:2;
        unsigned int c:3;
        unsigned int d:4;
    };

the compiler packs all those into a single unsigned int. But what if you needed a and b in one int, and c and d in a different one?

There’s a solution for that: put an unnamed bit-field of width 0 where you want the compiler to start anew with packing bits in a different int:

    struct foo {
        unsigned int a:1;
        unsigned int b:2;
        unsigned int :0;   // <--Zero-width unnamed bit-field
        unsigned int c:3;
        unsigned int d:4;
    };

It’s analogous to an explicit page break in a word processor. You’re telling the compiler, “Stop packing bits in this unsigned, and start packing them in the next one.”

By adding the zero-width unnamed bit field in that spot, the compiler puts a and b in one unsigned int, and c and d in another unsigned int. Two total, for a size of 8 bytes on my system (unsigned ints are 4 bytes each).

20.8 Unions

These are basically just like structs, except the fields overlap in memory. The union will be only large enough for the largest field, and you can only use one field at a time.

It’s a way to reuse the same memory space for different types of data.

You declare them just like structs, except it’s union. Take a look at this:

    union foo {
        int a, b, c, d, e, f;
        float g, h;
        char i, j, k, l;
    };

Now, that’s a lot of fields. If this were a struct, my system would tell me it took 36 bytes to hold it all.

But it’s a union, so all those fields overlap in the same stretch of memory. The biggest one is int (or float), taking up 4 bytes on my system. And, indeed, if I ask for the sizeof the union foo, it tells me 4!

The tradeoff is that you can only portably use one of those fields at a time. If you try to read from a field that was not the last one written to, the behavior is unspecified.

Let’s take that crazy union and first store an int in it, then a float. Then we’ll print out the int again to see what’s in there—even though, since it wasn’t the last value we stored, the result is unspecified.

#include <stdio.h>

union foo {
    int a, b, c, d, e, f;
    float g, h;
    char i, j, k, l;
};

int main(void)
{
    union foo x;

    x.a = 12;
    printf("%d\n", x.a);  // OK--x.a was the last thing we stored into

    x.g = 3.141592;
    printf("%f\n", x.g);  // OK--x.g was the last thing we stored into

    printf("%d\n", x.a);  // Unspecified behavior!
}

On my machine, this prints:

    12
    3.141592
    1078530008

Probably deep down the decimal value 1078530008 is probably the same pattern of bits as 3.141592, but the spec makes no guarantees about this.

20.8.1 Pointers to `union`s

If you have a pointer to a union, you can cast that pointer to any of the types of the fields in that union and get the values out that way.

In this example, we see that the union has ints and floats in it. And we get pointers to the union, but we cast them to int* and float* types (the cast silences compiler warnings). And then if we dereference those, we see that they have the values we stored directly in the union.

#include <stdio.h>

union foo {
    int a, b, c, d, e, f;
    float g, h;
    char i, j, k, l;
};

int main(void)
{
    union foo x;

    int *foo_int_p = (int *)&x;
    float *foo_float_p = (float *)&x;

    x.a = 12;
    printf("%d\n", x.a);           // 12
    printf("%d\n", *foo_int_p);    // 12, again

    x.g = 3.141592;
    printf("%f\n", x.g);           // 3.141592
    printf("%f\n", *foo_float_p);  // 3.141592, again
}

The reverse is also true. If we have a pointer to a type inside the union, we can cast that to a pointer to the union and access its members.

    union foo x;
    int *foo_int_p = (int *)&x;             // Pointer to int field
    union foo *p = (union foo *)foo_int_p;  // Back to pointer to union
    
    p->a = 12;  // This line the same as...
    x.a = 12;   // this one.

All this just lets you know that, under the hood, all these values in a union start at the same place in memory, and that’s the same as where the entire union is.

21 Characters and Strings II

We’ve talked about how char types are actually just small integer types… but it’s the same for a character in single quotes.

But a string in double quotes is type const char *.

Turns out there are few more types of strings and characters, and it leads down one of the most infamous rabbit holes in the language: the whole multibyte/wide/Unicode/localization thingy.

We’re going to peer into that rabbit hole, but not go in. …Yet!

21.1 Escape Sequences

We’re used to strings and characters with regular letters, punctuation, and numbers:

    char *s = "Hello!";
    char t = 'c';

But what if we want some special characters in there that we can’t type on the keyboard because they don’t exist (e.g. “€”), or even if we want a character that’s a single quote? We clearly can’t do this:

    char t = ''';

To do these things, we use something called escape sequences. These are the backslash character (\) followed by another character. The two (or more) characters together have special meaning.

For our single quote character example, we can put an escape (that is, \) in front of the central single quote to solve it:

    char t = '\'';

Now C knows that \' means just a regular quote we want to print, not the end of the character sequence.

You can say either “backslash” or “escape” in this context (“escape that quote”) and C devs will know what you’re talking about. Also, “escape” in this context is different than your Esc key or the ASCII ESC code.

21.1.1 Frequently-used Escapes

In my humble opinion, these escape characters make up 99.2%¹¹⁵ of all escapes.

Code	Description
`\n`	Newline character—when printing, continue subsequent output on the next line
`\'`	Single quote—used for a single quote character constant
`\"`	Double quote—used for a double quote in a string literal
`\\`	Backslash—used for a literal `\` in a string or character

Here are some examples of the escapes and what they output when printed.

    printf("Use \\n for newline\n");  // Use \n for newline
    printf("Say \"hello\"!\n");       // Say "hello"!
    printf("%c\n", '\'');             // '

21.1.2 Rarely-used Escapes

But there are more escapes! You just don’t see these as often.

Code	Description
`\a`	Alert. This makes the terminal make a sound or flash, or both!
`\b`	Backspace. Moves the cursor back a character. Doesn’t delete the character.
`\f`	Formfeed. This moves to the next “page”, but that doesn’t have much modern meaning. On my system, this behaves like `\v`.
`\r`	Return. Move to the beginning of the same line.
`\t`	Horizontal tab. Moves to the next horizontal tab stop. On my machine, this lines up on columns that are multiples of 8, but YMMV.
`\v`	Vertical tab. Moves to the next vertical tab stop. On my machine, this moves to the same column on the next line.
`\?`	Literal question mark. Sometimes you need this to avoid trigraphs, as shown below.

21.1.2.1 Single Line Status Updates

A use case for \b or \r is to show status updates that appear on the same line on the screen and don’t cause the display to scroll. Here’s an example that does a countdown from 10. (Note this makes use of the non-standard POSIX function sleep() from <unistd.h>—if you’re not on a Unix-like, search for your platform and sleep for the equivalent.)

#include <stdio.h>
#include <unistd.h>  // Non-standard Unix-likes only for sleep()

int main(void)
{
    for (int i = 10; i >= 0; i--) {
        printf("\rT minus %d second%s... \b", i, i != 1? "s": "");

        fflush(stdout);  // Force output to update

        sleep(1);        // Delay 1 second
    }

    printf("\rLiftoff!             \n");
}

Quite a few things are happening on line 7. First of all, we lead with a \r to get us to the beginning of the current line, then we overwrite whatever’s there with the current countdown. (There’s ternary operator out there to make sure we print 1 second instead of 1 seconds.)

Also, there’s a space after the ... That’s so that we properly overwrite the last . when i drops from 10 to 9 and we get a column narrower. Try it without the space to see what I mean.

And we wrap it up with a \b to back up over that space so the cursor sits at the exact end of the line in an aesthetically-pleasing way.

Note that line 14 also has a lot of spaces at the end to overwrite the characters that were already there from the countdown.

Finally, we have a weird fflush(stdout) in there, whatever that means. Short answer is that most terminals are line buffered by default, meaning they don’t actually display anything until a newline character is encountered. Since we don’t have a newline (we just have \r), without this line, the program would just sit there until Liftoff! and then print everything all in one instant. fflush() overrides this behavior and forces output to happen right now.

21.1.2.2 The Question Mark Escape

Why bother with this? After all, this works just fine:

    printf("Doesn't it?\n");

And it works fine with the escape, too:

    printf("Doesn't it\?\n");   // Note \?

So what’s the point??!

Let’s get more emphatic with another question mark and an exclamation point:

    printf("Doesn't it??!\n");

When I compile this, I get this warning:

    foo.c: In function ‘main’:
    foo.c:5:23: warning: trigraph ??! converted to | [-Wtrigraphs]
        5 |     printf("Doesn't it??!\n");
          |

And running it gives this unlikely result:

    Doesn't it|

So trigraphs? What the heck is this??!

I’m sure we’ll revisit this dusty corner of the language later, but the short of it is the compiler looks for certain triplets of characters starting with ?? and it substitutes other characters in their place. So if you’re on some ancient terminal without a pipe symbol (|) on the keyboard, you can type ??! instead.

You can fix this by escaping the second question mark, like so:

    printf("Doesn't it?\?!\n");

And then it compiles and works as-expected.

These days, of course, no one ever uses trigraphs. But that whole ??! does sometimes appear if you decide to use it in a string for emphasis.

21.1.3 Numeric Escapes

In addition, there are ways to specify numeric constants or other character values inside strings or character constants.

If you know an octal or hexadecimal representation of a byte, you can include that in a string or character constant.

The following table has example numbers, but any hex or octal numbers may be used. Pad with leading zeros if necessary to read the proper digit count.

Code	Description
`\123`	Embed the byte with octal value `123`, 3 digits exactly.
`\x4D`	Embed the byte with hex value `4D`, 2 digits.
`\u2620`	Embed the Unicode character at code point with hex value `2620`, 4 digits.
`\U0001243F`	Embed the Unicode character at code point with hex value `1243F`, 8 digits.

Here’s an example of the less-commonly used octal notation to represent the letter B in between A and C. Normally this would be used for some kind of special unprintable character, but we have other ways to do that, below, and this is just an octal demo:

    printf("A\102C\n");  // 102 is `B` in ASCII/UTF-8

Note there’s no leading zero on the octal number when you include it this way. But it does need to be three characters, so pad with leading zeros if you need to.

But far more common is to use hex constants these days. Here’s a demo that you shouldn’t use, but it demos embedding the UTF-8 bytes 0xE2, 0x80, and 0xA2 in a string, which corresponds to the Unicode “bullet” character (•).

    printf("\xE2\x80\xA2 Bullet 1\n");
    printf("\xE2\x80\xA2 Bullet 2\n");
    printf("\xE2\x80\xA2 Bullet 3\n");

Produces the following output if you’re on a UTF-8 console (or probably garbage if you’re not):

    • Bullet 1
    • Bullet 2
    • Bullet 3

But that’s a crummy way to do Unicode. You can use the escapes \u (16-bit) or \U (32-bit) to just refer to Unicode by code point number. The bullet is 2022 (hex) in Unicode, so you can do this and get more portable results:

    printf("\u2022 Bullet 1\n");
    printf("\u2022 Bullet 2\n");
    printf("\u2022 Bullet 3\n");

Be sure to pad \u with enough leading zeros to get to four characters, and \U with enough to get to eight.

For example, that bullet could be done with \U and four leading zeros:

    printf("\U00002022 Bullet 1\n");

But who has time to be that verbose?

22 Enumerated Types: `enum`

C offers us another way to have constant integer values by name: enum.

For example:

    enum {
      ONE=1,
      TWO=2
    };
    
    printf("%d %d", ONE, TWO);  // 1 2

In some ways, it can be better—or different—than using a #define. Key differences:

enums can only be integer types.
#define can define anything at all.
enums are often shown by their symbolic identifier name in a debugger.
#defined numbers just show as raw numbers which are harder to know the meaning of while debugging.

Since they’re integer types, they can be used any place integers can be used, including in array dimensions and case statements.

Let’s tear into this more.

22.1 Behavior of `enum`

22.1.1 Numbering

enums are automatically numbered unless you override them.

They start at 0, and autoincrement up from there, by default:

    enum {
        SHEEP,  // Value is 0
        WHEAT,  // Value is 1
        WOOD,   // Value is 2
        BRICK,  // Value is 3
        ORE     // Value is 4
    };
    
    printf("%d %d\n", SHEEP, BRICK);  // 0 2

You can force particular integer values, as we saw earlier:

    enum {
      X=2,
      Y=18,
      Z=-2
    };

Duplicates are not a problem:

    enum {
      X=2,
      Y=2,
      Z=2
    };

if values are omitted, numbering continues counting in the positive direction from whichever value was last specified. For example:

    enum {
      A,    // 0, default starting value
      B,    // 1
      C=4,  // 4, manually set
      D,    // 5
      E,    // 6
      F=3   // 3, manually set
      G,    // 4
      H     // 5
    }

22.1.2 Trailing Commas

This is perfectly fine, if that’s your style:

    enum {
      X=2,
      Y=18,
      Z=-2,   // <-- Trailing comma
    };

It’s gotten more popular in languages of the recent decades so you might be pleased to see it.

22.1.3 Scope

enums scope as you’d expect. If at file scope, the whole file can see it. If in a block, it’s local to that block.

It’s really common for enums to be defined in header files so they can be #included at file scope.

22.1.4 Style

As you’ve noticed, it’s common to declare the enum symbols in uppercase (with underscores).

This isn’t a requirement, but is a very, very common idiom.

22.2 Your `enum` is a Type

This is an important thing to know about enum: they’re a type, analogous to how a struct is a type.

You can give them a tag name so you can refer to the type later and declare variables of that type.

Now, since enums are integer types, why not just use int?

In C, the best reason for this is code clarity–it’s a nice, typed way to describe your thinking in code. C (unlike C++) doesn’t actually enforce any values being in range for a particular enum.

Let’s do an example where we declare a variable r of type enum resource that can hold those values:

    // Named enum, type is "enum resource"
    
    enum resource {
        SHEEP,
        WHEAT,
        WOOD,
        BRICK,
        ORE
    };
    
    // Declare a variable "r" of type "enum resource"
    
    enum resource r = BRICK;
    
    if (r == BRICK) {
        printf("I'll trade you a brick for two sheep.\n");
    }

You can also typedef these, of course, though I personally don’t like to.

    typedef enum {
        SHEEP,
        WHEAT,
        WOOD,
        BRICK,
        ORE
    } RESOURCE;
    
    RESOURCE r = BRICK;

Another shortcut that’s legal but rare is to declare variables when you declare the enum:

    // Declare an enum and some initialized variables of that type:
    
    enum {
        SHEEP,
        WHEAT,
        WOOD,
        BRICK,
        ORE
    } r = BRICK, s = WOOD;

You can also give the enum a name so you can use it later, which is probably what you want to do in most cases:

    // Declare an enum and some initialized variables of that type:
    
    enum resource {   // <-- type is now "enum resource"
        SHEEP,
        WHEAT,
        WOOD,
        BRICK,
        ORE
    } r = BRICK, s = WOOD;

In short, enums are a great way to write nice, scoped, typed, clean code.

23 Pointers III: Pointers to Pointers and More

Here’s where we cover some intermediate and advanced pointer usage. If you don’t have pointers down well, review the previous chapters on pointers and pointer arithmetic before starting on this stuff.

23.1 Pointers to Pointers

If you can have a pointer to a variable, and a variable can be a pointer, can you have a pointer to a variable that it itself a pointer?

Yes! This is a pointer to a pointer, and it’s held in variable of type pointer-pointer.

Before we tear into that, I want to try for a gut feel for how pointers to pointers work.

Remember that a pointer is just a number. It’s a number that represents an index in computer memory, typically one that holds a value we’re interested in for some reason.

That pointer, which is a number, has to be stored somewhere. And that place is memory, just like everything else¹¹⁶.

But because it’s stored in memory, it must have an index it’s stored at, right? The pointer must have an index in memory where it is stored. And that index is a number. It’s the address of the pointer. It’s a pointer to the pointer.

Let’s start with a regular pointer to an int, back from the earlier chapters:

#include <stdio.h>

int main(void)
{
    int x = 3490;  // Type: int
    int *p = &x;   // Type: pointer to an int

    printf("%d\n", *p);  // 3490
}

Straightforward enough, right? We have two types represented: int and int*, and we set up p to point to x. Then we can dereference p on line 8 and print out the value 3490.

But, like we said, we can have a pointer to any variable… so does that mean we can have a pointer to p?

In other words, what type is this expression?

    int x = 3490;  // Type: int
    int *p = &x;   // Type: pointer to an int
    
    &p  // <-- What type is the address of p? AKA a pointer to p?

If x is an int, then &x is a pointer to an int that we’ve stored in p which is type int*. Follow? (Repeat this paragraph until you do!)

And therefore &p is a pointer to an int*, AKA a “pointer to a pointer to an int”. AKA “int-pointer-pointer”.

Got it? (Repeat the previous paragraph until you do!)

We write this type with two asterisks: int **. Let’s see it in action.

#include <stdio.h>

int main(void)
{
    int x = 3490;  // Type: int
    int *p = &x;   // Type: pointer to an int
    int **q = &p;  // Type: pointer to pointer to int

    printf("%d %d\n", *p, **q);  // 3490 3490
}

Let’s make up some pretend addresses for the above values as examples and see what these three variables might look like in memory. The address values, below are just made up by me for example purposes:

Variable	Stored at Address	Value Stored There
`x`	`28350`	`3490`—the value from the code
`p`	`29122`	`28350`—the address of `x`!
`q`	`30840`	`29122`—the address of `p`!

Indeed, let’s try it for real on my computer¹¹⁷ and print out the pointer values with %p and I’ll do the same table again with actual references (printed in hex).

Variable	Stored at Address	Value Stored There
`x`	`0x7ffd96a07b94`	`3490`—the value from the code
`p`	`0x7ffd96a07b98`	`0x7ffd96a07b94`—the address of `x`!
`q`	`0x7ffd96a07ba0`	`0x7ffd96a07b98`—the address of `p`!

You can see those addresses are the same except the last byte, so just focus on those.

On my system, ints are 4 bytes, which is why we’re seeing the address go up by 4 from x to p¹¹⁸ and then goes up by 8 from p to q. On my system, all pointers are 8 bytes.

Does it matter if it’s an int* or an int**? Is one more bytes than the other? Nope! Remember that all pointers are addresses, that is indexes into memory. And on my machine you can represent an index with 8 bytes… doesn’t matter what’s stored at that index.

Now check out what we did there on line 9 of the previous example: we double dereferenced q to get back to our 3490.

This is the important bit about pointers and pointers to pointers:

You can get a pointer to anything with & (including to a pointer!)
You can get the thing a pointer points to with * (including a pointer!)

So you can think of & as being used to make pointers, and * being the inverse—it goes the opposite direction of &—to get to the thing pointed to.

In terms of type, each time you &, that adds another pointer level to the type.

If you have	Then you run	The result type is
`int x`	`&x`	`int *`
`int *x`	`&x`	`int **`
`int **x`	`&x`	`int ***`
`int ***x`	`&x`	`int ****`

And each time you use dereference (*), it does the opposite:

If you have	Then you run	The result type is
`int ****x`	`*x`	`int ***`
`int ***x`	`*x`	`int **`
`int **x`	`*x`	`int *`
`int *x`	`*x`	`int`

Note that you can use multiple *s in a row to quickly dereference, just like we saw in the example code with **q, above. Each one strips away one level of indirection.

If you have	Then you run	The result type is
`int ****x`	`***x`	`int *`
`int ***x`	`**x`	`int *`
`int **x`	`**x`	`int`

In general, &*E == E¹¹⁹. The dereference “undoes” the address-of.

But & doesn’t work the same way—you can only do those one at a time, and have to store the result in an intermediate variable:

    int x = 3490;     // Type: int
    int *p = &x;      // Type: int *, pointer to an int
    int **q = &p;     // Type: int **, pointer to pointer to int
    int ***r = &q;    // Type: int ***, pointer to pointer to pointer to int
    int ****s = &r;   // Type: int ****, you get the idea
    int *****t = &s;  // Type: int *****

23.1.1 Pointer Pointers and `const`

If you recall, declaring a pointer like this:

    int *const p;

means that you can’t modify p. Trying to p++ would give you a compile-time error.

But how does that work with int ** or int ***? Where does the const go, and what does it mean?

Let’s start with the simple bit. The const right next to the variable name refers to that variable. So if you want an int*** that you can’t change, you can do this:

    int ***const p;
    
    p++;  // Not allowed

But here’s where things get a little weird.

What if we had this situation:

int main(void)
{
    int x = 3490;
    int *const p = &x;
    int **q = &p;
}

When I build that, I get a warning:

    warning: initialization discards ‘const’ qualifier from pointer target type
        7 |     int **q = &p;
          |               ^

What’s going on? The

That is, we’re saying that q is type int **, and if you dereference that, the rightmost * in the type goes away. So after the dereference, we have type int *.

And we’re assigning &p into it which is a pointer to an int *const, or, in other words, int *const *.

But q is int **! A type with different constness on the first *! So we get a warning that the const in p’s int *const * is being ignored and thrown away.

We can fix that by making sure q’s type is at least as const as p.

    int x = 3490;
    int *const p = &x;
    int *const *q = &p;

And now we’re happy.

We could make q even more const. As it is, above, we’re saying, “q isn’t itself const, but the thing it points to is const.” But we could make them both const:

    int x = 3490;
    int *const p = &x;
    int *const *const q = &p;  // More const!

And that works, too. Now we can’t modify q, or the pointer q points to.

23.2 Multibyte Values

We kinda hinted at this in a variety of places earlier, but clearly not every value can be stored in a single byte of memory. Things take up multiple bytes of memory (assuming they’re not chars). You can tell how many bytes by using sizeof. And you can tell which address in memory is the first byte of the object by using the standard & operator, which always returns the address of the first byte.

And here’s another fun fact! If you iterate over the bytes of any object, you get its object representation. Two things with the same object representation in memory are equal.

If you want to iterate over the object representation, you should do it with pointers to unsigned char.

Let’s make our own version of memcpy() that does exactly this:

    void *my_memcpy(void *dest, const void *src, size_t n)
    {
        // Make local variables for src and dest, but of type unsigned char
    
        const unsigned char *s = src;
        unsigned char *d = dest;
    
        while (n-- > 0)   // For the given number of bytes
            *d++ = *s++;  // Copy source byte to dest byte
    
        // Most copy functions return a pointer to the dest as a convenience
        // to the caller
    
        return dest;
    }

(There are some good examples of post-increment and post-decrement in there for you to study, as well.)

It’s important to note that the version, above, is probably less efficient than the one that comes with your system.

But you can pass pointers to anything into it, and it’ll copy those objects. Could be int*, struct animal*, or anything.

Let’s do another example that prints out the object representation bytes of a struct so we can see if there’s any padding in there and what values it has¹²⁰.

#include <stdio.h>

struct foo {
    char a;
    int b;
};

int main(void)
{
    struct foo x = {0x12, 0x12345678};
    unsigned char *p = (unsigned char *)&x;

    for (size_t i = 0; i < sizeof x; i++) {
        printf("%02X\n", p[i]);
    }
}

What we have there is a struct foo that’s built in such a way that should encourage a compiler to inject padding bytes (though it doesn’t have to). And then we get an unsigned char * to the first byte of the struct foo variable x.

From there, all we need to know is the sizeof x and we can loop through that many bytes, printing out the values (in hex for ease).

Running this gives a bunch of numbers as output. I’ve annotated it below to identify where the values were stored:

    12  | x.a == 0x12
    
    AB  |
    BF  | padding bytes with "random" value
    26  |
    
    78  |
    56  | x.b == 0x12345678
    34  |
    12  |

On all systems, sizeof(char) is 1, and we see that first byte at the top of the output holding the value 0x12 that we stored there.

Then we have some padding bytes—for me, these varied from run to run.

Finally, on my system, sizeof(int) is 4, and we can see those 4 bytes at the end. Notice how they’re the same bytes as are in the hex value 0x12345678, but strangely in reverse order¹²¹.

So that’s a little peek under the hood at the bytes of a more complex entity in memory.

23.3 The `NULL` Pointer and Zero

These things can be used interchangeably:

NULL
0
'\0'
(void *)0

Personally, I always use NULL when I mean NULL, but you might see some other variants from time to time. Though '\0' (a byte with all bits set to zero) will also compare equal, it’s weird to compare it to a pointer; you should compare NULL against the pointer. (Of course, lots of times in string processing, you’re comparing the thing the pointer points to to '\0', and that’s right.)

0 is called the null pointer constant, and, when compared to or assigned into another pointer, it is converted to a null pointer of the same type.

23.4 Pointers as Integers

You can cast pointers to integers and vice-versa (since a pointer is just an index into memory), but you probably only ever need to do this if you’re doing some low-level hardware stuff. The results of such machinations are implementation-defined, so they aren’t portable. And weird things could happen.

C does make one guarantee, though: you can convert a pointer to a uintptr_t type and you’ll be able to convert it back to a pointer without losing any data.

uintptr_t is defined in <stdint.h>¹²².

Additionally, if you feel like being signed, you can use intptr_t to the same effect.

23.5 Pointer Differences

As you know from the section on pointer arithmetic, you can subtract one pointer from another¹²³ to get the difference between them in count of array elements.

Now the type of that difference is something that’s up to the implementation, so it could vary from system to system.

To be more portable, you can store the result in a variable of type ptrdiff_t defined in <stddef.h>.

    int cats[100];
    
    int *f = cats + 20;
    int *g = cats + 60;
    
    ptrdiff_t d = g - f;  // difference is 40

And you can print it by prefixing the integer format specifier with t:

    printf("%td\n", d);  // Print decimal: 40
    printf("%tX\n", d);  // Print hex:     28

23.6 Pointers to Functions

Functions are just collections of machine instructions in memory, so there’s no reason we can’t get a pointer to the first instruction of the function.

And then call it.

This can be useful for passing a pointer to a function into another function as an argument. Then the second one could call whatever was passed in.

The tricky part with these, though, is that C needs to know the type of the variable that is the pointer to the function.

And it would really like to know all the details.

Like “this is a pointer to a function that takes two int arguments and returns void”.

How do you write all that down so you can declare a variable?

Well, it turns out it looks very much like a function prototype, except with some extra parentheses:

    // Declare p to be a pointer to a function.
    // This function returns a float, and takes two ints as arguments.
    
    float (*p)(int, int);

Also notice that you don’t have to give the parameters names. But you can if you want; they’re just ignored.

    // Declare p to be a pointer to a function.
    // This function returns a float, and takes two ints as arguments.
    
    float (*p)(int a, int b);

So now that we know how to declare a variable, how do we know what to assign into it? How do we get the address of a function?

Turns out there’s a shortcut just like with getting a pointer to an array: you can just refer to the bare function name without parens. (You can put an & in front of this if you like, but it’s unnecessary and not idiomatic.)

Once you have a pointer to a function, you can call it just by adding parens and an argument list.

Let’s do a simple example where I effectively make an alias for a function by setting a pointer to it. Then we’ll call it.

This code prints out 3490:

#include <stdio.h>

void print_int(int n)
{
    printf("%d\n", n);
}

int main(void)
{
    // Assign p to point to print_int:

    void (*p)(int) = print_int;

    p(3490);          // Call print_int via the pointer
}

Notice how the type of p represents the return value and parameter types of print_int. It has to, or else C will complain about incompatible pointer types.

One more example here shows how we might pass a pointer to a function as an argument to another function.

We’ll write a function that takes a couple integer arguments, plus a pointer to a function that operates on those two arguments. Then it prints the result.

#include <stdio.h>

int add(int a, int b)
{
    return a + b;
}

int mult(int a, int b)
{
    return a * b;
}

void print_math(int (*op)(int, int), int x, int y)
{
    int result = op(x, y);

    printf("%d\n", result);
}

int main(void)
{
    print_math(add, 5, 7);   // 12
    print_math(mult, 5, 7);  // 35
}

Take a moment to digest that. The idea here is that we’re going to pass a pointer to a function to print_math(), and it’s going to call that function to do some math.

This way we can change the behavior of print_math() by passing another function into it. You can see we do that on lines 22-23 when we pass in pointers to functions add and mult, respectively.

Now, on line 13, I think we can all agree the function signature of print_math() is a sight to behold. And, if you can believe it, this one is actually pretty straight-forward compared to some things you can construct¹²⁴.

But let’s digest it. Turns out there are only three parameters, but they’re a little hard to see:

    //                      op             x      y
    //              |-----------------|  |---|  |---|
    void print_math(int (*op)(int, int), int x, int y)

The first, op, is a pointer to a function that takes two ints as arguments and returns an int. This matches the signatures for both add() and mult().

The second and third, x and y, are just standard int parameters.

Slowly and deliberately let your eyes play over the signature while you identify the working parts. One thing that always stands out for me is the sequence (*op)(, the parens and the asterisk. That’s the giveaway it’s a pointer to a function.

Finally, jump back to the Pointers II chapter for a pointer-to-function example using the built-in qsort().

24 Bitwise Operations

These numeric operations effectively allow you to manipulate individual bits in variables, fitting since C is such a low-level langauge¹²⁵.

If you’re not familiar with bitwise operations, Wikipedia has a good bitwise article ¹²⁶.

24.1 Bitwise AND, OR, XOR, and NOT

For each of these, the usual arithmetic conversions take place on the operands (which in this case must be an integer type), and then the appropriate bitwise operation is performed.

Operation	Operator	Example
AND	`&`	`a = b & c`
OR	`\|`	`a = b \| c`
XOR	`^`	`a = b ^ c`
NOT	`~`	`a = ~c`

Note how they’re similar to the Boolean operators && and ||.

These have assignment shorthand variants similar to += and -=:

Operator	Example	Longhand equivalent
`&=`	`a &= c`	`a = a & c`
`\|=`	`a \|= c`	`a = a \| c`
`^=`	`a ^= c`	`a = a ^ c`

24.2 Bitwise Shift

For these, the integer promotions are performed on each operand (which must be an integer type) and then a bitwise shift is executed. The type of the result is the type of the promoted left operand.

New bits are filled with zeros, with a possible exception noted in the implementation-defined behavior, below.

Operation	Operator	Example
Shift left	`<<`	`a = b << c`
Shift right	`>>`	`a = b >> c`

There’s also the same similar shorthand for shifting:

Operator	Example	Longhand equivalent
`>>=`	`a >>= c`	`a = a >> c`
`<<=`	`a <<= c`	`a = a << c`

Watch for undefined behavior: no negative shifts, and no shifts that are larger than the size of the promoted left operand.

Also watch for implementation-defined behavior: if you right-shift a negative number, the results are implementation-defined. (It’s perfectly fine to right-shift a signed int, just make sure it’s positive.)

25 Variadic Functions

Variadic is a fancy word for functions that take arbitrary numbers of arguments.

A regular function takes a specific number of arguments, for example:

    int add(int x, int y)
    {
        return x + y;
    }

You can only call that with exactly two arguments which correspond to parameters x and y.

    add(2, 3);
    add(5, 12);

But if you try it with more, the compiler won’t let you:

    add(2, 3, 4);  // ERROR
    add(5);        // ERROR

Variadic functions get around this limitation to a certain extent.

We’ve already seen a famous example in printf()! You can pass all kinds of things to it.

    printf("Hello, world!\n");
    printf("The number is %d\n", 2);
    printf("The number is %d and pi is %f\n", 2, 3.14159);

It seems to not care how many arguments you give it!

Well, that’s not entirely true. Zero arguments will give you an error:

    printf();  // ERROR

This leads us to one of the limitations of variadic functions in C: they must have at least one argument.

But aside from that, they’re pretty flexible, even allows arguments to have different types just like printf() does.

Let’s see how they work!

25.1 Ellipses in Function Signatures

So how does it work, syntactically?

What you do is put all the arguments that must be passed first (and remember there has to be at least one) and after that, you put .... Just like this:

    void func(int a, ...)   // Literally 3 dots here

Here’s some code to demo that:

    #include <stdio.h>
    
    void func(int a, ...)
    {
        printf("a is %d\n", a);  // Prints "a is 2"
    }
    
    int main(void)
    {
        func(2, 3, 4, 5, 6);
    }

So, great, we can get that first argument that’s in variable a, but what about the rest of the arguments? How do you get to them?

Here’s where the fun begins!

25.2 Getting the Extra Arguments

You’re going to want to include <stdarg.h> to make any of this work.

First things first, we’re going to use a special variable of type va_list (variable argument list) to keep track of which variable we’re accessing at a time.

The idea is that we first start processing arguments with a call to va_start(), process each argument in turn with va_arg(), and then, when done, wrap it up with va_end().

When you call va_start(), you need to pass in the last named parameter (the one just before the ...) so it knows where to start looking for the additional arguments.

And when you call va_arg() to get the next argument, you have to tell it the type of argument to get next.

Here’s a demo that adds together an arbitrary number of integers. The first argument is the number of integers to add together. We’ll make use of that to figure out how many times we have to call va_arg().

#include <stdio.h>
#include <stdarg.h>

int add(int count, ...)
{
    int total = 0;
    va_list va;

    va_start(va, count);   // Start with arguments after "count"

    for (int i = 0; i < count; i++) {
        int n = va_arg(va, int);   // Get the next int

        total += n;
    }

    va_end(va);  // All done

    return total;
}

int main(void)
{
    printf("%d\n", add(4, 6, 2, -4, 17));  // 6 + 2 - 4 + 17 = 21
    printf("%d\n", add(2, 22, 44));        // 22 + 44 = 66
}

When printf() is called, it uses the number of %ds (or whatever) in the format string to know how many more arguments there are!

If the syntax of va_arg() is looking strange to you (because of that loose type name floating around in there), you’re not alone. These are implemented with preprocessor macros in order to get all the proper magic in there.

25.3 `va_list` Functionality

What is that va_list variable we’re using up there? It’s an opaque variable¹²⁷ that holds information about which argument we’re going to get next with va_arg(). You see how we just call va_arg() over and over? The va_list variable is a placeholder that’s keeping track of progress so far.

But we have to initialize that variable to some sensible value. That’s where va_start() comes into play.

When we called va_start(va, count), above, we were saying, “Initialize the va variable to point to the variable argument immediately after count.”

And that’s why we need to have at least one named variable in our argument list¹²⁸.

Once you have that pointer to the initial parameter, you can easily get subsequent argument values by calling va_arg() repeatedly. When you do, you have to pass in your va_list variable (so it can keep on keeping track of where you are), as well as the type of argument you’re about to copy off.

It’s up to you as a programmer to figure out which type you’re going to pass to va_arg(). In the above example, we just did ints. But in the case of printf(), it uses the format specifier to determine which type to pull off next.

And when you’re done, call va_end() to wrap it up. You must (the spec says) call this on a particular va_list variable before you decide to call either va_start() or va_copy() on it again. I know we haven’t talked about va_copy() yet.

So the standard progression is:

va_start() to initialize your va_list variable
Repeatedly va_arg() to get the values
va_end() to deinitialize your va_list variable

I also mentioned va_copy() up there; it makes a copy of your va_list variable in the exact same state. That is, if you haven’t started with va_arg() with the source variable, the new one won’t be started, either. If you’ve consumed 5 variables with va_arg() so far, the copy will also reflect that.

va_copy() can be useful if you need to scan ahead through the arguments but need to also remember your current place.

25.4 Library Functions That Use `va_list`s

One of the other uses for these is pretty cool: writing your own custom printf() variant. It would be a pain to have to handle all those format specifiers right? All zillion of them?

Luckily, there are printf() variants that accept a working va_list as an argument. You can use these to wrap up and make your own custom printf()s!

These functions start with the letter v, such as vprintf(), vfprintf(), vsprintf(), and vsnprintf(). Basically all your printf() golden oldies except with a v in front.

Let’s make a function my_printf() that works just like printf() except it takes an extra argument up front.

#include <stdio.h>
#include <stdarg.h>

int my_printf(int serial, const char *format, ...)
{
    va_list va;

    // Do my custom work
    printf("The serial number is: %d\n", serial);

    // Then pass the rest off to vprintf()
    va_start(va, format);
    int rv = vprintf(format, va);
    va_end(va);

    return rv;
}

int main(void)
{
    int x = 10;
    float y = 3.2;

    my_printf(3490, "x is %d, y is %f\n", x, y);
}

See what we did there? On lines 12-14 we started a new va_list variable, and then just passed it right into vprintf(). And it knows just want to do with it, because it has all the printf() smarts built-in.

We still have to call va_end() when we’re done, though, so don’t forget that!

26 Locale and Internationalization

Localization is the process of making your app ready to work well in different locales (or countries).

As you might know, not everyone uses the same character for decimal points or for thousands separators… or for currency.

These locales have names, and you can select one to use. For example, a US locale might write a number like:

100,000.00

Whereas in Brazil, the same might be written with the commas and decimal points swapped:

100.000,00

Makes it easier to write your code so it ports to other nationalities with ease!

Well, sort of. Turns out C only has one built-in locale, and it’s limited. The spec really leaves a lot of ambiguity here; it’s hard to be completely portable.

But we’ll do our best!

26.1 Setting the Localization, Quick and Dirty

For these calls, include <locale.h>.

There is basically one thing you can portably do here in terms of declaring a specific locale. This is likely what you want to do if you’re going to do locale anything:

    set_locale(LC_ALL, "");  // Use this environment's locale for everything

You’ll want to call that so that the program gets initialized with your current locale.

Getting into more details, there is one more thing you can do and stay portable:

    set_locale(LC_ALL, "C");  // Use the default C locale

but that’s called by default every time your program starts, so there’s not much need to do it yourself.

In that second string, you can specify any locale supported by your system. This is completely system-dependent, so it will vary. On my system, I can specify this:

    setlocale(LC_ALL, "en_US.UTF-8");  // Non-portable!

And that’ll work. But it’s only portable to systems which have that exact same name for that exact same locale, and you can’t guarantee it.

By passing in an empty string ("") for the second argument, you’re telling C, “Hey, figure out what the current locale on this system is so I don’t have to tell you.”

26.2 Getting the Monetary Locale Settings

Because moving green pieces of paper around promises to be the key to happiness¹²⁹, let’s talk about monetary locale. When you’re writing portable code, you have to know what to type for cash, right? Whether that’s “$”, “€”, “¥”, or “£”.

How can you write that code without going insane? Luckily, once you call setlocale(LC_ALL, ""), you can just look these up with a call to localeconv():

    struct lconv *x = localeconv();

This function returns a pointer to a statically-allocated struct lconv that has all that juicy information you’re looking for.

Here are the fields of struct lconv and their meanings.

“negative”, and int_ means “international”. Though a lot of these are type char or char*, most (or the strings they point to) are actually treated as integers¹³⁰.

Before we go further, know that CHAR_MAX (from <limits.h>) is the maximum value that can be held in a char. And that many of the following char values use that to indicate the value isn’t available in the given locale.

Field	Description
`char *mon_decimal_point`	Decimal pointer character for money, e.g. `"."`.
`char *mon_thousands_sep`	Thousands separator character for money, e.g. `","`.
`char *mon_grouping`	Grouping description for money (see below).
`char *positive_sign`	Positive sign for money, e.g. `"+"` or `""`.
`char *negative_sign`	Negative sign for money, e.g. `"-"`.
`char *currency_symbol`	Currency symbol, e.g. `"$"`.
`char frac_digits`	When printing monetary amounts, how many digits to print past the decimal point, e.g. `2`.
`char p_cs_precedes`	`1` if the `currency_symbol` comes before the value for a non-negative monetary amount, `0` if after.
`char n_cs_precedes`	`1` if the `currency_symbol` comes before the value for a negative monetary amount, `0` if after.
`char p_sep_by_space`	Determines the separation of the `currency symbol` from the value for non-negative amounts (see below).
`char n_sep_by_space`	Determines the separation of the `currency symbol` from the value for negative amounts (see below).
`char p_sign_posn`	Determines the `positive_sign` position for non-negative values.
`char p_sign_posn`	Determines the `positive_sign` position for negative values.
`char *int_curr_symbol`	International currency symbol, e.g. `"USD "`.
`char int_frac_digits`	International value for `frac_digits`.
`char int_p_cs_precedes`	International value for `p_cs_precedes`.
`char int_n_cs_precedes`	International value for `n_cs_precedes`.
`char int_p_sep_by_space`	International value for `p_sep_by_space`.
`char int_n_sep_by_space`	International value for `n_sep_by_space`.
`char int_p_sign_posn`	International value for `p_sign_posn`.
`char int_n_sign_posn`	International value for `n_sign_posn`.

26.2.1 Monetary Digit Grouping

OK, this is a trippy one. mon_grouping is a char*, so you might be thinking it’s a string. But in this case, no, it’s really an array of chars. It should always end either with a 0 or CHAR_MAX.

These values describe how to group sets of numbers in currency to the left of the decimal (the whole number part).

For example, we might have:

      2   1   0
     --- --- ---
    $100,000,000.00

These are groups of three. Group 0 (just left of the decimal) has 3 digits. Group 1 (next group to the left) has 3 digits, and the last one also has 3.

So we could describe these groups, from the right (the decimal) to the left with a bunch of integer values representing the group sizes:

    3 3 3

And that would work for values up to $100,000,000.

But what if we had more? We could keep adding 3s…

    3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

but that’s crazy. Luckily, we can specify 0 to indicate that the previous group size repeats:

3 0

Which means to repeat every 3. That would handle $100, $1,000, $10,000, $10,000,000, $100,000,000,000, and so on.

You can go legitimately crazy with these to indicate some weird groupings.

For example:

    4 3 2 1 0

would indicate:

    $1,0,0,0,0,00,000,0000.00

One more value that can occur is CHAR_MAX. This indicates that no more grouping should occur, and can appear anywhere in the array, including the first value.

    3 2 CHAR_MAX

would indicate:

    100000000,00,000.00

for example.

And simply having CHAR_MAX in the first array position would tell you there was to be no grouping at all.

26.2.2 Separators and Sign Position

All the sep_by_space variants deal with spacing around the currency sign. Valid values are:

Value	Description
`0`	No space between currency symbol and value.
`1`	Separate the currency symbol (and sign, if any) from the value with a space.
`2`	Separate the sign symbol from the currency symbol (if adjacent) with a space, otherwise separate the sign symbol from the value with a space.

The sign_posn variants are determined by the following values:

Value	Description
`0`	Put parens around the value and the currency symbol.
`1`	Put the sign string in front of the currency symbol and value.
`2`	Put the sign string after the currency symbol and value.
`3`	Put the sign string directly in front of the currency symbol.
`4`	Put the sign string directly behind the currency symbol.

26.2.3 Example Values

When I get the values on my system, this is what I see (grouping string displayed as individual byte values):

    mon_decimal_point  = "."
    mon_thousands_sep  = ","
    mon_grouping       = 3 3 0
    positive_sign      = ""
    negative_sign      = "-"
    currency_symbol    = "$"
    frac_digits        = 2
    p_cs_precedes      = 1
    n_cs_precedes      = 1
    p_sep_by_space     = 0
    n_sep_by_space     = 0
    p_sign_posn        = 1
    n_sign_posn        = 1
    int_curr_symbol    = "USD "
    int_frac_digits    = 2
    int_p_cs_precedes  = 1
    int_n_cs_precedes  = 1
    int_p_sep_by_space = 1
    int_n_sep_by_space = 1
    int_p_sign_posn    = 1
    int_n_sign_posn    = 1

26.3 Localization Specifics

Notice how we passed the macro LC_ALL to setlocale() earlier… this hints that there might be some variant that allows you to be more precise about which parts of the locale you’re setting.

Let’s take a look at the values you can see for these:

Macro	Description
`LC_ALL`	Set all of the following to the given locale.
`LC_COLLATE`	Controls the behavior of the `strcoll()` and `strxfrm()` functions.
`LC_CTYPE`	Controls the behavior of the character-handling functions¹³¹.
`LC_MONETARY`	Controls the values returned by `localeconv()`.
`LC_NUMERIC`	Controls the decimal point for the `printf()` family of functions.
`LC_TIME`	Controls time formatting of the `strftime()` and `wcsftime()` time and date printing functions.

It’s pretty common to see LC_ALL being set, but, hey, at least you have options.

Also I should point out that LC_CTYPE is one of the biggies because it ties into wide characters, a significant can of worms that we’ll talk about later.

27 Unicode, Wide Characters, and All That

Before we begin, note that this is an active area of language development in C as it works to get past some, erm, growing pains. When C2x comes out, updates here are probable.

Most people are basically interested in the deceptively simple question, “How do I use such-and-such character set in C?” We’ll get to that. But as we’ll see, it might already work on your system. Or you might have to punt to a third-party library.

We’re going to take about a lot of things this chapter—some are platform agnostic, and some are C-specific.

Let’s get an outline first of what we’re going to look at:

Unicode background
Character encoding background
Source and Execution character Sets
Using Unicode and UTF-8
Using other character types like wchar_t, char16_t, and char32_t

Let’s dive in!

27.1 What is Unicode?

Back in the day, it was popular in the US and much of the world to use a 7-bit or 8-bit encoding for characters in memory. This meant we could have 128 or 256 characters (including non-printable characters) total. That was fine for a US-centric world, but it turns out there are actually other alphabets out there—who knew? Chinese has over 50,000 characters, and that’s not fitting in a byte.

So people came up with all kinds of alternate ways to represent their own custom character sets. And that was fine, but turned into a compatibility nightmare.

To escape it, Unicode was invented. One character set to rule them all. It extends off into infinity (effectively) so we’ll never run out of space for new characters. It has Chinese, Latin, Greek, cuniform, chess symbols, emojis… just about everything, really! And more is being added all the time!

27.2 Code Points

I want to talk about two concepts here. It’s confusing because they’re both numbers… different numbers for the same thing. But bear with me.

Let’s loosely define code point to mean a numeric value representing a character. (Code points can also represent unprintable control characters, but just assume I mean something like the letter “B” or the character “π”.)

Each code point represents a unique character. And each character has a unique numeric code point associated with it.

For example, in Unicode, the numeric value 66 represents “B”, and 960 represents “π”. Other character mappings that aren’t Unicode use different values, potentially, but let’s forget them and concentrate on Unicode, the future!

So that’s one thing: there’s a number that represents each character. In Unicode, these numbers run from 0 to over 1 million.

Got it?

Because we’re about to flip the table a little.

27.3 Encoding

If you recall, an 8-bit byte can hold values from 0-255, inclusive. That’s great for “B” which is 66—that fits in a byte. But “π” is 960, and that doesn’t fit in a byte! We need another byte. How do we store all that in memory? Or what about bigger numbers, like 195,024? That’s going to need a number of bytes to hold.

The Big Question: how are these numbers represented in memory? This is what we call the encoding of the characters.

So we have two things: one is the code point which tells us effectively the serial number of a particular character. And we have the encoding which tells us how we’re going to represent that number in memory.

There are plenty of encodings. You can make up your own right now, if you want¹³². But we’re going to look at some really common encodings that are in use with Unicode.

Encoding	Description
UTF-8	A byte-oriented encoding that uses a variable number of bytes per character. This is the one to use.
UTF-16	A 16-bit per character¹³³ encoding.
UTF-32	A 32-bit per character encoding.

With UTF-16 and UTF-32, the byte order matters, so you might see UTF-16BE for big-endian and UTF-16LE for little-endian. Same for UTF-32. Technically, if unspecified, you should assume big-endian. But since Windows uses UTF-16 extensively and is little-endian, sometimes that is assumed¹³⁴.

Let’s look at some examples. I’m going to write the values in hex because that’s exactly two digits per 8-bit byte, and it makes it easier to see how things are arranged in memory.

Character	Code Point	UTF-16BE	UTF-32BE	UTF-16LE	UTF-32LE	UTF-8
`A`	41	0041	00000041	4100	41000000	41
`B`	42	0042	00000042	4200	42000000	42
`~`	7E	007E	0000007E	7E00	7E000000	7E
`π`	3C0	03C0	000003C0	C003	C0030000	CF80
`€`	20AC	20AC	000020AC	AC20	AC200000	E282AC

Look in there for the patterns. Note that UTF-16BE and UTF-32BE are simply the code point represented directly as 16- and 32-bit values¹³⁵.

Little-endian is the same, except the bytes are in little-endian order.

Then we have UTF-8 at the end. First you might notice that the single-byte code points are represented as a single byte. That’s nice. You might also notice that different code points take different number of bytes. This is a variable-width encoding.

So as soon as we get above a certain value, UTF-8 starts using additional bytes to store the values. And they don’t appear to correlate with the code point value, either.

The details of UTF-8 encoding ¹³⁶ are beyond the scope of this guide, but it’s enough to know that it has a variable number of bytes per code point, and those byte values don’t match up with the code point except for the first 128 code points. If you really want to learn more, Computerphile has a great UTF-8 video with Tom Scott ¹³⁷.

That last bit is a neat thing about Unicode and UTF-8 from a North American perspective: it’s backward compatible with 7-bit ASCII encoding! So if you’re used to ASCII, UTF-8 is the same! Every ASCII-encoded document is also UTF-8 encoded! (But not the other way around, obviously.)

It’s probably that last point more than any other that is driving UTF-8 to take over the world.

27.4 Source and Execution Character Sets

When programming in C, there are (at least) three character sets that are in play:

The one that your code exists on disk as.
The one the compiler translates that into just as compilation begins (the source character set). This might be the same as the one on disk, or it might not.
The one the compiler translates the source character set into for execution (the execution character set). This might be the same as the source character set, or it might not.

Your compiler probably has options to select these character sets at build-time.

The basic character set for both source and execution will contain the following characters:

    A B C D E F G H I J K L M
    N O P Q R S T U V W X Y Z
    a b c d e f g h i j k l m
    n o p q r s t u v w x y z
    0 1 2 3 4 5 6 7 8 9
    ! " # % & ' ( ) * + , - . / :
    ; < = > ? [ \ ] ^ _ { | } ~
    space tab vertical-tab
    form-feed end-of-line

Those are the characters you can use in your source and remain 100% portable.

The execution character set will additionally have characters for alert (bell/flash), backspace, carriage return, and newline.

But most people don’t go to that extreme and freely use their extended character sets in source and executable, especially now that Unicode and UTF-8 are getting more common. I mean, the basic character set doesn’t even allow for @, $, or `!

Notably, it’s a pain (though possible with escape sequences) to enter Unicode characters using only the basic character set.

27.5 Unicode in C

Before I get into encoding in C, let’s talk about Unicode from a code point standpoint. There is a way in C to specify Unicode characters and these will get translated by the compiler into the execution character set¹³⁸.

So how do we do it?

How about the euro symbol, code point 0x20AC. (I’ve written it in hex because both ways of representing it in C require hex.) How can we put that in our C code?

Use the \u escape to put it in a string, e.g. "\u20AC" (case for the hex doesn’t matter). You must put exactly four hex digits after the \u, padding with leading zeros if necessary.

Here’s an example:

    char *s = "\u20AC1.23";
    
    printf("%s\n", s);  // €1.23

So \u works for 16-bit Unicode code points, but what about ones bigger than 16 bits? For that, we need capitals: \U.

For example:

    char *s = "\U0001D4D1";
    
    printf("%s\n", s);  // Prints a mathematical letter "B"

It’s the same as \u, just with 32 bits instead of 16. These are equivalent:

    \u03C0
    \U000003C0

Again, these are translated into the execution character set during compilation. They represent Unicode code points, not any specific encoding. Furthermore, if a Unicode code point is not representable in the execution character set, the compiler can do whatever it wants with it.

Now, you might wonder why you can’t just do this:

    char *s = "€1.23";
    
    printf("%s\n", s);  // €1.23

And you probably can, given a modern compiler. The source character set will be translated for you into the execution character set by the compiler. But compilers are free to puke out if they find any characters that aren’t included in their extended character set, and the € symbol certainly isn’t in the basic character set.

Caveat from the spec: you can’t use \u or \U to encode any code points below 0xA0 except for 0x24 ($), 0x40 (@), and 0x60 (`)—yes, those are precisely the trio of common punctuation marks missing from the basic character set. Apparently this restriction is relaxed in the upcoming version of the spec.

Finally, you can also use these in identifiers in your code, with some restrictions. But I don’t want to get into that here. We’re all about string handling in this chapter.

And that’s about it for Unicode in C (except encoding).

27.6 A Quick Note on UTF-8 Before We Swerve into the Weeds

It could be that your source file on disk, the extended source characters, and the extended execution characters are all in UTF-8 format. And the libraries you use expect UTF-8. This is the glorious future of UTF-8 everywhere.

If that’s the case, and you don’t mind being non-portable to systems that aren’t like that, then just run with it. Stick Unicode characters in your source and data at will. Use regular C strings and be happy.

A lot of things will just work (albeit non-portably) because UTF-8 strings can safely be NUL-terminated just like any other C string. But maybe losing portability in exchange for easier character handling is a tradeoff that’s worth it to you.

There are some caveats, however:

Things like strlen() report the number of bytes in a string, not the number of characters, necessarily. (Use mbstowcs() with a NULL first argument to get the number of characters in a multibyte string.)
The following won’t work properly with characters of more than one byte: strtok(), strchr() (use strstr() instead), strspn()-type functions, toupper(), tolower(), isalpha()-type functions, and probably more. Beware anything that operates on bytes.
printf() variants allow for a way to only print so many bytes of a string¹³⁹. You want to make certain you print the correct number of bytes to end on a character boundary.
If you want to malloc() space for a string, or declare an array of chars for one, be aware that the maximum size could be more than you were expecting. Each character could take up to MB_LEN_MAX bytes (from <limits.h>)—except characters in the basic character set which are guaranteed to be one byte.

And probably others I haven’t discovered. Let me know what pitfalls there are out there…

27.7 Different Character Types

I want to introduce more character types. We’re used to char, right?

But that’s too easy. Let’s make things a lot more difficult! Yay!

27.7.1 Multibyte Characters

First of all, I want to potentially change your thinking about what a string (array of chars) is. These are multibyte strings made up of multibyte characters.

That’s right—your run-of-the-mill string of characters is multibyte.

Even if a particular character in the string is only a single byte, or if a string is made up of only single characters, it’s known as multibyte.

For example:

    char c[128] = "Hello, world!";  // Multibyte string

What we’re saying here is that a particular character that’s not in the basic character set could be composed of multiple bytes. Up to MB_LEN_MAX of them (from <limits.h>). Sure, it only looks like one character on the screen, but it could be multiple bytes.

You can throw Unicode values in there, as well, as we saw earlier:

    char *s = "\u20AC1.23";
    
    printf("%s\n", s);  // €1.23

But here we’re getting into some weirdness, because check this out:

    char *s = "\u20AC1.23";  // €1.23
    
    printf("%zu\n", strlen(s));  // 7!

The string length of "€1.23" is 7?! Yes! Well, on my system, yes! Remember that strlen() returns the number of bytes in the string, not the number of characters. (When we get to “wide characters”, coming up, we’ll see a way to get the number of characters in the string.)`

Note that while C allows individual multibyte char constants, the behavior of these varies by implementation and your compiler might warn on it.

GCC, for example, warns of multi-character character constants for the following two lines (and, on my system, prints out the UTF-8 encoding):

    printf("%x\n", '€');
    printf("%x\n", '\u20ac');

27.7.2 Wide Characters

If you’re not a multibyte character, then you’re a wide character.

A wide character is a single value that can uniquely represent any character in the current locale. It’s analogous to Unicode code points. But it might not be. Or it might be.

Basically, where multibyte character strings are arrays of bytes, wide character strings are arrays of characters. So you can start thinking on a character-by-character basis rather than a byte-by-byte basis (the latter of which gets all messy when characters start taking up variable numbers of bytes).

Wide characters can be represented by a number of types, but the big standout one is wchar_t. It’s the main one.

You might be wondering if you can’t tell if it’s Unicode or not, how does that allow you much flexibility in terms of writing code? wchar_t opens some of those doors, as there are a rich set of function you can use to deal with wchar_t strings (like getting the length, etc.) without caring about the encoding.

27.8 Using Wide Characters and `wchar_t`

Time for a new type: wchar_t. This is the main wide character typre. Remember how a char is only one byte? And a byte’s not enough to represent all characters, potentially? Well, this one is enough.

To use wchar_t, #include <wchar.h>.

How many bytes big is it? Well, it’s not totally clear. Could be 16 bits. Could be 32 bits.

But wait, you’re saying—if it’s only 16 bits, it’s not big enough to hold all the Unicode code points, is it? You’re right—it’s not. The spec doesn’t require it to be. It just has to be able to represent all the characters in the current locale.

This can cause grief with Unicode on platforms with 16-bit wchar_ts (ahem—Windows). But that’s out of scope for this guide.

You can declare a string or character of this type with the L prefix, and you can print them with the %ls (“ell ess”) format specifier. Or print an individual wchar_t with %lc.

    wchar_t *s = L"Hello, world!";
    wchar_t c = L'B';
    
    printf("%ls %lc\n", s, c);

Now—are those characters stored are Unicode code points, or not? Depends on the implementation. But you can test if they are with the macro __STDC_ISO_10646__. If this is defined, the answer is, “It’s Unicode!”

More detailedly, the value in that macro is an integer in the form yyyymm that lets you know what Unicode standard you can rely on—whatever was in effect on that date.

But how do you use them?

27.8.1 Multibyte to `wchar_t` Conversions

So how do we get from the byte-oriented standard strings to the character-oriented wide strings and back?

We can use a couple string conversion functions to make this happen.

First, some naming conventions you’ll see in these functions:

mb: multibyte
wc: wide character
mbs: multibyte string
wcs: wide character string

So if we want to convert a multibyte string to a wide character string, we can call the mbstowcs(). And the other way around: wcstombs().

Conversion Function	Description
`mbtowc()`	Convert a multibyte character to a wide character.
`wctomb()`	Convert a wide character to a multibyte character.
`mbstowcs()`	Convert a multibyte string to a wide string.
`wcstombs()`	Convert a wide string to a multibyte string.

Let’s do a quick demo where we convert a multibyte string to a wide character string, and compare the string lengths of the two using their respective functions.

#include <stdio.h>
#include <stdlib.h>
#include <wchar.h>
#include <string.h>
#include <locale.h>

int main(void)
{
    // Get out of the C locale to one that likely has the euro symbol
    setlocale(LC_ALL, "");

    // Original multibyte string with a euro symbol (Unicode point 20ac)
    char *mb_string = "The cost is \u20ac1.23";  // €1.23
    size_t mb_len = strlen(mb_string);

    // Wide character array that will hold the converted string
    wchar_t wc_string[128];  // Holds up to 128 wide characters

    // Convert the MB string to WC; this returns the number of wide chars
    size_t wc_len = mbstowcs(wc_string, mb_string, 128);

    // Print result--note the %ls for wide char strings
    printf("multibyte: \"%s\" (%zu bytes)\n", mb_string, mb_len);
    printf("wide char: \"%ls\" (%zu characters)\n", wc_string, wc_len);
}

On my system, this outputs:

    multibyte: "The cost is €1.23" (19 bytes)
    wide char: "The cost is €1.23" (17 characters)

(Your system might vary on the number of bytes depending on your locale.)

One interesting thing to note is that mbstowcs(), in addition to converting the multibyte string to wide, returns the length (in characters) of the wide character string. And, in fact, it has a special mode where it only returns the length-in-characters of a given multibyte string: you just pass NULL to the destination, and 0 to the maximum number of characters to convert (this value is ignored).

(In the code below, I’m using my extended source character set—you might have to replace those with \u escapes.)

    setlocale(LC_ALL, "");
    
    // The following string has 7 characters
    size_t len_in_chars = mbstowcs(NULL, "§¶°±π€•", 0));
    
    printf("%zu", len_in_chars);  // 7

And, of course, if you want to convert the other way, it’s wcstombs().

27.9 Wide Character Functionality

Once we’re in wide character land, we have all kinds of functionality at our disposal. I’m just going to summarize a bunch of the functions here, but basically what we have here are the wide character versions of the multibyte string functions that we’re use to. (For example, we know strlen() for multibyte strings; there’s an wcslen() for wide character strings.)

27.9.1 `wint_t`

A lot of these functions use a wint_t to hold single characters, whether they are passed in or returned.

It is related to wchar_t in nature. A wint_t is an integer that can represent all values in the extended character set, and also a special end-of-file character, WEOF.

This is used by a number of single-character-oriented wide character functions.

27.9.2 I/O Stream Orientation

The tl;dr here is to not mix and match byte-oriented functions (like fprintf() with wide-oriented functions (like fwprintf()). Decide if a stream will be byte-oriented or wide-oriented and stick with those types of I/O functions.

In more detail: streams can be either byte-oriented or wide-oriented. When a stream is first created, it has no orientation, but the first read or write will set the orientation.

If you first use a wide operation (like fwprintf()) it will orient the stream wide.

If you first use a byte operation (like fprintf()) it will orient the stream by bytes.

You can manually set an unoriented stream one way or the other with a call to fwide(). You can use that same function to get the orientation of a stream.

If you need to change the orientation mid-flight, you can do it with freopen().

27.9.3 I/O Functions

Typically include <stdio.h> and <wchar.h> for these.

I/O Function	Description
`wprintf()`	Formatted console output.
`wscanf()`	Formatted console input.
`getwchar()`	Character-based console input.
`putwchar()`	Character-based console output.
`fwprintf()`	Formatted file output.
`fwscanf()`	Formatted file input.
`fgetwc()`	Character-based file input.
`fputwc()`	Character-based file output.
`fgetws()`	String-based file input.
`fputws()`	String-based file output.
`swprintf()`	Formatted string output.
`swscanf()`	Formatted string input.
`vfwprintf()`	Variadic formatted file output.
`vfwscanf()`	Variadic formatted file input.
`vswprintf()`	Variadic formatted string output.
`vswscanf()`	Variadic formatted string input.
`vwprintf()`	Variadic formatted console output.
`vwscanf()`	Variadic formatted console input.
`ungetwc()`	Push a wide character back on an output stream.
`fwide()`	Get or set stream multibyte/wide orientation.

27.9.4 Type Conversion Functions

Typically include <wchar.h> for these.

Conversion Function	Description
`wcstod()`	Convert string to `double`.
`wcstof()`	Convert string to `float`.
`wcstold()`	Convert string to `long double`.
`wcstol()`	Convert string to `long`.
`wcstoll()`	Convert string to `long long`.
`wcstoul()`	Convert string to `unsigned long`.
`wcstoull()`	Convert string to `unsigned long long`.

27.9.5 String and Memory Copying Functions

Typically include <wchar.h> for these.

Copying Function	Description
`wcscpy()`	Copy string.
`wcsncpy()`	Copy string, length-limited.
`wmemcpy()`	Copy memory.
`wmemmove()`	Copy potentially-overlapping memory.
`wcscat()`	Concatenate strings.
`wcsncat()`	Concatenate strings, length-limited.

27.9.6 String and Memory Comparing Functions

Typically include <wchar.h> for these.

Comparing Function	Description
`wcscmp()`	Compare strings lexicographically.
`wcsncmp()`	Compare strings lexicographically, length-limited.
`wcscoll()`	Compare strings in dictionary order by locale.
`wmemcmp()`	Compare memory lexicographically.
`wcsxfrm()`	Transform strings into versions such that `wcscmp()` behaves like `wcscoll()`¹⁴⁰.

27.9.7 String Searching Functions

Typically include <wchar.h> for these.

Searching Function	Description
`wcschr()`	Find a character in a string.
`wcsrchr()`	Find a character in a string from the back.
`wmemchr()`	Find a character in memory.
`wcsstr()`	Find a substring in a string.
`wcspbrk()`	Find any of a set of characters in a string.
`wcsspn()`	Find length of substring including any of a set of characters.
`wcscspn()`	Find length of substring before any of a set of characters.
`wcstok()`	Find tokens in a string.

27.9.8 Length/Miscellaneous Functions

Typically include <wchar.h> for these.

Length/Misc Function	Description
`wcslen()`	Return the length of the string.
`wmemset()`	Set characters in memory.
`wcsftime()`	Formatted date and time output.

27.9.9 Character Classification Functions

Include <wctype.h> for these.

Length/Misc Function	Description
`iswalnum()`	True if the character is alphanumeric.
`iswalpha()`	True if the character is alphabetic.
`iswblank()`	True if the character is blank (space-ish, but not a newline).
`iswcntrl()`	True if the character is a control character.
`iswdigit()`	True if the character is a digit.
`iswgraph()`	True if the character is printable (except space).
`iswlower()`	True if the character is lowercase.
`iswprint()`	True if the character is printable (including space).
`iswpunct()`	True if the character is punctuation.
`iswspace()`	True if the character is whitespace.
`iswupper()`	True if the character is uppercase.
`iswxdigit()`	True if the character is a hex digit.
`towlower()`	Convert character to lowercase.
`towupper()`	Convert character to uppercase.

27.10 Parse State, Restartable Functions

We’re going to get a little bit into the guts of multibyte conversion, but this is a good thing to understand, conceptually.

Imagine how your program takes a sequence of multibyte characters and turns them into wide characters, or vice-versa. It might, at some point, be partway through parsing a character, or it might have to wait for more bytes before it makes the determination of the final value.

This parse state is stored in an opaque variable of type mbstate_t and is used every time conversion is performed. That’s how the conversion functions keep track of where they are mid-work.

And if you change to a different character sequence mid-stream, or try to seek to a different place in your input sequence, it could get confused over that.

Now you might want to call me on this one: we just did some conversions, above, and I never mentioned any mbstate_t anywhere.

That’s because the conversion functions like mbstowcs(), wctomb(), etc. each have their own mbstate_t variable that they use. There’s only one per function, though, so if you’re writing multithreaded code, they’re not safe to use.

Fortunately, C defines restartable versions of these functions where you can pass in your own mbstate_t on per-thread basis if you need to. If you’re doing multithreaded stuff, use these!

Quick note on initializing an mbstate_t variable: just memset() it to zero. There is no built-in function to force it to be initialized.

    mbstate_t mbs;
    
    // Set the state to the initial state
    memset(&mbs, 0, sizeof mbs);

Here is a list of the restartable conversion functions—note the naming convension of putting an “r” after the “from” type:

mbrtowc()—multibyte to wide character
wcrtomb()—wide character to multibyte
mbsrtowcs()—multibyte string to wide character string
wcsrtombs()—wide character string to multibyte string

These are really similar to their non-restartable counterparts, except they require you pass in a pointer to your own mbstate_t variable. And also they modify the source string pointer (to help you out if invalid bytes are found), so it might be useful to save a copy of the original.

Here’s the example from earlier in the chapter reworked to pass in our own mbstate_t.

#include <stdio.h>
#include <stdlib.h>
#include <stddef.h>
#include <wchar.h>
#include <string.h>
#include <locale.h>

int main(void)
{
    // Get out of the C locale to one that likely has the euro symbol
    setlocale(LC_ALL, "");

    // Original multibyte string with a euro symbol (Unicode point 20ac)
    char *mb_string = "The cost is \u20ac1.23";  // €1.23
    size_t mb_len = strlen(mb_string);

    // Wide character array that will hold the converted string
    wchar_t wc_string[128];  // Holds up to 128 wide characters

    // Set up the conversion state
    mbstate_t mbs;
    memset(&mbs, 0, sizeof mbs);  // Initial state

    // mbsrtowcs() modifies the input pointer to point at the first
    // invalid character, or NULL if successful. Let's make a copy of
    // the pointer for mbsrtowcs() to mess with so our original is
    // unchanged.
    //
    // This example will probably be successful, but we check farther
    // down to see.
    const char *invalid = mb_string;

    // Convert the MB string to WC; this returns the number of wide chars
    size_t wc_len = mbsrtowcs(wc_string, &invalid, 128, &mbs);

    if (invalid == NULL) {
        printf("No invalid characters found\n");

        // Print result--note the %ls for wide char strings
        printf("multibyte: \"%s\" (%zu bytes)\n", mb_string, mb_len);
        printf("wide char: \"%ls\" (%zu characters)\n", wc_string, wc_len);
    } else {
        ptrdiff_t offset = invalid - mb_string;
        printf("Invalid character at offset %td\n", offset);
    }
}

For the conversion functions that manage their own state, you can reset their internal state to the initial one by passing in NULL for their char* arguments, for example:

    mbstowcs(NULL, NULL, 0);   // Reset the parse state for mbstowcs()
    mbstowcs(dest, src, 100);  // Parse some stuff

For I/O, each wide stream manages its own mbstate_t and uses that for input and output conversions as it goes.`

And some of the byte-oriented I/O functions like printf() and scanf() keep their own internal state while doing their work.

Finally, these restartable conversion functions do actually have their own internal state if you pass in NULL for the mbstate_t parameter. This makes them behave more like their non-restartable counterparts.

27.11 Unicode Encodings and C

In this section, we’ll see what C can (and can’t) do when it comes to three specific Unicode encodings: UTF-8, UTF-16, and UTF-32.

27.11.1 UTF-8

To refresh before this section, read the UTF-8 quick note, above.

Aside from that, what are C’s UTF-8 capabilities?

Well, not much, unfortunately.

You can tell C that you specifically want a string literal to be UTF-8 encoded, and it’ll do it for you. You can prefix a string with u8:

    char *s = u8"Hello, world!";
    
    printf("%s\n", s);   // Hello, world!--if you can output UTF-8

Now, can you put Unicode characters in there?

    char *s = u8"€123";

Sure! If the extended source character set supports it. (gcc does.)

What if it doesn’t? You can specify a Unicode code point with your friendly neighborhood \u and \U, as noted above.

But that’s about it. There’s no portable way in the standard library to take arbirary input and turn it into UTF-8 unless your locale is UTF-8. Or to parse UTF-8 unless your locale is UTF-8.

So if you want to do it, either be in a UTF-8 locale and:

    setlocale(LC_ALL, "");

or figure out a UTF-8 locale name on your local machine and set it explicitly like so:

    setlocale(LC_ALL, "en_US.UTF-8");  // Non-portable name

Or use a third-party library.

27.11.2 UTF-16, UTF-32, `char16_t`, and `char32_t`

char16_t and char32_t are a couple other potentially wide character types with sizes of 16 bits and 32 bits, respectively. Not necessarily wide, because if they can’t represent every character in the current locale, they lose their wide character nature. But the spec refers them as “wide character” types all over the place, so there we are.

These are here to make things a little more Unicode-friendly, potentially.

To use, include <uchar.h>. (That’s “u”, not “w”.)

You can declare a string or character of these types with the u and U prefixes:

    char16_t *s = u"Hello, world!";
    char16_t c = u'B';
    
    char32_t *t = U"Hello, world!";
    char32_t d = U'B';

Now—are values in these stored in UTF-16 or UTF-32? Depends on the implementation.

But you can test to see if they are. If the macros __STDC_UTF_16__ or __STDC_UTF_32__ are defined (to 1) it means the types hold UTF-16 or UTF-32, respectively.

If you’re curious, and I know you are, the values, if UTF-16 or UTF-32, are stored in the native endianess. That is, you should be able to compare them straight up to Unicode code point values:

    char16_t pi = u"\u03C0";  // pi symbol
    
    #if __STDC_UTF_16__
    pi == 0x3C0;  // Always true
    #else
    pi == 0x3C0;  // Probably not true
    #endif

27.11.3 Multibyte Conversions

You can convert from your multibyte encoding to char16_t or char32_t with a number of helper functions.

(Like I said, though, the result might not be UTF-16 or UTF-32 unless the corresponding macro is set to 1.)

All of these functions are restartable (i.e. you pass in your own mbstate_t), and all of them operate character by character¹⁴¹.

Conversion Function	Description
`mbrtoc16()`	Convert a multibyte character to a `char16_t` character.
`mbrtoc32()`	Convert a multibyte character to a `char32_t` character.
`c16rtomb()`	Convert a `char16_t` character to a multibyte character.
`c32rtomb()`	Convert a `char32_t` character to a multibyte character.

27.11.4 Third-Party Libraries

For heavy-duty conversion between different specific encodings, there are a couple mature libraries worth checking out. Note that I haven’t used either of these.

iconv ¹⁴²—Internationalization Conversion, a common POSIX-standard API available on the major platforms.
ICU ¹⁴³—International Components for Unicode. At least one blogger found this easy to use.

If you have more noteworthy libraries, let me know.

28 Exiting a Program

Turns out there are a lot of ways to do this, and even ways to set up “hooks” so that a function runs when a program exits.

In this chapter we’ll dive in and check them out.

We already covered the meaning of the exit status code in the Exit Status section, so jump back there and review if you have to.

All the functions in this section are in <stdlib.h>.

28.1 Normal Exits

We’ll start with the regular ways to exit a program, and then jump to some of the rarer, more esoteric ones.

When you exit a program normally, all open I/O streams are flushed and temporary files removed. Basically it’s a nice exit where everything gets cleaned up and handled. It’s what you want to do almost all the time unless you have reasons to do otherwise.

28.1.1 Returning From `main()`

If you’ve noticed, main() has a return type of int… and yet I’ve rarely, if ever, been returning anything from main() at all.

This is because for main() only (and I can’t stress enough this special case only applies to main() and no other functions anywhere) has an implicit return 0 if you fall off the end.

You can explicitly return from main() any time you want, and some programmers feel it’s more Right to always have a return at the end of main(). But if you leave it off, C will put one there for you.

So… here are the return rules for main():

You can return an exit status from main() with a return statement. main() is the only function with this special behavior. Using return in any other function just returns from that function to the caller.
If you don’t explicitly return and just fall off the end of main(), it’s just as if you’d returned 0 or EXIT_SUCCESS.

28.1.2 `exit()`

This one has also made an appearance a few times. If you call exit() from anywhere in your program, it will exit at that point.

The argument you pass to exit() is the exit status.

28.1.3 Setting Up Exit Handlers with `atexit()`

You can register functions to be called when a program exits whether by returning from main() or calling the exit() function.

A call to atexit() with the handler function name will get it done. You can register multiple exit handlers, and they’ll be called in the reverse order of registration.

Here’s an example:

#include <stdio.h>
#include <stdlib.h>

void on_exit_1(void)
{
    printf("Exit handler 1 called!\n");
}

void on_exit_2(void)
{
    printf("Exit handler 2 called!\n");
}

int main(void)
{
    atexit(on_exit_1);
    atexit(on_exit_2);
    
    printf("About to exit...\n");
}

And the output is:

    About to exit...
    Exit handler 2 called!
    Exit handler 1 called!

28.2 Quicker Exits with `quick_exit()`

This is similar to a normal exit, except:

Open files might not be flushed.
Temporary files might not be removed.
atexit() handlers won’t be called.

But there is a way to register exit handlers: call at_quick_exit() analogously to how you’d call atexit().

#include <stdio.h>
#include <stdlib.h>

void on_quick_exit_1(void)
{
    printf("Quick exit handler 1 called!\n");
}

void on_quick_exit_2(void)
{
    printf("Quick exit handler 2 called!\n");
}

void on_exit(void)
{
    printf("Normal exit--I won't be called!\n");
}

int main(void)
{
    at_quick_exit(on_quick_exit_1);
    at_quick_exit(on_quick_exit_2);

    atexit(on_exit);  // This won't be called

    printf("About to quick exit...\n");

    quick_exit(0);
}

Which gives this output:

    About to quick exit...
    Quick exit handler 2 called!
    Quick exit handler 1 called!

It works just like exit()/atexit(), except for the fact that file flushing and cleanup might not be done.

28.3 Nuke it from Orbit: `_Exit()`

Calling _Exit() exits immediately, period. No on-exit callback functions are executed. Files won’t be flushed. Temp files won’t be removed.

Use this if you have to exit right fargin’ now.

28.4 Exiting Sometimes: `assert()`

The assert() statement is used to insist that something be true, or else the program will exit.

Devs often use an assert to catch Should-Never-Happen type errors.

    #define PI 3.14159
    
    assert(PI > 3);   // Sure enough, it is, so carry on

versus:

    goats -= 100;
    
    assert(goats >= 0);  // Can't have negative goats

In that case, if I try to run it and goats falls under 0, this happens:

    goat_counter: goat_counter.c:8: main: Assertion `goats >= 0' failed.
    Aborted

and I’m dropped back to the command line.

This isn’t very user-friendly, so it’s only used for things the user will never see. And often people write their own assert macros that can more easily be turned off.

28.5 Abnormal Exit: `abort()`

You can use this if something has gone horribly wrong and you want to indicate as much to the outside environment. This also won’t necessarily clean up any open files, etc.

I’ve rarely seen this used.

Some foreshadowing about signals: this actually works by raising a SIGABRT which will end the process.

What happens after that is up to the system, but on Unix-likes, it was common to dump core ¹⁴⁴ as the program terminated.

29 Signal Handling

Before we start, I’m just going to advise you to generally ignore this entire chapter and use your OS’s (very likely) superior signal handling functions. Unix-likes have the sigaction() family of functions, and Windows has… whatever it does¹⁴⁵.

With that out of the way, what are signals?

29.1 What Are Signals?

A signal is raised on a variety of external events. Your program can be configured to be interrupted to handle the signal, and, optionally, continue where it left off once the signal has been handled.

Think of it like a function that’s automatically called when one of these external events occurs.

What are these events? On your system, there are probably a lot of them, but in the C spec there are just a few:

Signal	Description
`SIGABRT`	Abnormal termination—what happens when `abort()` is called.
`SIGFPE`	Floating point exception.
`SIGILL`	Illegal instruction.
`SIGINT`	Interrupt—usually the result of `CTRL-C` being hit.
`SIGSEGV`	“Segmentation Violation”: invalid memory access.
`SIGTERM`	Termination requested.

You can set up your program to ignore, handle, or allow the default action for each of these by using the signal() function.

29.2 Handling Signals with `signal()`

The signal() call takes two parameters: the signal in question, and an action to take when that signal is raised.

The action can be one of three things:

A pointer to a handler function.
SIG_IGN to ignore the signal.
SIG_DFL to restore the default handler for the signal.

Let’s write a program that you can’t CTRL-C out of. (Don’t fret—in the following program, you can also hit RETURN and it’ll exit.)

#include <stdio.h>
#include <signal.h>

int main(void)
{
    char s[1024];

    signal(SIGINT, SIG_IGN);    // Ignore SIGINT, caused by ^C

    printf("Try hitting ^C...\n");

    // Wait for a line of input so the program doesn't just exit
    fgets(s, sizeof s, stdin);
}

Check out line 8—we tell the program to ignore SIGINT, the interrupt signal that’s raised when CTRL-C is hit. No matter how much you hit it, the signal remains ignored. If you comment out line 8, you’ll see you can CTRL-C with impunity and quit the program on the spot.

29.3 Writing Signal Handlers

I mentioned you could also write a handler function that gets called with the signal is raised.

These are pretty straightforward, are also very capability-limited when it comes to the spec.

Before we start, let’s look at the function prototype for the signal() call:

    void (*signal(int sig, void (*func)(int)))(int);

Pretty easy to read, right?

WRONG! :)

Let’s take a moment to take it apart for practice.

signal() takes two arguments: an integer sig representing the signal, and a pointer func to the handler (the handler returns void and takes an int as an argument), highlighted below:

                    sig          func
                  |-----|  |---------------|
    void (*signal(int sig, void (*func)(int)))(int);

Basically, we’re going to pass in the signal number we’re interesting in catching, and we’re going to pass a pointer to a function of the form:

    void f(int x);

that will do the actual catching.

Now—what about the rest of that prototype? It’s basically all the return type. See, signal() will return whatever you passed as func on success… so that means it’s returning a pointer to a function that returns void and takes an int as an argument.

    returned
    function    indicates we're              and
    returns     returning a                  that function
    void        pointer to function          takes an int
    |--|        |                                   |---|
    void       (*signal(int sig, void (*func)(int)))(int);

Also, it can return SIG_ERR in case of an error.

Let’s do an example where we make it so you have to hit CTRL-C twice to exit.

I want to be clear that this program engages in undefined behavior in a couple ways. But it’ll probably work for you, and it’s hard to come up with portable non-trivial demos.

#include <stdio.h>
#include <stdlib.h>
#include <signal.h>

int count = 0;

void sigint_handler(int signum)
{
    // The compiler is allowed to run:
    //
    //   signal(signum, SIG_DFL)
    //
    // when the handler is first called. So we reset the handler here:
    signal(SIGINT, sigint_handler);

    (void)signum;   // Get rid of unused variable warning

    count++;                       // Undefined behavior
    printf("Count: %d\n", count);  // Undefined behavior

    if (count == 2) {
        printf("Exiting!\n");      // Undefined behavior
        exit(0);
    }
}

int main(void)
{
    signal(SIGINT, sigint_handler);

    printf("Try hitting ^C...\n");

    for(;;);  // Wait here forever
}

One of the things you’ll notice is that on line 14 we reset the signal handler. This is because C has the option of resetting the signal handler to its SIG_DFL behavior before running your custom handler. In other words, it could be a one-off. So we reset it first thing so that we handle it again for the next one.

We’re ignoring the return value from signal() in this case. If we’d set it to a different handler earlier, it would return a pointer to that handler, which we could get like this:

    // old_handler is type "pointer to function that takes a single
    // int parameter and returns void":
    
    void (*old_handler)(int);
    
    old_handler = signal(SIGINT, sigint_handler);

That said, I’m not sure of a common use case for this. But if you need the old handler for some reason, you can get it that way.

Quick note on line 16—that’s just to tell the compiler to not warn that we’re not using this variable. It’s like saying, “I know I’m not using it; you don’t have to warn me.”

And lastly you’ll see that I’ve marked undefined behavior in a couple places. More on that in the next section.

29.4 What Can We Actually Do?

Turns out we’re pretty limited in what we can and can’t do in our signal handlers. This is one of the reasons why I say you shouldn’t even bother with this and instead use your OS’s signal handling instead (e.g. sigaction() for Unix-like systems).

Wikipedia goes so far as to say the only really portable thing you can do is call signal() with SIG_IGN or SIG_DFL and that’s it.

Here’s what we can’t portably do:

Call any standard library function.
- Like printf(), for example.
- I think it’s probably safe to call restartable/reentrant functions, but the spec doesn’t allow that liberty.
Get or set values from a local static, file scope, or thread-local variable.
- Unless it’s a lock-free atomic object or…
- You’re assigning into a variable of type volatile sig_atomic_t.

That last bit–sig_atomic_t–is your ticket to getting data out of a signal handler. (Unless you want to use lock-free atomic objects, which is outside the scope of this section¹⁴⁶.) It’s an integer type that might or might not be signed. And it’s bounded by what you can put in there.

You can look at the minimum and maximum allowable values in the macros SIG_ATOMIC_MIN and SIG_ATOMIC_MAX¹⁴⁷.

Confusingly, the spec also says you can’t refer “to any object with static or thread storage duration that is not a lock-free atomic object other than by assigning a value to an object declared as volatile sig_atomic_t […]”

My read on this is that you can’t read or write anything that’s not a lock-free atomic object. Also you can assign to an object that’s volatile sig_atomic_t.

But can you read from it? I honestly don’t see why not, except that the spec is very pointed about mentioning assigning into. But if you have to read it and make any kind of decision based on it, you might be opening up room for some kind of race conditions.

With that in mind, we can rewrite our “hit CTRL-C twice to exit” code to be a little more portable, albeit less verbose on the output.

Let’s change our SIGINT handler to do nothing except increment a value that’s of type volatile sig_atomic_t. So it’ll count the number of CTRL-Cs that have been hit.

Then in our main loop, we’ll check to see if that counter is over 2, then bail out if it is.

#include <stdio.h>
#include <signal.h>

volatile sig_atomic_t count = 0;

void sigint_handler(int signum)
{
    (void)signum;                    // Unused variable warning

    signal(SIGINT, sigint_handler);  // Reset signal handler

    count++;                         // Undefined behavior
}

int main(void)
{
    signal(SIGINT, sigint_handler);

    printf("Hit ^C twice to exit.\n");

    while(count < 2);
}

Undefined behavior again? It’s my read that this is, because we have to read the value in order to increment and store it. We can do some ridiculous contortions so that we’re only assigning into the values and manage to avoid undefined behavior.

#include <stdio.h>
#include <signal.h>

void sigint_handler_2(int signum)
{
    (void)signum;                      // Unused variable warning
    signal(SIGINT, SIG_DFL);           // Reset signal handler
}

void sigint_handler_1(int signum)
{
    (void)signum;                      // Unused variable warning
    signal(SIGINT, sigint_handler_2);  // Set to second handler
}

int main(void)
{
    signal(SIGINT, sigint_handler_1);

    printf("Hit ^C twice to exit.\n");

    while(1);
}

That’s pretty ugly, all right. Later when we look at lock-free atomic variables, we’ll see a way to fix the count version (assuming lock-free atomic variables are available on your particular system), but we’re getting into zanyland here.

This is why at the beginning, I was suggesting checking out your OS’s built-in signal system as a probably-superior alternative.

29.5 Friends Don’t Let Friends `signal()`

Again, use your OS’s built-in signal handling or the equivalent. It’s not in the spec, not as portable, but probably is far more capable. Plus your OS probably has a number of signals defined that aren’t in the C spec. And it’s difficult to write portable code using signal() anyway.

30 Variable-Length Arrays (VLAs)

C provides a way for you to declare an array whose size is determined at runtime. This gives you the benefits of dynamic runtime sizing like you get with malloc(), but without needing to worry about free()ing the memory after.

Now, a lot of people don’t like VLAs. They’ve been banned from the Linux kernel, for example. We’ll dig into more of that rationale later.

This is an optional feature of the language. The macro __STDC_NO_VLA__ is set to 1 if VLAs are not present. (They were mandatory in C99, and then became optional in C11.)

    #if __STDC_NO_VLA__ == 1
       #error Sorry, need VLAs for this program!
    #endif

Let’s dive in first with an example, and then we’ll look for the devil in the details.

30.1 The Basics

A normal array is declared with a constant size, like this:

    int v[10];

But with VLAs, we can use a size determined at runtime to set the array, like this:

    int n = 10;
    int v[n];

Now, that looks like the same thing, and in many ways is, but this gives you the flexibility to compute the size you need, and then get an array of exactly that size.

Let’s ask the user to input the size of the array, and then store the index-times-10 in each of those array elements:

#include <stdio.h>

int main(void)
{
    int n;

    printf("Enter a number: "); fflush(stdout);
    scanf(" %d", &n);

    int v[n];

    for (int i = 0; i < n; i++)
        v[i] = i * 10;

    for (int i = 0; i < n; i++)
        printf("v[%d] = %d\n", i, v[i]);
}

(On line 7, I have an fflush() that should force the line to output even though I don’t have a newline at the end.)

Line 10 is where we declare the VLA—once execution gets past that line, the size of the array is set to whatever n was at that moment. The array length can’t be changed later.

You can put an expression in the brackets, as well:

    int v[x * 100];

Some restrictions:

You can’t declare a VLA at file scope, and you can’t make a static one in block scope¹⁴⁸.
You can’t use an initializer list to initialize the array.

Also, entering a negative value for the size of the array invokes undefined behavior—in this universe, anyway.

30.2 `sizeof` and VLAs

We’re used to sizeof giving us the size in bytes of any particular object, including arrays. And VLAs are no exception.

The main difference is that sizeof on a VLA is executed at runtime, whereas on a non-variably-sized variable it is computed at compile time.

But the usage is the same.

You can even compute the number of elements in a VLA with the usual array trick:

    size_t num_elems = sizeof v / sizeof v[0];

There’s a subtle and correct implication from the above line: pointer arithmetic works just like you’d expect for a regular array. So go ahead and use it to your heart’s content:

#include <stdio.h>

int main(void)
{
    int n = 5;
    int v[n];

    int *p = v;

    *(p+2) = 12;
    printf("%d\n", v[2]);  // 12

    p[3] = 34;
    printf("%d\n", v[3]);  // 34
}

30.3 Multidimensional VLAs

You can go ahead and make all kinds of VLAs with one or more dimensions set to a variable

    int w = 10;
    int h = 20;
    
    int x[h][w];
    int y[5][w];
    int z[10][w][20];

Again, you can navigate these just like you would a regular array.

30.4 Passing One-Dimensional VLAs to Functions

Passing single-dimensional VLAs into a function can be no different than passing a regular array in. You just go for it.

#include <stdio.h>

int sum(int count, int *v)
{
    int total = 0;

    for (int i = 0; i < count; i++)
        total += v[i];

    return total;
}

int main(void)
{
    int x[5];   // Standard array

    int a = 5;
    int y[a];   // VLA

    for (int i = 0; i < a; i++)
        x[i] = y[i] = i + 1;

    printf("%d\n", sum(5, x));
    printf("%d\n", sum(a, y));
}

But there’s a bit more to it than that. You can also let C know that the array is a specific VLA size by passing that in first and then giving that dimension in the parameter list:

    int sum(int count, int v[count])
    {
        // ...
    }

Incidentally, there are a couple ways of listing a prototype for the above function; one of them involves an * if you don’t want to specifically name the value in the VLA. It just indicates that the type is a VLA as opposed to a regular pointer.

VLA prototypes:

    void do_something(int count, int v[count]);  // With names
    void do_something(int, int v[*]);            // Without names

Again, that * thing only works with the prototype—in the function itself, you’ll have to put the explicit size.

Now—let’s get multidimensional! This is where the fun begins.

30.5 Passing Multi-Dimensional VLAs to Functions

Same thing as we did with the second form of one-dimensional VLAs, above, but this time we’re passing in two dimensions and using those.

In the following example, we build a multiplication table matrix of a variable width and height, and then pass it into a function to print it out.

#include <stdio.h>

void print_matrix(int h, int w, int m[h][w])
{
    for (int row = 0; row < h; row++) {
        for (int col = 0; col < w; col++)
            printf("%2d ", m[row][col]);
        printf("\n");
    }
}

int main(void)
{
    int rows = 4;
    int cols = 7;

    int matrix[rows][cols];

    for (int row = 0; row < rows; row++)
        for (int col = 0; col < cols; col++)
            matrix[row][col] = row * col;

    print_matrix(rows, cols, matrix);
}

30.5.1 Partial Multidimensional VLAs

You can have some of the dimensions fixed and some variable. Let’s say we have a record length fixed at 5 elements, but we don’t know how many records there are.

#include <stdio.h>

void print_records(int count, int record[count][5])
{
    for (int i = 0; i < count; i++) {
        for (int j = 0; j < 5; j++)
            printf("%2d ", record[i][j]);
        printf("\n");
    }
}

int main(void)
{
    int rec_count = 3;
    int records[rec_count][5];

    // Fill with some dummy data
    for (int i = 0; i < rec_count; i++)
        for (int j = 0; j < 5; j++)
            records[i][j] = (i+1)*(j+2);

    print_records(rec_count, records);
}

30.6 Compatibility with Regular Arrays

Because VLAs are just like regular arrays in memory, it’s perfectly permissible to pass them interchangeably… as long as the dimensions match.

For example, if we have a function that specifically wants a $3\times5$ array, we can still pass a VLA into it.

    int foo(int m[5][3]) {...}
    
    \\ ...
    
    int w = 3, h = 5;
    int matrix[h][w];
    
    foo(matrix);   // OK!

Likewise, if you have a VLA function, you can pass a regular array into it:

    int foo(int h, int w, int m[h][w]) {...}
    
    \\ ...
    
    int matrix[3][5];
    
    foo(3, 5, matrix);   // OK!

Beware, through: if your dimensions mismatch, you’re going to have some undefined behavior going on, likely.

30.7 `typedef` and VLAs

You can typedef a VLA, but the behavior might not be as you expect.

Basically, typedef makes a new type with the values as they existed the moment the typedef was executed.

So it’s not a typedef of a VLA so much as a new fixed size array type of the dimensions at the time.

#include <stdio.h>

int main(void)
{
    int w = 10;

    typedef int goat[w];

    // goat is an array of 10 ints
    goat x;

    // Init with squares of numbers
    for (int i = 0; i < w; i++)
        x[i] = i*i;

    // Print them
    for (int i = 0; i < w; i++)
        printf("%d\n", x[i]);

    // Now let's change w...

    w = 20;

    // But goat is STILL an array of 10 ints, because that was the
    // value of w when the typedef executed.
}

So it acts like an array of fixed size.

But you still can’t use an initializer list on it.

30.8 Jumping Pitfalls

You have to watch out when using goto near VLAs because a lot of things aren’t legal.

And when you’re using longjmp() there’s a case where you could leak memory with VLAs.

But both of these things we’ll cover in their respective chapters.

30.9 General Issues

VLAs have been banned from the Linux kernel due for a few reasons:

Lots of places they were used should have just been fixed-size.
The code behind VLAs is slower (to a degree that most people wouldn’t notice, but makes a difference in an operating system).
VLAs are not supported to the same degree by all C compilers.
Stack size is limited, and VLAs go on the stack. If some code accidentally (or maliciously) passes a large value into a kernel function that allocates a VLA, Bad Things™ could happen.

Other folks online point out that there’s no way to detect a VLA’s failure to allocate, and programs that suffered such problems would likely just crash. While fixed-size arrays also have the same issue, it’s far more likely that someone accidentally make a VLA Of Unusual Size than somehow accidentally declare a fixed-size, say, 30 megabyte array.

31 `goto`

The goto statement is universally revered and can be here presented without contest.

Just kidding! Over the years, there has been a lot of back-and-forth over whether or not (often not) goto is considered harmful ¹⁴⁹.

In this programmer’s opinion, you should use whichever constructs leads to the best code, factoring in maintainability and speed. And sometimes this might be goto!

In this chapter, we’ll see how goto works in C, and then check out some of the common cases where it is used¹⁵⁰.

31.1 A Simple Example

In this example, we’re going to use goto to skip a line of code and jump to a label. The label is the identifier that can be a goto target—it ends with a colon (:).

#include <stdio.h>

int main(void)
{
    printf("One\n");
    printf("Two\n");

    goto skip_3;

    printf("Three\n");

skip_3:

    printf("Five!\n");
}

The output is:

    One
    Two
    Five!

goto sends execution jumping to the specified label, skipping everything in between.

You can jump forward or backward with goto.

    infinite_loop:
        print("Hello, world!\n");
        goto infinite_loop;

Labels are skipped over during execution. The following will print all three numbers in order just as if the labels weren’t there:

        printf("Zero\n");
    label_1:
    label_2:
        printf("One\n");
    label 3:
        printf("Two\n");
    label 4:
        printf("Three\n");

As you’ve noticed, it’s common convention to justify the labels all the way on the left. This increases readability because a reader can quickly scan to find the destination.

Labels have function scope. That is, no matter how many levels deep in blocks they appear, you can still goto them from anywhere in the function.

It also means you can only goto labels that are in the same function as the goto itself. Labels in other functions are out of scope from goto’s perspective. And it means you can use the same label name in two functions—just not the same label name in the same function.

31.2 Labeled `continue`

In some languages, you can actually specify a label for a continue statement. C doesn’t allow it, but you can easily use goto instead.

To show the issue, check out continue in this nested loop:

    for (int i = 0; i < 3; i++) {
        for (int j = 0; j < 3; j++) {
            printf("%d, %d\n", i, j);
            continue;   // Always goes to next j
        }
    }

As we see, that continue, like all continues, goes to the next iteration of the nearest enclosing loop. What if we want to continue in the next loop out, the loop with i?

Well, we can break to get back to the outer loop, right?

    for (int i = 0; i < 3; i++) {
        for (int j = 0; j < 3; j++) {
            printf("%d, %d\n", i, j);
            break;     // Gets us to the next iteration of i
        }
    }

That gets us two levels of nested loop. But then if we nest another loop, we’re out of options. What about this, where we don’t have any statement that will get us out to the next iteration of i?

    for (int i = 0; i < 3; i++) {
        for (int j = 0; j < 3; j++) {
            for (int k = 0; k < 3; k++) {
                printf("%d, %d, %d\n", i, j, k);
    
                continue;  // Gets us to the next iteration of k
                break;     // Gets us to the next iteration of j
                ????;      // Gets us to the next iteration of i???
    
            }
        }
    }

The goto statement offers us a way!

        for (int i = 0; i < 3; i++) {
            for (int j = 0; j < 3; j++) {
                for (int k = 0; k < 3; k++) {
                    printf("%d, %d, %d\n", i, j, k);
    
                    goto continue_i;   // Now continuing the i loop!!
                }
            }
    continue_i: ;
        }

We have a ; at the end there—that’s because you can’t have a label pointing to the plain end of a compound statement (or before a variable declaration).

31.3 Bailing Out

When you’re super nested in the middle of some code, you can use goto to get out of it in a manner that’s often cleaner than nesting more ifs and using flag variables.

        // Pseudocode
    
        for(...) {
            for (...) {
                while (...) {
                    do {
                        if (some_error_condition)
                            goto bail;
    
                    } while(...);
                }
            }
        }
    
    bail:
        // Cleanup here

Without goto, you’d have to check an error condition flag in all of the loops to get all the way out.

31.4 Labeled `break`

This is a very similar situation to how continue only continues the innermost loop. break also only breaks out of the innermost loop.

        for (int i = 0; i < 3; i++) {
            for (int j = 0; j < 3; j++) {
                printf("%d, %d\n", i, j);
                break;   // Only breaks out of the j loop
            }
        }
    
        printf("Done!\n");

But we can use goto to break farther:

        for (int i = 0; i < 3; i++) {
            for (int j = 0; j < 3; j++) {
                printf("%d, %d\n", i, j);
                goto break_i;   // Now breaking out of the i loop!
            }
        }
    
    break_i:
    
        printf("Done!\n");

31.5 Multi-level Cleanup

If you’re calling multiple functions to initialize multiple systems and one of them fails, you should only de-initialize the ones that you’ve gotten to so far.

Let’s do a fake example where we start initializing systems and checking to see if any returns an error (we’ll use -1 to indicate an error). If one of them does, we have to shutdown only the systems we’ve initialized so far.

        if (init_system_1() == -1)
            goto shutdown;
    
        if (init_system_2() == -1)
            goto shutdown_1;
    
        if (init_system_3() == -1)
            goto shutdown_2;
    
        if (init_system_4() == -1)
            goto shutdown_3;
    
        do_main_thing();   // Run our program
    
        shutdown_system4();
    
    shutdown_3:
        shutdown_system3();
    
    shutdown_2:
        shutdown_system2();
    
    shutdown_1:
        shutdown_system1();
    
    shutdown:
        print("All subsystems shut down.\n");

Note that we’re shutting down in the reverse order that we initialized the subsystems. So if subsystem 4 fails to start up, it will shut down 3, 2, then 1 in that order.

31.6 Restarting Interrupted System Calls

This is outside the spec, but commonly seen in Unix-like systems.

Certain long-lived system calls might return an error if they’re interrupted by a signal, and errno will be set to EINTR to indicate the syscall was doing fine; it was just interrupted.

In those cases, it’s really common for the programmer to want to restart the call and try it again.

    retry:
        byte_count = read(0, buf, sizeof(buf) - 1);  // Unix read() syscall
    
        if (byte_count == -1) {            // An error occurred...
            if (errno == EINTR) {          // But it was just interrupted
                printf("Restarting...\n");
                goto retry;
            }

Many Unix-likes have an SA_RESTART flag you can pass to sigaction() to request the OS automatically restart any slow syscalls instead of failing with EINTR.

Again, this is Unix-specific and is outside the C standard.

That said, it’s possible to use a similar technique any time any function should be restarted.

31.7 `goto` and Variable Scope

We’ve already seen that labels have function scope, but weird things can happen if we jump past some variable initialization.

Look at this example where we jump from a place where the variable x is out of scope into the middle of its scope (in the block).

        goto label;
    
        {
            int x = 12345;
    
    label:
            printf("%d\n", x);
        }

This will compile and run, but gives me a warning:

    warning: ‘x’ is used uninitialized in this function

And then it prints out 0 when I run it (your mileage may vary).

Basically what has happened is that we jumped into x’s scope (so it was OK to reference it in the printf()) but we jumped over the line that actually initialized it to 12345. So the value was indeterminate.

The fix is, of course, to get the initialization after the label one way or another.

        goto label;
    
        {
            int x;
    
    label:
            x = 12345;
            printf("%d\n", x);
        }

Let’s look at one more example.

        {
            int x = 10;
    
    label:
    
            printf("%d\n", x);
        }
    
        goto label;

What happens here?

The first time through the block, we’re good. x is 10 and that’s what prints.

But after the goto, we’re jumping into the scope of x, but past its initialization. Which means we can still print it, but the value is indeterminate (since it hasn’t been reinitialized).

On my machine, it prints 10 again (to infinity), but that’s just luck. It could print any value after the goto since x is uninitialized.

31.8 `goto` and Variable-Length Arrays

When it comes to VLAs and goto, there’s one rule: you can’t jump from outside the scope of a VLA into the scope of that VLA.

If I try to do this:

        int x = 10;
    
        goto label;
    
        {
            int v[x];
    
    label:
    
            printf("Hi!\n");
        }

I get an error:

    error: jump into scope of identifier with variably modified type

You can jump in ahead of the VLA declaration, like this:

        int x = 10;
    
        goto label;
    
        {
    label:  ;
            int v[x];
    
            printf("Hi!\n");
        }

Because that way the VLA gets allocated properly before its inevitable deallocation once it falls out of scope.

32 Types Part V: Compound Literals and Generic Selections

This is the final chapter for types! We’re going to talk about two things:

How to have “anonymous” unnamed objects and how that’s useful.
How to generate type-dependent code.

They’re not particularly related, but don’t really each warrant their own chapters. So I crammed them in here like a rebel!

32.1 Compound Literals

This is a neat feature of the language that allows you to create an object of some type on the fly without ever assigning it to a variable. You can make simple types, arrays, structs, you name it.

One of the main uses for this is passing complex arguments to functions when you don’t want to make a temporary variable to hold the value.

The way you create a compound literal is to put the type name in parentheses, and then put an initializer list after. For example, an unnamed array of ints, might look like this:

    (int []){1,2,3,4}

Now, that line of code doesn’t do anything on its own. It creates an unnamed array of 4 ints, and then throws them away without using them.

We could use a pointer to store a reference to the array…

    int *p = (int []){1 ,2 ,3 ,4};
    
    printf("%d\n", p[1]);  // 2

But that seems a little like a long-winded way to have an array. I mean, we could have just done this¹⁵¹:

    int p[] = {1, 2, 3, 4};
    
    printf("%d\n", p[1]);  // 2

So let’s take a look at a more useful example.

32.1.1 Passing Unnamed Objects to Functions

Let’s say we have a function to sum an array of ints:

    int sum(int p[], int count)
    {
        int total = 0;
    
        for (int i = 0; i < count; i++)
            total += p[i];
    
        return total;
    }

If we wanted to call it, we’d normally have to do something like this, declaring an array and storing values in it to pass to the function:

    int a[] = {1, 2, 3, 4};
    
    int s = sum(a, 4);

But unnamed objects give us a way to skip the variable by passing it directly in (parameter names listed above). Check it out—we’re going to replace the variable a with an unnamed array that we pass in as the second argument:

    //                   p[]         count
    //           |-----------------|  |
    int s = sum((int []){1, 2, 3, 4}, 4));

Pretty slick!

32.1.2 Unnamed `struct`s

We can do something similar with structs.

First, let’s do things without unnamed objects. We’ll define a struct to hold some x/y coordinates. Then we’ll define one, passing in values into its initializer. Finally, we’ll pass it to a function to print the values:

#include <stdio.h>

struct coord {
    int x, y;
};

void print_coord(struct coord c)
{
    printf("%d, %d\n", c.x, c.y);
}

int main(void)
{
    struct coord t = {.x=10, .y=20};

    print_coord(t);   // prints "10, 20"
}

Straightforward enough?

Let’s modify it to use an unnamed object instead of the variable t we’re passing to print_coord().

We’ll just take t out of there and replace it with an unnamed struct:

    //struct coord t = {.x=10, .y=20};

    print_coord((struct coord){.x=10, .y=20});   // prints "10, 20"

Still works!

32.1.3 Pointers to Unnamed Objects

You might have noticed in the last example that even through we were using a struct, we were passing a copy of the struct to print_coord() as opposed to passing a pointer to the struct.

Turns out, we can just take the address of an unnamed object with & like always.

This is because, in general, if an operator would have worked on a variable of that type, you can use that operator on an unnamed object of that type.

Let’s modify the above code so that we pass a pointer to an unnamed object

#include <stdio.h>

struct coord {
    int x, y;
};

void print_coord(struct coord *c)
{
    printf("%d, %d\n", c->x, c->y);
}

int main(void)
{
    //     Note the &
    //          |
    print_coord(&(struct coord){.x=10, .y=20});   // prints "10, 20"
}

Additionally, this can be a nice way to pass even pointers to simple objects:

    // Pass a pointer to an int with value 3490
    foo(&(int){3490});

Easy as that.

32.1.4 Unnamed Objects and Scope

The lifetime of an unnamed object ends at the end of its scope. The biggest way this could bite you is if you make a new unnamed object, get a pointer to it, and then leave the object’s scope. In that case, the pointer will refer to a dead object.

So this is undefined behavior:

    int *p;
    
    {
        p = &(int){10};
    }
    
    printf("%d\n", *p);  // INVALID: The (int){10} fell out of scope

Likewise, you can’t return a pointer to an unnamed object from a function. The object is deallocated when it falls out of scope:

#include <stdio.h>

int *get3490(void)
{
    // Don't do this
    return &(int){3490};
}

int main(void)
{
    printf("%d\n", *get3490());  // INVALID: (int){3490} fell out of scope
}

Just think of their scope like that of an ordinary local variable. You can’t return a pointer to a local variable, either.

32.1.5 Silly Unnamed Object Example

You can put any type in there and make an unnamed object.

For example, these are effectively equivalent:

    int x = 3490;
    
    printf("%d\n", x);               // 3490 (variable)
    printf("%d\n", 3490);            // 3490 (constant)
    printf("%d\n", (int){3490});     // 3490 (unnamed object)

That last one is unnamed, but it’s silly. Might as well do the simple one on the line before.

But hopefully that provides a little more clarity on the syntax.

32.2 Generic Selections

This is an expression that allows you select different pieces of code depending on the type of the first argument to the expression.

We’ll look at an example in just a second, but it’s important to know this is processed at compile time, not at runtime. There’s no runtime analysis going on here.

The expression begins with _Generic, works kinda like a switch, and it takes at least two arguments.

The first argument is an expression (or variable¹⁵²) that has a type. All expressions have a type. The remaining arguments to _Generic are the cases of what to substitute in for the result of the expression if the first argument is that type.

Wat?

Let’s try it out and see.

#include <stdio.h>

int main(void)
{
    int i;
    float f;
    char c;

    char *s = _Generic(i,
                    int: "that variable is an int",
                    float: "that variable is a float",
                    default: "that variable is some type"
                );

    printf("%s\n", s);
}

Check out the _Generic expression starting on line 9.

When the compiler sees it, it look at the type of the first argument. (In this example, the type of the variable i.) It then looks through the cases for something of that type. And then it substitutes the argument in place of the entire _Generic expression.

In this case, i is an int, so it matches that case. Then the string is substituted in for the expression. So the line turns into this when the compiler sees it:

        char *s = "that variable is an int";

If the compiler can’t find a type match in the _Generic, it looks for the optional default case and uses that.

If it can’t find a type match and there’s no default, you’ll get a compile error. The first expression must match one of the types or default.

Because it’s inconvenient to write _Generic over and over, it’s often used to make the body of a macro that can be easily repeatedly reused.

Let’s make a macro TYPESTR(x) that takes an argument and returns a string with the type of the argument.

So TYPESTR(1) will return the string "int", for example.

Here we go:

    #include <stdio.h>
    
    #define TYPESTR(x) _Generic((x), \
                            int: "int", \
                            long: "long", \
                            float: "float", \
                            double: "double", \
                            default: "something else")
    
    int main(void)
    {
        int i;
        long l;
        float f;
        double d;
        char c;
    
        printf("i is type %s\n", TYPESTR(i));
        printf("l is type %s\n", TYPESTR(l));
        printf("f is type %s\n", TYPESTR(f));
        printf("d is type %s\n", TYPESTR(d));
        printf("c is type %s\n", TYPESTR(c));
    }

This outputs:

    i is type int
    l is type long
    f is type float
    d is type double
    c is type something else

Which should be no surprise, because, like we said, that code in main() is replaced with the following when it is compiled:

        printf("i is type %s\n", "int");
        printf("l is type %s\n", "long");
        printf("f is type %s\n", "float");
        printf("d is type %s\n", "double");
        printf("c is type %s\n", "something else");

And that’s exactly the output we see.

Let’s do one more. I’ve included some macros here so that when you run:

    int i = 10;
    char *s = "Foo!";
    
    PRINT_VAL(i);
    PRINT_VAL(s);

you get the output:

    i = 10
    s = Foo!

We’ll have to make use of some macro magic to do that.

#include <stdio.h>
#include <string.h>

// Macro that gives back a format specifier for a type
#define FMTSPEC(x) _Generic((x), \
                        int: "%d", \
                        long: "%ld", \
                        float: "%f", \
                        double: "%f", \
                        char *: "%s")
                        // TODO: add more types
                        
// Macro that prints a variable in the form "name = value"
#define PRINT_VAL(x) { \
    char fmt[512]; \
    snprintf(fmt, sizeof fmt, #x " = %s\n", FMTSPEC(x)); \
    printf(fmt, (x)); \
}

int main(void)
{
    int i = 10;
    float f = 3.14159;
    char *s = "Hello, world!";

    PRINT_VAL(i);
    PRINT_VAL(f);
    PRINT_VAL(s);
}

for the output:

    i = 10
    f = 3.141590
    s = Hello, world!

We could have crammed that all in one big macro, but I broke it into two to prevent eye bleeding.

33 Arrays Part II

We’re going to go over a few extra misc things this chapter concerning arrays.

Type qualifiers with array parameters
The static keyword with array parameters
Partial multi-dimensional array initializers

They’re not super-commonly seen, but we’ll peek at them since they’re part of the newer spec.

33.1 Type Qualifiers for Arrays in Parameter Lists

If you recall from earlier, these two things are equivalent in function parameter lists:

    int func(int *p) {...}
    int func(int p[]) {...}

And you might also recall that you can add type qualifiers to a pointer variable like so:

    int *const p;
    int *volatile p;
    int *const volatile p;
    // etc.

But how can we do that when we’re using array notation in your parameter list?

Turns out it goes in the brackets. And you can put the optional count after. The two following lines are equivalent:

    int func(int *const volatile p) {...}
    int func(int p[const volatile]) {...}
    int func(int p[const volatile 10]) {...}

If you have a multidimensional array, you need to put the type qualifiers in the first set of brackets.

33.2 `static` for Arrays in Parameter Lists

Similarly, you can use the keyword static in the array in a parameter list.

This is something I’ve never seen in the wild. It is always followed by a dimension:

    int func(int p[static 4]) {...}

What this means, in the above example, is the compiler is going to assume that any array you pass to the function will be at least 4 elements.

Anything else is undefined behavior.

    int func(int p[static 4]) {...}
    
    int main(void)
    {
        int a[] = {11, 22, 33, 44};
        int b[] = {11, 22, 33, 44, 55};
        int c[] = {11, 22};
    
        func(a);  // OK! a is 4 elements, the minimum
        func(b);  // OK! b is at least 4 elements
        func(c);  // Undefined behavior! c is under 4 elements!
    }

This basically sets the minimum size array you can have.

Important note: there is nothing in the compiler that prohibits you from passing in a smaller array. The compiler probably won’t warn you, and it won’t detect it at runtime.

By putting static in there, you’re saying, “I double secret PROMISE that I will never pass in a smaller array than this.” And the compiler says, “Yeah, fine,” and trusts you to not do it.

And then the compiler can make certain code optimizations, safe in the knowledge that you, the programmer, will always do the right thing.

33.3 Equivalent Initializers

C is a little bit, shall we say, flexible when it comes to array initializers.

We’ve already seen some of this, where any missing values are replaced with zero.

For example, we can initialize a 5 element array to 1,2,0,0,0 with this:

    int a[5] = {1, 2};

Or set an array entirely to zero with:

    int a[5] = {0};

But things get interesting when initializing multidimensional arrays.

Let’s make an array of 3 rows and 2 columns:

    int a[3][2];

Let’s write some code to initialize it and print the result:

    #include <stdio.h>
    
    int main(void)
    {
        int a[3][2] = {
            {1, 2},
            {3, 4},
            {5, 6}
        };
    
        for (int row = 0; row < 3; row++) {
            for (int col = 0; col < 2; col++)
                printf("%d ", a[row][col]);
            printf("\n");
        }
    }

And when we run it, we get the expected:

    1 2
    3 4
    5 6

Let’s leave off some of the initializer elements and see they get set to zero:

        int a[3][2] = {
            {1, 2},
            {3},    // Left off the 4!
            {5, 6}
        };

which produces:

    1 2
    3 0
    5 6

Now let’s leave off the entire last middle element:

        int a[3][2] = {
            {1, 2},
            // {3, 4},   // Just cut this whole thing out
            {5, 6}
        };

And now we get this, which might not be what you expect:

    1 2
    5 6
    0 0

But if you stop to think about it, we only provided enough initializers for two rows, so they got used for the first two rows. And the remaining elements were initialized to zero.

So far so good. Generally, if we leave off parts of the initializer, the compiler sets the corresponding elements to 0.

But let’s get crazy.

        int a[3][2] = { 1, 2, 3, 4, 5, 6 };

What—? That’s a 2D array, but it only has a 1D initializer!

Turns out that’s legal (though GCC will warn about it with the proper warnings turned on).

Basically, what it does is starts filling in elements in row 0, then row 1, then row 2 from left to right.

So when we print, it prints in order:

    1 2
    3 4
    5 6

If we leave some off:

        int a[3][2] = { 1, 2, 3 };

they fill with 0:

    1 2
    3 0
    0 0

So if you want to fill the whole array with 0, then go ahead and:

        int a[3][2] = {0};

But my recommendation is if you have a 2D array, use a 2D initializer. It just makes the code more readable. (Except for initializing the whole array with 0, in which case it’s idiomatic to use {0} no matter the dimension of the array.)

34 Long Jumps with `setjmp`, `longjmp`

We’ve already seen goto, which jumps in function scope. But longjmp() allows you to jump back to an earlier point in execution, back to a function that called this one.

There are a lot of limitations and caveats, but this can be a useful function for bailing out from deep in the call stack back up to an earlier state.

In my experience, this is very rarely-used functionality.

34.1 Using `setjmp` and `longjmp`

The dance we’re going to do here is to basically put a bookmark in execution with setjmp(). Later on, we’ll call longjmp() and it’ll jump back to the earlier point in execution where we set the bookmark with setjmp().

And it can do this even if you’ve called subfunctions.

Here’s a quick demo where we call into functions a couple levels deep and then bail out of it.

We’re going to use a file scope variable env to keep the state of things when we call setjmp() so we can restore them when we call longjmp() later. This is the variable in which we remember our “place”.

The variable env is of type jmp_buf, an opaque type declared in <setjmp.h>.

#include <stdio.h>
#include <setjmp.h>

jmp_buf env;

void depth2(void)
{
    printf("Entering depth 2\n");
    longjmp(env, 3490);           // Bail out
    printf("Leaving depth 2\n");  // This won't happen
}

void depth1(void)
{
    printf("Entering depth 1\n");
    depth2();
    printf("Leaving depth 1\n");  // This won't happen
}

int main(void)
{
    switch (setjmp(env)) {
    case 0:
        printf("Calling into functions, setjmp() returned 0\n");
        depth1();
        printf("Returned from functions\n");  // This won't happen
        break;

    case 3490:
        printf("Bailed back to main, setjmp() returned 3490\n");
        break;
    }
}

When run, this outputs:

    Calling into functions, setjmp() returned 0
    Entering depth 1
    Entering depth 2
    Bailed back to main, setjmp() returned 3490

If you try to take that output and match it up with the code, it’s clear there’s some really funky stuff going on.

One of the most notable things is that setjmp() returns twice. What the actual frank? What is this sorcery?!

So here’s the deal: if setjmp() returns 0, it means that you’ve successfully set the “bookmark” at that point.

If it returns non-zero, it means you’ve just returned to the “bookmark” set earlier. (And the value returned is the one you pass into longjmp().)

This way you can tell the difference between setting the bookmark and returning to it later.

So when the code, above, calls setjmp() the first time, setjmp() stores the state in the env variable and returns 0. Later when we call longjmp() with that same env, it restores the state and setjmp() returns the value longjmp() was passed.

34.2 Pitfalls

Under the hood, this is pretty straightforward. Typically the stack pointer keeps track of the locations in memory that local variables are stored, and the program counter keeps track of the address of the currently-executing instruction¹⁵³.

So if we want to jump back to an earlier function, it’s basically only a matter of restoring the stack pointer and program counter to the values kept in the jmp_buf variable, and making sure the return value is set correctly. And then execution will resume there.

But a variety of factors confound this, making a significant number of undefined behavior traps.

34.2.1 The Values of Local Variables

If you want the values of automatic (non-static and non-extern) local variables to persist in the function that called setjmp() after a longjmp() happens, you must declare those variables to be volatile.

Technically, they only have to be volatile if they change between the time setjmp() is called and longjmp() is called¹⁵⁴.

For example, if we run this code:

    int x = 20;
    
    if (setjmp(env) == 0) {
        x = 30;
    }

and then later longjmp() back, the value of x will be indeterminate.

If we want to fix this, x must be volatile:

    volatile int x = 20;
    
    if (setjmp(env) == 0) {
        x = 30;
    }

Now the value will be the correct 30 after a longjmp() returns us to this point.

34.2.2 How Much State is Saved?

When you longjmp(), execution resumes at the point of the corresponding setjmp(). And that’s it.

The spec points out that it’s just as if you’d jumped back into the function at that point with local variables set to whatever values they had when the longjmp() call was made.

Things that aren’t restored include, paraphrasing the spec:

Floating point status flags
Open files
Any other component of the abstract machine (including values in local variables when setjmp() was called)

34.2.3 You Can’t Name Anything `setjmp`

You can’t have any extern identifiers with the name setjmp. Or, if setjmp is a macro, you can’t undefine it.

Both are undefined behavior.

34.2.4 You Can’t `setjmp()` in a Larger Expression

That is, you can’t do something like this:

    if (x == 12 && setjmp(env) == 0) { ... }

That’s too complex to be allowed by the spec due to the machinations that must occur when unrolling the stack and all that. We can’t longjmp() back into some complex expression that’s only been partially executed.

So there are limits on the complexity of that expression.

It can be the entire controlling expression of the conditional.

    if (setjmp(env)) {...}

    switch (setjmp(env)) {...}

It can be part of a relational or equality expression, as long as the other operand is an integer constant. And the whole thing is the controlling expression of the conditional.
```
    if (setjmp(env) == 0) {...}
```
The operand to a logical NOT (!) operation, being the entire controlling expression.
```
    if (!setjmp(env)) {...}
```
A standalone expression, possibly cast to void.
```
    setjmp(env);
```
```
    (void)setjmp(env);
```

34.2.5 When Can’t You `longjmp()`?

It’s undefined behavior if:

You didn’t call setjmp() earlier
You called setjmp() from another thread
You called setjmp() in the scope of a variable length array (VLA), and execution left the scope of that VLA before longjmp() was called.
The function containing the setjmp() exited before longjmp() was called.

On that last one, “exited” includes normal returns from the function, as well as the case if another longjmp() jumped back to “earlier” in the call stack than the function in question.

34.2.6 You Can’t Pass `0` to `longjmp()`

If you try to pass the value 0 to longjmp(), it will silently change that value to 1.

Since setjmp() ultimately returns this value, and having setjmp() return 0 has special meaning, returning 0 is prohibited.

34.2.7 `longjmp() and Variable Length Arrays

If you are in scope of a VLA and longjmp() out there, the memory allocated to the VAL could leak¹⁵⁵.

Same thing happens if you longjmp() back over any earlier functions that had VLAs still in scope.

This is one thing that really bugged me able VLAs—that you could write perfectly legitimate C code that squandered memory. But, hey—I’m not in charge of the spec.

35 Incomplete Types

It might surprise you to learn that this builds without error:

extern int a[];

int main(void)
{
    struct foo *x;
    union bar *y;
    enum baz *z;
}

We never gave a size for a. And we have pointers to structs foo, bar, and baz that never seem to be declared anywhere.

And the only warnings I get are that x, y, and z are unused.

These are examples of incomplete types.

An incomplete type is a type the size (i.e. the size you’d get back from sizeof) for which is not known. Another way to think of it is a type that you haven’t finished declaring.

You can have a pointer to an incomplete type, but you can’t dereference it or use pointer arithmetic on it. And you can’t sizeof it.

So what can you do with it?

35.1 Use Case: Self-Referential Structures

I only know of one real use case: forward references to structs or unions with self-referential or co-dependent structures. (I’m going to use struct for the rest of these examples, but they all apply equally to unions, as well.)

Let’s do the classic example first.

But before I do, know this! As you declare a struct, the struct is incomplete until the closing brace is reached!

    struct antelope {              // struct antelope is incomplete here
        int leg_count;             // Still incomplete
        float stomach_fullness;    // Still incomplete
        float top_speed;           // Still incomplete
        char *nickname;            // Still incomplete
    };                             // NOW it's complete.

So what? Seems sane enough.

But what if we’re doing a linked list? Each linked list node needs to have a reference to another node. But how can we create a reference to another node if we haven’t finished even declaring the node yet?

C’s allowance for incomplete types makes it possible. We can’t declare a node, but we can declare a pointer to one, even if it’s incomplete!

struct node {
    int val;
    struct node *next;  // struct node is incomplete, but that's OK!
};

Even though the struct node is incomplete on line 3, we can still declare a pointer to one¹⁵⁶.

We can do the same thing if we have two different structs that refer to each other:

    struct a {
        struct b *x;  // Refers to a `struct b`
    };
    
    struct b {
        struct a *x;  // Refers to a `struct a`
    };

We’d never be able to make that pair of structures without the relaxed rules for incomplete types.

35.2 Incomplete Type Error Messages

Are you getting errors like these?

    invalid application of ‘sizeof’ to incomplete type
    
    invalid use of undefined type
    
    dereferencing pointer to incomplete type

Most likely culprit: you probably forgot to #include the header file that declares the type.

35.3 Other Incomplete Types

Declaring a struct or union with no body makes an incomplete type, e.g. struct foo;.

enums are incomplete until the closing brace.

void is an incomplete type.

Arrays declared extern with no size are incomplete, e.g.:

    extern int a[];

If it’s a non-extern array with no size followed by an initializer, it’s incomplete until the closing brace of the initializer.

35.4 Use Case: Arrays in Header Files

It can be useful to declare incomplete array types in header files. In those cases, the actual storage (where the complete array is declared) should be in a single .c file. If you put it in the .h file, it will be duplicated every time the header file is included.

So what you can do is make a header file with an incomplete type that refers to the array, like so:

// File: bar.h

#ifndef BAR_H
#define BAR_H

extern int my_array[];  // Incomplete type

#endif

And the in the .c file, actually define the array:

// File: bar.c

int my_array[1024];     // Complete type!

Then you can include the header from as many places as you’d like, and every one of those places will refer to the same underlying my_array.

// File: foo.c

#include <stdio.h>
#include "bar.h"    // includes the incomplete type for my_array

int main(void)
{
    my_array[0] = 10;

    printf("%d\n", my_array[0]);
}

When compiling multiple files, remember to specific all the .c files to the compiler, but not the .h files, e.g.:

    gcc -o foo foo.c bar.c

35.5 Completing Incomplete Types

If you have an incomplete type, you can complete it by defining the complete struct, union, enum, or array in the same scope.

    struct foo;        // incomplete type
    
    struct foo *p;     // pointer, no problem
    
    // struct foo f;   // Error: incomplete type!
    
    struct foo {
        int x, y, z;
    };                 // Now the struct foo is complete!
    
    struct foo f;      // Success!

Note that though void is an incomplete type, there’s no way to complete it. Not that anyone ever thinks of doing that weird thing. But it does explain why you can do this:

    void *p;             // OK: pointer to incomplete type

and not either of these:

    void v;              // Error: declare variable of incomplete type
    
    printf("%d\n", *p);  // Error: dereference incomplete type

The more you know…

36 Complex Numbers

A tiny primer on Complex numbers ¹⁵⁷ stolen directly from Wikipedia:

A complex number is a number that can be expressed in the form $a+bi$, where $a$ and $b$ are real numbers [i.e. floating point types in C], and $i$ represents the imaginary unit, satisfying the equation $i^2=−1$. Because no real number satisfies this equation, $i$ is called an imaginary number. For the complex number $a+bi$, $a$ is called the real part, and $b$ is called the imaginary part.

But that’s as far as I’m going to go. We’ll assume that if you’re reading this chapter, you know what a complex number is and what you want to do with them.

And all we need to cover is C’s faculties for doing so.

Turns out, though, that complex number support in a compiler is an optional feature. Not all compliant compilers can do it. And the ones that do, might do it to various degrees of completeness.

You can test if your system supports complex numbers with:

    #ifdef __STDC_NO_COMPLEX__
    #error Complex numbers not supported!
    #endif

Furthermore, there is a macro that indicates adherence to the ISO 60559 (IEEE 754) standard for floating point math with complex numbers, as well as the presence of the _Imaginary type.

    #if __STDC_IEC_559_COMPLEX__ != 1
    #error Need IEC 60559 complex support!
    #endif

More details on that are spelled out in Annex G in the C11 spec.

36.1 Complex Types

To use complex numbers, #include <complex.h>.

With that, you get at least two types:

    _Complex
    complex

Those both mean the same thing, so you might as well use the prettier complex.

You also get some types for imaginary numbers if you implementation is IEC 60559-compliant:

    _Imaginary
    imaginary

These also both mean the same thing, so you might as well use the prettier imaginary.

You also get values for the imaginary number $i$, itself:

    I
    _Complex_I
    _Imaginary_I

The macro I is set to _Imaginary_I (if available), or _Complex_I. So just use I for the imaginary number.

One aside: I’ve said that if a compiler has __STDC_IEC_559_COMPLEX__ set to 1, it must support _Imaginary types to be compliant. That’s my read of the spec. However, I don’t know of a single compiler that actually supports _Imaginary even though they have __STDC_IEC_559_COMPLEX__ set. So I’m going to write some code with that type in here I have no way of testing. Sorry!

OK, so now we know there’s a complex type, how can we use it?

36.2 Assigning Complex Numbers

Since the complex number has a real and imaginary part, but both of them rely on floating point numbers to store values, we need to also tell C what precision to use for those parts of the complex number.

We do that by just pinning a float, double, or long double to the complex, either before or after it.

Let’s define a complex number that uses float for its components:

    float complex c;   // Spec prefers this way
    complex float c;   // Same thing--order doesn't matter

So that’s great for declarations, but how do we initialize them or assign to them?

Turns out we get to use some pretty natural notation. Example!

    double complex x = 5 + 2*I;
    double complex y = 10 + 3*I;

For $5+2i$ and $10+3i$, respectively.

36.3 Constructing, Deconstructing, and Printing

We’re getting there…

We’ve already seen one way to write a complex number:

    double complex x = 5 + 2*I;

There’s also no problem using other floating point numbers to build it:

    double a = 5;
    double b = 2;
    double complex x = a + b*I;

There is also a set of macros to help build these. The above code could be written using the CMPLX() macro, like so:

    double complex x = CMPLX(5, 2);

As far as I can tell in my research, these are almost equivalent:

    double complex x = 5 + 2*I;
    double complex x = CMPLX(5, 2);

But the CMPLX() macro will handle negative zeros in the imaginary part correctly every time, whereas the other way might convert them to positive zeros. I think¹⁵⁸ This seems to imply that if there’s a chance the imaginary part will be zero, you should use the macro… but someone should correct me on this if I’m mistaken!

The CMPLX() macro works on double types. There are two other macros for float and long double: CMPLXF() and CMPLXL(). (These “f” and “l” suffixes appear in virtually all the complex-number-related functions.)

Now let’s try the reverse: if we have a complex number, how do we break it apart into its real and imaginary parts?

Here we have a couple functions that will extract the real and imaginary parts from the number: creal() and cimag():

    double complex x = 5 + 2*I;
    double complex y = 10 + 3*I;
    
    printf("x = %f + %fi\n", creal(x), cimag(x));
    printf("y = %f + %fi\n", creal(y), cimag(y));

for the output:

    x = 5.000000 + 2.000000i
    y = 10.000000 + 3.000000i

Note that the i I have in the printf() format string is a literal i that gets printed—it’s not part of the format specifier. Both return values from creal() and cimag() are double.

And as usual, there are float and long double variants of these functions: crealf(), cimagf(), creall(), and cimagl().

36.4 Complex Arithmetic and Comparisons

Arithmetic can be performed on complex numbers, though how this works mathematically is beyond the scope of the guide.

#include <stdio.h>
#include <complex.h>

int main(void)
{
    double complex x = 1 + 2*I;
    double complex y = 3 + 4*I;
    double complex z;

    z = x + y;
    printf("x + y = %f + %fi\n", creal(z), cimag(z));

    z = x - y;
    printf("x - y = %f + %fi\n", creal(z), cimag(z));

    z = x * y;
    printf("x * y = %f + %fi\n", creal(z), cimag(z));

    z = x / y;
    printf("x / y = %f + %fi\n", creal(z), cimag(z));
}

for a result of:

    x + y = 4.000000 + 6.000000i
    x - y = -2.000000 + -2.000000i
    x * y = -5.000000 + 10.000000i
    x / y = 0.440000 + 0.080000i

You can also compare two complex numbers for equality (or inequality):

#include <stdio.h>
#include <complex.h>

int main(void)
{
    double complex x = 1 + 2*I;
    double complex y = 3 + 4*I;

    printf("x == y = %d\n", x == y);  // 0
    printf("x != y = %d\n", x != y);  // 1
}

with the output:

    x == y = 0
    x != y = 1

They are equal if both components test equal. Note that as with all floating point, they could be equal if they’re close enough due to rounding error¹⁵⁹.

36.5 Complex Math

But wait! There’s more than just simple complex arithmetic!

Here’s a summary table of all the math functions available to you with complex numbers.

I’m only going to list the double version of each function, but for all of them there is a float version that you can get by appending f to the function name, and a long double version that you can get by appending l.

For example, the cabs() function for computing the absolute value of a complex number also has cabsf() and cabsl() variants. I’m omitting them for brevity.

36.5.1 Trigonometry Functions

Function	Description
`ccos()`	Cosine
`csin()`	Sine
`ctan()`	Tangent
`cacos()`	Arc cosine
`casin()`	Arc sine
`catan()`	Play Settlers of Catan
`ccosh()`	Hyperbolic cosine
`csinh()`	Hyperbolic sine
`ctanh()`	Hyperbolic tangent
`cacosh()`	Arc hyperbolic cosine
`casinh()`	Arc hyperbolic sine
`catanh()`	Arc hyperbolic tangent

36.5.2 Exponential and Logarithmic Functions

Function	Description
`cexp()`	Base-$e$ exponential
`clog()`	Natural (base-$e$) logarithm

36.5.3 Power and Absolute Value Functions

Function	Description
`cabs()`	Absolute value
`cpow()`	Power
`csqrt()`	Square root

36.5.4 Manipulation Functions

Function	Description
`creal()`	Return real part
`cimag()`	Return imaginary part
`CMPLX()`	Construct a complex number
`carg()`	Argument/phase angle
`conj()`	Conjugate¹⁶⁰
`cproj()`	Projection on Riemann sphere

37 Fixed Width Integer Types

C has all those small, bigger, and biggest integer types like int and long and all that. And you can look in the section on limits to see what the largest int is with INT_MAX and so on.

How big are those types? That is, how many bytes do they take up? We could use sizeof to get that answer.

But what if I wanted to go the other way? What if I needed a type that was exactly 32 bits (4 bytes) or at least 16 bits or somesuch?

How can we declare a type that’s a certain size?

The header <stdint>.h gives us a way.

37.1 The Bit-Sized Types

For both signed and unsigned integers, we can specify a type that is a certain number of bits, with some caveats, of course.

And there are three main classes of these types (in these examples, the N would be replaced by a certain number of bits):

Integers of exactly a certain size (intN_t)
Integers that are at least a certain size (int_leastN_t)
Integers that are at least a certain size and are as fast as possible (int_fastN_t)¹⁶¹

How much faster is fast? Definitely maybe some amount faster. Probably. The spec doesn’t say how much faster, just that they’ll be the fastest on this architecture. Most C compilers are pretty good, though, so you’ll probably only see this used in places where the most possible speed needs to be guaranteed (rather than just hoping the compiler is producing pretty-dang-fast code, which it is).

Finally, these unsigned number types have a leading u to differentiate them.

For example, these types have the corresponding listed meaning:

    int32_t w;        // x is exactly 32 bits, signed
    uint16_t x;       // y is exactly 16 bits, unsigned
    
    int_least8_t y;   // y is at least 8 bits, signed
    
    uint_fast64_t z;  // z is the fastest representation at least 64 bits, unsigned

The following types are guaranteed to be defined:

    int_least8_t      uint_least8_t
    int_least16_t     uint_least16_t
    int_least32_t     uint_least32_t
    int_least64_t     uint_least64_t
    
    int_fast8_t       uint_fast8_t
    int_fast16_t      uint_fast16_t
    int_fast32_t      uint_fast32_t
    int_fast64_t      uint_fast64_t

There might be others of different widths, as well, but those are optional.

Hey! Where are the fixed types like int16_t? Turns out those are entirely optional…unless certain conditions are met¹⁶². And if you have an average run-of-the-mill modern computer system, those conditions probably are met. And if they are, you’ll have these types:

    int8_t      uint8_t
    int16_t     uint16_t
    int32_t     uint32_t
    int64_t     uint64_t

Other variants with different widths might be defined, but they’re optional.

37.2 Maximum Integer Size Type

There’s a type you can use that holds the largest representable integers available on the system, both signed and unsigned:

    intmax_t
    uintmax_t

Use these types when you want to go as big as possible.

Obviously values from any other integer types of the same sign will fit in this type, necessarily.

37.3 Using Fixed Size Constants

If you have a constant that you want to have fit in a certain number of bits, you can use these macros to automatically append the proper suffix onto the number (e.g. 22L or 3490ULL).

    INT8_C(x)     UINT8_C(x)
    INT16_C(x)    UINT16_C(x)
    INT32_C(x)    UINT32_C(x)
    INT64_C(x)    UINT64_C(x)
    INTMAX_C(x)   UINTMAX_C(x)

Again, these work only with constant integer values.

For example, we can use one of these to assign constant values like so:

    uint16_t x = UINT16_C(12);
    intmax_t y = INTMAX_C(3490);

37.4 Limits of Fixed Size Integers

We also have some limits defined so you can get the maximum and minimum values for these types:

    INT8_MAX           INT8_MIN           UINT8_MAX
    INT16_MAX          INT16_MIN          UINT16_MAX
    INT32_MAX          INT32_MIN          UINT32_MAX
    INT64_MAX          INT64_MIN          UINT64_MAX
    
    INT_LEAST8_MAX     INT_LEAST8_MIN     UINT_LEAST8_MAX
    INT_LEAST16_MAX    INT_LEAST16_MIN    UINT_LEAST16_MAX
    INT_LEAST32_MAX    INT_LEAST32_MIN    UINT_LEAST32_MAX
    INT_LEAST64_MAX    INT_LEAST64_MIN    UINT_LEAST64_MAX
    
    INT_FAST8_MAX      INT_FAST8_MIN      UINT_FAST8_MAX
    INT_FAST16_MAX     INT_FAST16_MIN     UINT_FAST16_MAX
    INT_FAST32_MAX     INT_FAST32_MIN     UINT_FAST32_MAX
    INT_FAST64_MAX     INT_FAST64_MIN     UINT_FAST64_MAX
    
    INTMAX_MAX         INTMAX_MIN         UINTMAX_MAX

Note the MIN for all the unsigned types is 0, so, as such, there’s no macro for it.

37.5 Format Specifiers

In order to print these types, you need to send the right format specifier to printf(). (And the same issue for getting input with scanf().)

But how are you going to know what size the types are under the hood? Luckily, once again, C provides some macros to help with this.

All this can be found in <inttypes.h>.

Now, we have a bunch of macros. Like a complexity explosion of macros. So I’m going to stop listing out every one and just put the lowercase letter n in the place where you should put 8, 16, 32, or 64 depending on your needs.

Let’s look at the macros for printing signed integers:

    PRIdn    PRIdLEASTn    PRIdFASTn    PRIdMAX
    PRIin    PRIiLEASTn    PRIiFASTn    PRIiMAX

Look for the patterns there. You can see there are variants for the fixed, least, fast, and max types.

And you also have a lowercase d and a lowercase i. Those correspond to the printf() format specifiers %d and %i.

So if I have something of type:

    int_least16_t x = 3490;

I can print that with the equivalent format specifier for %d by using PRId16.

But how? How do we use that macro?

First of all, that macro specifies a string containing the letter or letters printf() needs to use to print that type. Like, for example, it could be "d" or "ld".

So all we need to do is embed that in our format string to the printf() call.

To do this, we can take advantage of a fact about C that you might have forgotten: adjacent string literals are automatically concatenated to a single string. E.g.:

    printf("Hello, " "world!\n");   // Prints "Hello, world!"

And since these macros are string literals, we can use them like so:

#include <stdio.h>
#include <stdint.h>
#include <inttypes.h>

int main(void)
{
    int_least16_t x = 3490;

    printf("The value is %" PRIdLEAST16 "!\n", x);
}

We also have a pile of macros for printing unsigned types:

    PRIon    PRIoLEASTn    PRIoFASTn    PRIoMAX
    PRIun    PRIuLEASTn    PRIuFASTn    PRIuMAX
    PRIxn    PRIxLEASTn    PRIxFASTn    PRIxMAX
    PRIXn    PRIXLEASTn    PRIXFASTn    PRIXMAX

In this case, o, u, x, and X correspond to the documented format specifiers in printf().

And, as before, the lowercase n should be substituted with 8, 16, 32, or 64.

But just when you think you had enough of the macros, it turns out we have a complete complementary set of them for scanf()!

    SCNdn    SCNdLEASTn    SCNdFASTn    SCNdMAX
    SCNin    SCNiLEASTn    SCNiFASTn    SCNiMAX
    SCNon    SCNoLEASTn    SCNoFASTn    SCNoMAX
    SCNun    SCNuLEASTn    SCNuFASTn    SCNuMAX
    SCNxn    SCNxLEASTn    SCNxFASTn    SCNxMAX

Remember: when you want to print out a fixed size integer type with printf() or scanf(), grab the correct corresponding format specifer from <inttypes.h>.

38 Date and Time Functionality

“Time is an illusion. Lunchtime doubly so.” —Ford Prefect, The Hitchhikers Guide to the Galaxy

This isn’t too complex, but it can be a little intimidating at first, both with the different types available and the way we can convert between them.

Mix in GMT (UTC) and local time and we have all the Usual Fun™ one gets with times and dates.

And of course never forget the golden rule of dates and times: Never attempt to write your own date and time functionality. Only use what the library gives you.

Time is too complex for mere mortal programmers to handle correctly. Seriously, we all owe a point to everyone who worked on any date and time library, so put that in your budget.

38.1 Quick Terminology and Information

Just a couple quick terms in case you don’t have them down.

UTC: Coordinated Universal Time is a universally¹⁶³ agreed upon, absolute time. Everyone on the planet thinks it’s the same time right now in UTC… even though they have different local times.
GMT: Greenwich Mean Time, effectively the same as UTC¹⁶⁴. You probably want to say UTC, or “universal time”. If you’re talking specifically about the GMT timezone, say GMT. Confusingly, many of C’s UTC functions predate UTC and still refer to Greenwich Mean Time. When you see that, know the C means UTC.
Local time: what time it is where the computer running the program is located. This is described as an offset from UTC. Although there are many timezones in the world, most computers do work in either local time or UTC.

If you are describing an event that happens one time, like a log entry, or a rocket launch, or when pointers finally clicked for you, use UTC.

On the other hand, if it’s something that happens the same time in every timezone, like New Year’s Eve or dinner time, use local time.

Since a lot of languages are only good at converting between UTC and local time, you can cause yourself a lot of pain by choosing to store your dates in the wrong form. (Ask me how I know.)

38.2 Date Types

There are two¹⁶⁵ main types in C when it comes to dates: time_t and struct tm.

The spec doesn’t actually say much about them:

time_t: a real type capable of holding a time. So by the spec, this could be a floating type or integer type. In POSIX (Unix-likes), it’s an integer. This holds calendar time. Which you can think of as UTC time.
struct tm: holds the components of a calendar time. This is a broken-down time, i.e. the components of the time, like hour, minute, second, day, month, year, etc.

On a lot of systems, time_t represents the number of seconds since Epoch¹⁶⁶. Epoch is in some ways the start of time from the computer’s perspective, which is commonly January 1, 1970 UTC. time_t can go negative to represent times before Epoch. Windows behaves the same way as Unix from what I can tell.

And what’s in a struct tm? The following fields:

    struct tm {
        int tm_sec;    // seconds after the minute -- [0, 60]
        int tm_min;    // minutes after the hour -- [0, 59]
        int tm_hour;   // hours since midnight -- [0, 23]
        int tm_mday;   // day of the month -- [1, 31]
        int tm_mon;    // months since January -- [0, 11]
        int tm_year;   // years since 1900
        int tm_wday;   // days since Sunday -- [0, 6]
        int tm_yday;   // days since January 1 -- [0, 365]
        int tm_isdst;  // Daylight Saving Time flag
    };

Note that everything is zero-based except the day of the month.

It’s important to know that you can put any values in these types you want. There are functions to help get the time now, but the types hold a time, not the time.

So the question becomes: “How do you initialize data of these types, and how do you convert between them?”

38.3 Initialization and Conversion Between Types

First, you can get the current time and store it in a time_t with the time() function.

    time_t now;  // Variable to hold the time now
    
    now = time(NULL);  // You can get it like this...
    
    time(&now);        // ...or this. Same as the previous line.

Great! You have a variable that gets you the time now.

Amusingly, there’s only one portable way to print out what’s in a time_t, and that’s the rarely-used ctime() function that prints the value in local time:

    now = time(NULL);
    printf("%s", ctime(&now));

This returns a string with a very specific form that includes a newline at the end:

    Sun Feb 28 18:47:25 2021

So that’s kind of inflexible. If you want more control, you should convert that time_t into a struct tm.

38.3.1 Converting `time_t` to `struct tm`

There are two amazing ways to do this conversion:

localtime(): this function converts a time_t to a struct tm in local time.
gmtime(): this function converts a time_t to a struct tm in UTC. (See ye olde GMT creeping into that function name?)

Let’s see what time it is now by printing out a struct tm with the asctime() function:

    printf("Local: %s", asctime(localtime(&now)));
    printf("  UTC: %s", asctime(gmtime(&now)));

Output (I’m in Pacific Standard Time, out of daylight ):

    Local: Sun Feb 28 20:15:27 2021
      UTC: Mon Mar  1 04:15:27 2021

Once you have your time_t in a struct tm, it opens all kinds of doors. You can print out the time in a variety of ways, figure out which day of the week a date is, and so on. Or convert it back into a time_t.

38.3.2 Converting `struct tm` to `time_t`

If you want to go the other way, you can use mktime() to get that information.

mktime() sets the values of tm_wday and tm_yday for you, so don’t bother filling them out because they’ll just be overwritten.

Also, you can set tm_isdst to -1 to have it make the determination for you. Or you can manually set it to true or false.

    struct tm some_time = {
        .tm_year=82,   // years since 1900
        .tm_mon=3,     // months since January -- [0, 11]
        .tm_mday=12,   // day of the month -- [1, 31]
        .tm_hour=12,   // hours since midnight -- [0, 23]
        .tm_min=00,    // minutes after the hour -- [0, 59]
        .tm_sec=04,    // seconds after the minute -- [0, 60]
        .tm_isdst=-1,  // Daylight Saving Time flag
    };
    
    time_t some_time_epoch;
    
    some_time_epoch = mktime(&some_time);
    
    printf("%s", ctime(&some_time_epoch));
    printf("Is DST: %d\n", some_time.tm_isdst);

Output:

    Mon Apr 12 12:00:04 1982
    Is DST: 0

When you manually load a struct tm like that, it should be in local time. mktime() will convert that local time into a time_t calendar time.

Weirdly, however, the standard doesn’t give us a way to load up a struct tm with a UTC time and convert that to a time_t. If you want to do that with Unix-likes, try the non-standard timegm(). On Windows, _mkgmtime().

38.4 Formatted Date Output

We’ve already seen a couple ways to print formatted date output to the screen. With time_t we can use ctime(), and with struct tm we can use asctime().

    time_t now = time(NULL);
    struct tm *local = localtime(&now);
    struct tm *utc = gmtime(&now);
    
    printf("Local time: %s", ctime(&now));     // Local time with time_t
    printf("Local time: %s", asctime(local));  // Local time with struct tm
    printf("UTC       : %s", asctime(utc));    // UTC with a struct tm

But what if I told you, dear reader, that there’s a way to have much more control over how the date was printed?

Sure, we could fish individual fields out of the struct tm, but there’s a great function called strftime() that will do a lot of the hard work for you. It’s like printf(), except for dates!

Let’s see some examples. In each of these, we pass in a destination buffer, a maximum number of characters to write, and then a format string (in the style of—but not the same as—printf()) which tells strftime() which components of a struct tm to print and how.

You can add other constant characters to include in the output in the format string, as well, just like with printf().

We get a struct tm in this case from localtime(), but any source works fine.

#include <stdio.h>
#include <time.h>

int main(void)
{
    char s[128];
    time_t now = time(NULL);

    // %c: print date as per current locale
    strftime(s, sizeof s, "%c", localtime(&now));
    puts(s);   // Sun Feb 28 22:29:00 2021

    // %A: full weekday name
    // %B: full month name
    // %d: day of the month
    strftime(s, sizeof s, "%A, %B %d", localtime(&now));
    puts(s);   // Sunday, February 28

    // %I: hour (12 hour clock)
    // %M: minute
    // %S: second
    // %p: AM or PM
    strftime(s, sizeof s, "It's %I:%M:%S %p", localtime(&now));
    puts(s);   // It's 10:29:00 PM

    // %F: ISO 8601 yyyy-mm-dd
    // %T: ISO 8601 hh:mm:ss
    // %z: ISO 8601 timezone offset
    strftime(s, sizeof s, "ISO 8601: %FT%T%z", localtime(&now));
    puts(s);   // ISO 8601: 2021-02-28T22:29:00-0800
}

There are a ton of date printing format specifiers for strftime(), so be sure to check them out in the strftime() reference page.

38.5 More Resolution with `timespec_get()`

You can get the number of seconds and nanoseconds since Epoch with timespec_get().

Maybe.

Implementations might not have nanosecond resolution (that’s one billionth of a second) so who knows how many significant places you’ll get, but give it a shot and see.

timespec_get() takes two arguments. One is a pointer to a struct timespec to hold the time information. And the other is the base, which the spec lets you set to TIME_UTC indicating that you’re interested in seconds since Epoch. (Other implementations might give you more options for the base.)

And the structure itself has two fields:

    struct timespec {
        time_t tv_sec;   // Seconds
        long   tv_nsec;  // Nanoseconds (billionths of a second)
    };

Here’s an example where we get the time and print it out both as integer values and also a floating value:

    struct timespec ts;
    
    timespec_get(&ts, TIME_UTC);
    
    printf("%ld s, %ld ns\n", ts.tv_sec, ts.tv_nsec);
    
    double float_time = ts.tv_sec + ts.tv_nsec/1000000000.0;
    printf("%f seconds since epoch\n", float_time);

Example output:

    1614581530 s, 806325800 ns
    1614581530.806326 seconds since epoch

struct timespec also makes an appearance in a number of the threading functions that need to be able to specify time with that resolution.

38.6 Differences Between Times

One quick note about getting the difference between two time_ts: since the spec doesn’t dictate how that type represents a time, you might not be able to simply subtract two time_ts and get anything sensible¹⁶⁷.

Luckily you can use difftime() to compute the difference in seconds between two dates.

In the following example, we have two events that occur some time apart, and we use difftime() to compute the difference.

    #include <stdio.h>
    #include <time.h>
    
    int main(void)
    {
        struct tm time_a = {
            .tm_year=82,   // years since 1900
            .tm_mon=3,     // months since January -- [0, 11]
            .tm_mday=12,   // day of the month -- [1, 31]
            .tm_hour=4,    // hours since midnight -- [0, 23]
            .tm_min=00,    // minutes after the hour -- [0, 59]
            .tm_sec=04,    // seconds after the minute -- [0, 60]
            .tm_isdst=-1,  // Daylight Saving Time flag
        };
    
        struct tm time_b = {
            .tm_year=120,  // years since 1900
            .tm_mon=10,    // months since January -- [0, 11]
            .tm_mday=15,   // day of the month -- [1, 31]
            .tm_hour=16,   // hours since midnight -- [0, 23]
            .tm_min=27,    // minutes after the hour -- [0, 59]
            .tm_sec=00,    // seconds after the minute -- [0, 60]
            .tm_isdst=-1,  // Daylight Saving Time flag
        };
    
        time_t cal_a = mktime(&time_a);
        time_t cal_b = mktime(&time_b);
    
        double diff = difftime(cal_b, cal_a);
    
        double years = diff / 60 / 60 / 24 / 365.2425;  // close enough
    
        printf("%f seconds (%f years) between events\n", diff, years);
    }

Output:

    1217996816.000000 seconds (38.596783 years) between events

And there you have it! Remember to use difftime() to take the time difference. Even though you can just subtract on a POSIX system, might as well stay portable.

39 Multithreading

C11 introduced, formally, multithreading to the C language. It’s very eerily similar to POSIX threads ¹⁶⁸, if you’ve ever used those.

And if you’re not, no worries. We’ll talk it through.

Do note, however, that I’m not intending this to be a full-blown classic multithreading how-to¹⁶⁹; you’ll have to pick up a different very thick book for that, specifically. Sorry!

Threading is an optional feature. If a C11+ compiler defines __STDC_NO_THREADS__, threads will not be present in the library. Why they decided to go with a negative sense in that macro is beyond me, but there we are.

You can test for it like this:

    #ifdef __STDC_NO_THREADS__
    #error I need threads to build this program!
    #endif

Also, you might need to specify certain linker options when building. In the case of Unix-likes, try appending a -lpthreads to the end of the command line to link the pthreads library¹⁷⁰:

    gcc -std=c11 -o foo foo.c -lpthreads

If you’re getting linker errors on your system, it could be because the appropriate library wasn’t included.

39.1 Background

Threads are a way to have all those shiny CPU cores you paid for do work for you in the same program.

Normally, a C program just runs on a single CPU core. But if you know how to split up the work, you can give pieces of it to a number of threads and have them do the work simultaneously.

Though the spec doesn’t say it, on your system it’s very likely that C (or the OS at its behest) will attempt to balance the threads over all your CPU cores.

And if you have more threads than cores, that’s OK. You just won’t realize all those gains if they’re all trying to compete for CPU time.

39.2 Things You Can Do

You can create a thread. It will begin running the function you specify. The parent thread that spawned it will also continue to run.

And you can wait for the thread to complete. This is called joining.

Or if you don’t care when the thread completes and don’t want to wait, you can detach it.

A thread can explicitly exit, or it can implicitly call it quits by returning from it’s main function.

A thread can also sleep for a period of time, doing nothing while other threads run.

The main() program is a thread, as well.

Additionally, we have thread local storage, mutexes, and conditional variables. But more on those later. Let’s just look at the basics for now.

39.3 Data Races and the Standard Library

Some of the functions in the standard library (e.g. asctime() and strtok()) return or use static data elements that aren’t threadsafe. But in general unless it’s said otherwise, the standard library makes an effort to be so¹⁷¹

But keep an eye out. If a standard library function is maintaining state between calls in a variable you don’t own, or if a function is returning a pointer to a thing that you didn’t pass in, it’s not threadsafe.

39.4 Creating and Waiting for Threads

Let’s hack something up!

We’ll make some threads (create) and wait for them to complete (join).

We have a tiny bit to understand first, though.

Every single thread is identified by an opaque variable of type thrd_t. It’s a unique identifier per thread in your program. When you create a thread, it’s given a new ID.

Also when you make the thread, you have to give it a pointer to a function to run, and a pointer to an argument to pass to it (or NULL if you don’t have anything to pass).

The thread will begin execution on the function you specify.

When you want to wait for a thread to complete, you have to specify it’s thread ID so C knows which one to wait for.

So the basic idea is:

Write a function to act as the thread’s “main”. It’s not main()-proper, but analogous to it. The thread will start running there.
From the main thread, launch a new thread with thrd_create(), and pass it a pointer to the function to run.
In that function, have the thread do whatever it has to do.
Meantimes, the main thread can continue doing whatever it has to do.
When the main thread decides to, it can wait for the child thread to complete by calling thrd_join(). Generally you must thrd_join() the thread to clean up after it or else you’ll leak memory¹⁷²

thrd_create() takes a pointer to the function to run, and it’s of type thrd_start_t, which is int (*)(void *). That’s Greek for “a pointer to a function that takes an void* as an argument, and returns an int.”

Let’s make a thread! We’ll launch it from the main thread with thrd_create() to run a function, do some other things, then wait for it to complete with thrd_join(). I’ve named the thread’s main function run(), but you can name it anything as long as the types match thrd_start_t.

#include <stdio.h>
#include <threads.h>

// This is the function the thread will run. It can be called anything.
//
// arg is the argument pointer passed to `thrd_create()`.
//
// The parent thread will get the return value back from `thrd_join()`'
// later.

int run(void *arg)
{
    int *a = arg;  // We'll pass in an int* from thrd_create()

    printf("THREAD: Running thread with arg %d\n", *a);

    return 12;  // Value to be picked up by thrd_join() (chose 12 at random)
}

int main(void)
{
    thrd_t t;  // t will hold the thread ID
    int arg = 3490;

    printf("Launching a thread\n");

    // Launch a thread to the run() function, passing a pointer to 3490
    // as an argument. Also stored the thread ID in t:

    thrd_create(&t, run, &arg);

    printf("Doing other things while the thread runs\n");

    printf("Waiting for thread to complete...\n");

    int res;  // Holds return value from the thread exit

    // Wait here for the thread to complete; store the return value
    // in res:

    thrd_join(t, &res);

    printf("Thread exited with return value %d\n", res);
}

See how we did the thrd_create() there to call the run() function? Then we did other things in main() and then stopped and waited for the thread to complete with thrd_join().

Sample output (yours might vary):

    Launching a thread
    Doing other things while the thread runs
    Waiting for thread to complete...
    THREAD: Running thread with arg 3490
    Thread exited with return value 12

The arg that you pass to the function has to have a lifetime long enough so that the thread can pick it up before it goes away. Also, it needs to not be overwritten by the main thread before the new thread can use it.

Let’s look at an example that launches 5 threads. One thing to note here is how we use an array of thrd_ts to keep track of all the thread IDs.

#include <stdio.h>
#include <threads.h>

int run(void *arg)
{
    int i = *(int*)arg;

    printf("THREAD %d: running!\n", i);

    return i;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    int i;

    printf("Launching threads...\n");
    for (i = 0; i < THREAD_COUNT; i++)

        // NOTE! In the following line, we pass a pointer to i, 
        // but each thread sees the same pointer. So they'll
        // print out weird things as i changes value here in
        // the main thread! (More in the text, below.)

        thrd_create(t + i, run, &i);

    printf("Doing other things while the thread runs...\n");
    printf("Waiting for thread to complete...\n");

    for (int i = 0; i < THREAD_COUNT; i++) {
        int res;
        thrd_join(t[i], &res);

        printf("Thread %d complete!\n", res);
    }

    printf("All threads complete!\n");
}

When I run the threads, I count i up from 0 to 4. And pass a pointer to it to thrd_create(). This pointer ends up in the run() routine where we make a copy of it.

Simple enough? Here’s the output:

    Launching threads...
    THREAD 2: running!
    THREAD 3: running!
    THREAD 4: running!
    THREAD 2: running!
    Doing other things while the thread runs...
    Waiting for thread to complete...
    Thread 2 complete!
    Thread 2 complete!
    THREAD 5: running!
    Thread 3 complete!
    Thread 4 complete!
    Thread 5 complete!
    All threads complete!

Whaaa—? Where’s THREAD 0? And why do we have a THREAD 5 when clearly i is never more than 4 when we call thrd_create()? And two THREAD 2s? Madness!

This is getting into the fun land of race conditions. The main thread is modifying i before the thread has a chance to copy it. Indeed, i makes it all the way to 5 and ends the loop before the last thread gets a chance to copy it.

We’ve got to have a per-thread variable that we can refer to so we can pass it in as the arg.

We could have a big array of them. Or we could malloc() space (and free it somewhere—maybe in the thread itself.)

Let’s give that a shot:

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

int run(void *arg)
{
    int i = *(int*)arg;  // Copy the arg

    free(arg);  // Done with this

    printf("THREAD %d: running!\n", i);

    return i;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    int i;

    printf("Launching threads...\n");
    for (i = 0; i < THREAD_COUNT; i++) {

        // Get some space for a per-thread argument:

        int *arg = malloc(sizeof *arg);
        *arg = i;

        thrd_create(t + i, run, arg);
    }

    // ...

Notice on lines 27-30 we malloc() space for an int and copy the value of i into it. Each new thread gets its own freshly-malloc()d variable and we pass a pointer to that into the run() function.

Once run() makes its own copy of the arg on line 7, it free()s the malloc()d int. And now that it has its own copy, it can do with it what it pleases.

And a run shows the result:

    Launching threads...
    THREAD 0: running!
    THREAD 1: running!
    THREAD 2: running!
    THREAD 3: running!
    Doing other things while the thread runs...
    Waiting for thread to complete...
    Thread 0 complete!
    Thread 1 complete!
    Thread 2 complete!
    Thread 3 complete!
    THREAD 4: running!
    Thread 4 complete!
    All threads complete!

There we go! Threads 0-4 all in effect!

Your run might vary—how the threads get scheduled to run is beyond the C spec. We see in the above example that thread 4 didn’t even begin until threads 0-1 had completed. Indeed, if I run this again, I likely get different output. We cannot guarantee a thread execution order.

39.5 Detaching Threads

If you want to fire-and-forget a thread (i.e. so you don’t have to thrd_join() it later), you can do that with thrd_detach().

This removes the parent thread’s ability to get the return value from the child thread, but if you don’t care about that and just want threads to clean up nicely on their own, this is the way to go.

Basically we’re going to do this:

    thrd_create(&t, run, NULL);
    thrd_detach(t);

where the thrd_detach() call is the parent thread saying, “Hey, I’m not going to wait for this child thread to complete with thrd_join(). So go ahead and clean it up on your own when it completes.”

#include <stdio.h>
#include <threads.h>

int run(void *arg)
{
    (void)arg;

    //printf("Thread running! %lu\n", thrd_current()); // non-portable!
    printf("Thread running!\n");

    return 0;
}

#define THREAD_COUNT 10

int main(void)
{
    thrd_t t;

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_create(&t, run, NULL);
        thrd_detach(t);               // <-- DETACH!
    }

    // Sleep for a second to let all the threads finish
    thrd_sleep(&(struct timespec){.tv_sec=1}, NULL);
}

Note that in this code, we put the main thread to sleep for 1 second with thrd_sleep()—more on that later.

Also in the run() function, I have a commented-out line in there that prints out the thread ID as an unsigned long. This is non-portable, because the spec doesn’t say what type a thrd_t is under the hood—it could be a struct for all we know. But that line works on my system.

Something interesting I saw when I ran the code, above, and printed out the thread IDs was that some threads had duplicate IDs! This seems like it should be impossible, but C is allowed to reuse thread IDs after the corresponding thread has exited. So what I was seeing was that some threads completed their run before other threads were launched.

39.6 Thread Local Data

Threads are interesting because they don’t have their own memory beyond local variables. If you want a static variable or file scope variable, all threads will see that same variable.

This can lead to race conditions, where you get Weird Things™ happening.

Check out this example. We have a static variable foo in block scope in run(). This variable will be visible to all threads that pass through the run() function. And the various threads can effectively step on each other’s toes.

Each thread copies foo into a local variable x (which is not shared between threads—all the threads have their own call stacks). So they should be the same, right?

And the first time we print them, they are¹⁷³. But then right after that, we check to make sure they’re still the same.

And they usually are. But not always!

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

int run(void *arg)
{
    int n = *(int*)arg;  // Thread number for humans to differentiate

    free(arg);

    static int foo = 10;  // Static value shared between threads

    int x = foo;  // Automatic local variable--each thread has its own

    // We just assigned x from foo, so they'd better be equal here.
    // (In all my test runs, they were, but even this isn't guaranteed!)

    printf("Thread %d: x = %d, foo = %d\n", n, x, foo);

    // And they should be equal here, but they're not always!
    // (Sometimes they were, sometimes they weren't!)

    // What happens is another thread gets in and increments foo
    // right now, but this thread's x remains what it was before!

    if (x != foo) {
        printf("Thread %d: Craziness! x != foo! %d != %d\n", n, x, foo);
    }

    foo++;  // Increment shared value

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }
}

Here’s an example output (though this varies from run to run):

    Thread 0: x = 10, foo = 10
    Thread 1: x = 10, foo = 10
    Thread 1: Craziness! x != foo! 10 != 11
    Thread 2: x = 12, foo = 12
    Thread 4: x = 13, foo = 13
    Thread 3: x = 14, foo = 14

In thread 1, between the two printf()s, the value of foo somehow changed from 10 to 11, even though clearly there’s no increment between the printf()s!

It was another thread that got in there (probably thread 0, from the look of it) and incremented the value of foo behind thread 1’s back!

Let’s solve this problem two different ways. (If you want all the threads to share the variable and not step on each other’s toes, you’ll have to read on to the mutex section.)

39.6.1 `_Thread_local` Storage-Class

First things first, let’s just look at the easy way around this: the _Thread_local storage-class.

Basically we’re just going to slap this on the front of our block scope static variable and things will work! It tells C that every thread should have its own version of this variable, so none of them step on each other’s toes.

The <threads.h> header defines thread_local as an alias to _Thread_local so your code doesn’t have to look so ugly.

Let’s take the previous example and make foo into a thread_local variable so that we don’t share that data.

int run(void *arg)
{
    int n = *(int*)arg;  // Thread number for humans to differentiate

    free(arg);

    thread_local static int foo = 10;  // <-- No longer shared!!

And running we get:

    Thread 0: x = 10, foo = 10
    Thread 1: x = 10, foo = 10
    Thread 2: x = 10, foo = 10
    Thread 4: x = 10, foo = 10
    Thread 3: x = 10, foo = 10

No more weird problems!

One thing: if a thread_local variable is block scope, it must be static. Them’s the rules. (But this is OK because non-static variables are per-thread already since each thread has it’s own non-static variables.)

A bit of a lie there: block scope thread_local variables can also be extern.

39.6.2 Another Option: Thread-Specific Storage

Thread-specific storage (TSS) is another way of getting per-thread data.

One additional feature is that these functions allow you to specify a destructor that will be called on the data when the TSS variable is deleted. Commonly this destructor is free() to automatically clean up malloc()d per-thread data. Or NULL if you don’t need to destroy anything.

The destructor is type tss_dtor_t which is a pointer to a function that returns void and takes a void* as an argument (the void* points to the data stored in the variable). In other words, it’s a void (*)(void*), if that clears it up. Which I admit it probably doesn’t. Check out the example, below.

Generally, thread_local is probably your go-to, but if you like the destructor idea, then you can make use of that.

The usage is a bit weird in that we need a variable of type tss_t to be alive to represent the value on a per thread basis. Then we initialize it with tss_create(). Eventually we get rid of it with tss_delete(). Note that calling tss_delete() doesn’t run all the destructors—it’s thrd_exit() (or returning from the run function) that does that. tss_delete() just releases any memory allocated by tss_create().

In the middle, threads can call tss_set() and tss_get() to set and get the value.

In the following code, we set up the TSS variable before creating the threads, then clean up after the threads.

In the run() function, the threads malloc() some space for a string and store that pointer in the TSS variable.

When the thread exits, the destructor function (free() in this case) is called for all the threads.

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

tss_t str;

void some_function(void)
{
    // Retrieve the per-thread value of this string
    char *tss_string = tss_get(str);

    // And print it
    printf("TSS string: %s\n", tss_string);
}

int run(void *arg)
{
    int serial = *(int*)arg;  // Get this thread's serial number
    free(arg);

    // malloc() space to hold the data for this thread
    char *s = malloc(64);
    sprintf(s, "thread %d! :)", serial);  // Happy little string

    // Set this TSS variable to point at the string
    tss_set(str, s);

    // Call a function that will get the variable
    some_function();

    return 0;   // Equivalent to thrd_exit(0)
}

#define THREAD_COUNT 15

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Make a new TSS variable, the free() function is the destructor
    tss_create(&str, free);

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // All threads are done, so we're done with this
    tss_delete(str);
}

Again, this is kind of a painful way of doing things compared to thread_local, so unless you really need that destructor functionality, I’d use that instead.

39.7 Mutexes

If you want to only allow a single thread into a critical section of code at a time, you can protect that section with a mutex¹⁷⁴.

For example, if we had a static variable and we wanted to be able to get and set it in two operations without another thread jumping in the middle and corrupting it, we could use a mutex for that.

You can acquire a mutex or release it. If you attempt to acquire the mutex and succeed, you may continue execution. If you attempt and fail (because someone else holds it), you will block¹⁷⁵ until the mutex is released.

If multiple threads are blocked waiting for a mutex to be released, one of them will be chosen to run (at random, from our perspective), and the others will continue to sleep.

The gameplan is that first we’ll initialize a mutex variable to make it ready to use with mtx_init().

Then subsequent threads can call mtx_lock() and mtx_unlock() to get and release the mutex.

When we’re completely done with the mutex, we can destroy it with mtx_destroy(), the logical opposite of mtx_init().

First, let’s look at some code that does not use a mutex, and endeavors to print out a shared (static) serial number and then increment it. Because we’re not using a mutex over the getting of the value (to print it) and the setting (to increment it), threads might get in each other’s way in that critical section.

#include <stdio.h>
#include <threads.h>

int run(void *arg)
{
    (void)arg;

    static int serial = 0;   // Shared static variable!

    printf("Thread running! %d\n", serial);

    serial++;

    return 0;
}

#define THREAD_COUNT 10

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_create(t + i, run, NULL);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }
}

When I run this, I get something that looks like this:

    Thread running! 0
    Thread running! 0
    Thread running! 0
    Thread running! 3
    Thread running! 4
    Thread running! 5
    Thread running! 6
    Thread running! 7
    Thread running! 8
    Thread running! 9

Clearly multiple threads are getting in there and running the printf() before anyone gets a change to update the serial variable.

What we want to do is wrap the getting of the variable and setting of it into a single mutex-protected stretch of code.

We’ll add a new variable to represent the mutex of type mtx_t in file scope, initialize it, and then the threads can lock and unlock it in the run() function.

#include <stdio.h>
#include <threads.h>

mtx_t serial_mtx;     // <-- MUTEX VARIABLE

int run(void *arg)
{
    (void)arg;

    static int serial = 0;   // Shared static variable!

    // Acquire the mutex--all threads will block on this call until
    // they get the lock:

    mtx_lock(&serial_mtx);           // <-- ACQUIRE MUTEX

    printf("Thread running! %d\n", serial);

    serial++;

    // Done getting and setting the data, so free the lock. This will
    // unblock threads on the mtx_lock() call:

    mtx_unlock(&serial_mtx);         // <-- RELEASE MUTEX

    return 0;
}

#define THREAD_COUNT 10

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Initialize the mutex variable, indicating this is a normal
    // no-frills, mutex:

    mtx_init(&serial_mtx, mtx_plain);        // <-- CREATE MUTEX

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_create(t + i, run, NULL);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // Done with the mutex, destroy it:

    mtx_destroy(&serial_mtx);                // <-- DESTROY MUTEX
}

See how on lines 38 and 50 of main() we initialize and destroy the mutex.

But each individual thread acquires the mutex on line 15 and releases it on line 24.

In between the mtx_lock() and mtx_unlock() is the critical section, the area of code where we don’t want multiple threads mucking about at the same time.

And now we get proper output!

    Thread running! 0
    Thread running! 1
    Thread running! 2
    Thread running! 3
    Thread running! 4
    Thread running! 5
    Thread running! 6
    Thread running! 7
    Thread running! 8
    Thread running! 9

If you need multiple mutexes, no problem: just have multiple mutex variables.

And always remember the Number One Rule of Multiple Mutexes: Unlock mutexes in the opposite order in which you lock them!

39.7.1 Different Mutex Types

As hinted earlier, we have a few mutex types that you can create with mtx_init(). (Some of these types are the result of a bitwise-OR operation, as noted in the table.)

Type	Description
`mtx_plain`	Regular ol’ mutex
`mtx_timed`	Mutex that supports timeouts
`mtx_plain\|mtx_recursive`	Recursive mutex
`mtx_timed\|mtx_recursive`	Recursive mutex that supports timeouts

“Recursive” means that the holder of a lock can call mtx_lock() multiple times on the same lock. (They have to unlock it an equal number of times before anyone else can take the mutex.) This might ease coding from time to time, especially if you call a function that needs to lock the mutex when you already hold the mutex.

And the timeout gives a thread a chance to try to get the lock for a while, but then bail out if it can’t get it in that timeframe.

For a timeout mutex, be sure to create it with mtx_timed:

    mtx_init(&serial_mtx, mtx_timed);

And then when you wait for it, you have to specify a time in UTC when it will unlock¹⁷⁶.

The function timespec_get() from <time.h> can be of assistance here. It’ll get you the current time in UTC in a struct timespec which is just what we need. In fact, it seems to exist merely for this purpose.

It has two fields: tv_sec has the current time in seconds since epoch, and tv_nsec has the nanoseconds (billionths of a second) as the “fractional” part.

So you can load that up with the current time, and then add to it to get a specific timeout.

Then call mtx_timedlock() instead of mtx_lock(). If it returns the value thrd_timedout, it timed out.

    struct timespec timeout;
    
    timespec_get(&timeout, TIME_UTC);  // Get current time
    timeout.tv_sec += 1;               // Timeout 1 second after now
    
    int result = mtx_timedlock(&serial_mtx, &timeout));
    
    if (result == thrd_timedout) {
        printf("Mutex lock timed out!\n");
    }

Other than that, timed locks are the same as regular locks.

39.8 Condition Variables

Condition Variables are the last piece of the puzzle we need to make performant multithreaded applications and to compose more complex multithreaded structures.

A condition variable provides a way for threads to go to sleep until some event on another thread occurs.

In other words, we might have a number of threads that are rearing to go, but they have to wait until some event is true before they continue. Basically they’re being told “wait for it!” until they get notified.

And this works hand-in-hand with mutexes since what we’re going to wait on generally depends on the value of some data, and that data generally needs to be protected by a mutex.

It’s important to note that the condition variable itself isn’t the holder of any particular data from our perspective. It’s merely the variable by which C keeps track of the waiting/not-waiting status of a particular thread or group of threads.

Let’s write a contrived program that reads in groups of 5 numbers from the main thread one at a time. Then, when 5 numbers have been entered, the child thread wakes up, sums up those 5 numbers, and prints the result.

The numbers will be stored in a global, shared array, as will the index into the array of the about-to-be-entered number.

Since these are shared values, we at least have to hide them behind a mutex for both the main and child threads. (The main will be writing data to them and the child will be reading data from them.)

But that’s not enough. The child thread needs to block (“sleep”) until 5 numbers have been read into the array. And then the parent thread needs to wake up the child thread so it can do its work.

And when it wakes up, it needs to be holding that mutex. And it will! When a thread waits on a condition variable, it also acquires a mutex when it wakes up.

How’s that work? Let’s look at the outline of what the child thread will do:

Lock the mutex with mtx_lock()
If we haven’t entered all the numbers, wait on the condition variable with cnd_wait()
Do the work that needs doing
Unlock the mutex with mtx_unlock()

Meanwhile the main thread will be doing this:

Lock the mutex with mtx_lock()
Store the recently-read number into the array
If the array is full, signal the child to wake up with cnd_signal()
Unlock the mutex with mtx_unlock()

If you didn’t skim that too hard (it’s OK—I’m not offended), you might notice something weird: how can the main thread hold the mutex lock and signal the child, if the child has to hold the mutex lock to wait for the signal? They can’t both hold the lock!

And indeed they don’t! There’s some behind-the-scenes magic with condition variables: when you cnd_wait(), it releases the mutex that you specify and the thread goes to sleep. And when someone signals that thread to wake up, it reacquires the lock as if nothing had happened.

It’s a little different on the cnd_signal() side of things. This doesn’t do anything with the mutex. The signalling thread still must manually release the mutex before the waiting threads can wake up.

One more thing on the cnd_wait(). You’ll probably be calling cnd_wait() if some condition¹⁷⁷ is not yet met (e.g. in this case, if not all the numbers have yet been entered). Here’s the deal: this condition should be in a while loop, not an if statement. Why?

It’s because of a mysterious phenomenon called a spurious wakeup. Sometimes, in some implementations, a thread can be woken up out of a cnd_wait() sleep for seemingly no reason. [X-Files music]¹⁷⁸. And so we have to check to see that the condition we need is still actually met when we wake up. And if it’s not, back to sleep with us!

So let’s do this thing! Starting with the main thread:

The main thread will set up the mutex and condition variable, and will launch the child thread.
Then it will, in an infinite loop, get numbers as input from the console.
It will also acquire the mutex to store the inputted number into a global array.
When the array has 5 numbers in it, the main thread will signal the child thread that it’s time to wake up and do its work.
Then the main thread will unlock the mutex and go back to reading the next number from the console.

Meanwhile, the child thread has been up to its own shenanigans:

The child thread grabs the mutex
While the condition is not met (i.e. while the shared array doesn’t yet have 5 numbers in it), the child thread sleeps by waiting on the condition variable. When it waits, it unlocks the mutex.
Once the main thread signals the child thread to wake up, it wakes up to do the work and gets the mutex lock back.
The child thread sums the numbers and resets the variable that is the index into the array.
It then releases the mutex and runs again in an infinite loop.

And here’s the code! Give it some study so you can see where all the above pieces are being handled:

#include <stdio.h>
#include <threads.h>

#define VALUE_COUNT_MAX 5

int value[VALUE_COUNT_MAX];  // Shared global
int value_count = 0;   // Shared global, too

mtx_t value_mtx;   // Mutex around value
cnd_t value_cnd;   // Condition variable on value

int run(void *arg)
{
    (void)arg;

    for (;;) {
        mtx_lock(&value_mtx);      // <-- GRAB THE MUTEX

        while (value_count < VALUE_COUNT_MAX) {
            printf("Thread: is waiting\n");
            cnd_wait(&value_cnd, &value_mtx);  // <-- CONDITION WAIT
        }

        printf("Thread: is awake!\n");

        int t = 0;

        // Add everything up
        for (int i = 0; i < VALUE_COUNT_MAX; i++)
            t += value[i];

        printf("Thread: total is %d\n", t);

        // Reset input index for main thread
        value_count = 0;

        mtx_unlock(&value_mtx);   // <-- MUTEX UNLOCK
    }

    return 0;
}

int main(void)
{
    thrd_t t;

    // Spawn a new thread

    thrd_create(&t, run, NULL);
    thrd_detach(t);

    // Set up the mutex and condition variable

    mtx_init(&value_mtx, mtx_plain);
    cnd_init(&value_cnd);

    for (;;) {
        int n;

        scanf("%d", &n);

        mtx_lock(&value_mtx);    // <-- LOCK MUTEX

        value[value_count++] = n;

        if (value_count == VALUE_COUNT_MAX) {
            printf("Main: signaling thread\n");
            cnd_signal(&value_cnd);  // <-- SIGNAL CONDITION
        }

        mtx_unlock(&value_mtx);  // <-- UNLOCK MUTEX
    }

    // Clean up (I know that's an infinite loop above here, but I
    // want to at least pretend to be proper):

    mtx_destroy(&value_mtx);
    cnd_destroy(&value_cnd);
}

And here’s some sample output (individual numbers on lines are my input):

    Thread: is waiting
    1
    1
    1
    1
    1
    Main: signaling thread
    Thread: is awake!
    Thread: total is 5
    Thread: is waiting
    2
    8
    5
    9
    0
    Main: signaling thread
    Thread: is awake!
    Thread: total is 24
    Thread: is waiting

It’s a common use of condition variables in producer-consumer situations like this. If we didn’t have a way to put the child thread to sleep while it waited for some condition to be met, it would be force to poll which is a big waste of CPU.

39.8.1 Timed Condition Wait

There’s a variant of cnd_wait() that allows you to specify a timeout so you can stop waiting.

Since the child thread must relock the mutex, this doesn’t necessarily mean that you’ll be popping back to life the instant the timeout occurs; you still must wait for any other threads to release the mutex.

But it does mean that you won’t be waiting until the cnd_signal() happens.

To make this work, call cnd_timedwait() instead of cnd_wait(). If it returns the value thrd_timedout, it timed out.

The timestamp is an absolute time in UTC, not a time-from-now. Thankfully the timespec_get() function in <time.h> seems custom-made for exactly this case.

    struct timespec timeout;
    
    timespec_get(&timeout, TIME_UTC);  // Get current time
    timeout.tv_sec += 1;               // Timeout 1 second after now
    
    int result = cnd_timedwait(&condition, &mutex, &timeout));
    
    if (result == thrd_timedout) {
        printf("Condition variable timed out!\n");
    }

39.8.2 Broadcast: Wake Up All Waiting Threads

cnd_signal() only wakes up one thread to continue working. Depending on how you have your logic done, it might make sense to wake up more than one thread to continue once the condition is met.

Of course only one of them can grab the mutex, but if you have a situation where:

The newly-awoken thread is responsible for waking up the next one, and—
There’s a chance the spurious-wakeup loop condition will prevent it from doing so, then—

you’ll want to broadcast the wake up so that you’re sure to get at least one of the threads out of that loop to launch the next one.

How, you ask?

Simply use cnd_broadcast() instead of cnd_signal(). Exact same usage, except cnd_broadcast() wakes up all the sleeping threads that were waiting on that condition variable.

39.9 Running a Function One Time

Let’s say you have a function that could be run by many threads, but you don’t know when, and it’s not work trying to write all that logic.

There’s a way around it: use call_once(). Tons of threads could try to run the function, but only the first one counts¹⁷⁹

To work with this, you need a special flag variable you declare to keep track of whether or not the thing’s been run. And you need a function to run, which takes no parameters and returns no value.

    once_flag of = ONCE_FLAG_INIT;  // Initialize it like this
    
    void run_once_function(void)
    {
        printf("I'll only run once!\n");
    }
    
    int run(void *arg)
    {
        (void)arg;
    
        call_once(&of, run_once_function);
    
        // ...

In this example, no matter how many threads get into the run() function, the run_once_function() will only be called a single time.

40 `<stdio.h>` Standard I/O Library

Function	Description
`clearerr()`	Clear the `feof` and `ferror` status flags
`fclose()`	Close an open file
`feof()`	Return the file end-of-file status
`ferror()`	Return the file error status
`fflush()`	Flush all buffered output to a file
`fgetc()`	Read a character in a file
`fgetpos()`	Get the file I/O position
`fgets()`	Read a line from a file
`fopen()`	Open a file
`fprintf()`	Print formatted output to a file
`fputc()`	Print a character to a file
`fputs()`	Print a string to a file
`fread()`	Read binary data from a file
`freopen()`	Change file associated with a stream
`fscanf()`	Read formatted input from a file
`fseek()`	Set the file I/O position
`fsetpos()`	Set the file I/O position
`ftell()`	Get the file I/O position
`fwrite()`	Write binary data to a file
`getc()`	Get a character from `stdin`
`getchar()`	Get a character from `stdin`
`gets()`	Get a string from `stdin` (removed in C11)
`perror()`	Print a human-formatted error message
`printf()`	Print formatted output to `stdout`
`putc()`	Print a character to `stdout`
`putchar()`	Print a character to `stdout`
`puts()`	Print a string to `stdout`
`remove()`	Delete a file from disk
`rename()`	Rename or move a file on disk
`rewind()`	Set the I/O position to the beginning of a file
`scanf()`	Read formatted input from `stdin`
`setbuf()`	Configure buffering for I/O operations
`setvbuf()`	Configure buffering for I/O operations
`snprintf()`	Print length-limited formatted output to a string
`sprintf()`	Print formatted output to a string
`sscanf()`	Read formatted input from a string
`tmpfile()`	Create a temporary file
`tmpnam()`	Generate a unique name for a temporary file
`ungetc()`	Push a character back on the input stream
`vfprintf()`	Variadic print formatted output to a file
`vfscanf()`	Variadic read formatted input from a file
`vprintf()`	Variadic print formatted output to `stdout`
`vscanf()`	Variadic read formatted input from `stdin`
`vsnprintf()`	Variadic length-limited print formatted output to a string
`vsprintf()`	Variadic print formatted output to a string
`vsscanf()`	Variadic read formatted input to a string

The most basic of all libraries in the whole of the standard C library is the standard I/O library. It’s used for reading from and writing to files. I can see you’re very excited about this.

So I’ll continue. It’s also used for readingand writing to the console, as we’ve already often seen with the printf() function.

(A little secret here—many many things in various operating systems are secretly files deep down, and the console is no exception. “Everything in Unix is a file!” :-))

You’ll probably want some prototypes of the functions you can use, right? To get your grubby little mittens on those, you’ll want to include stdio.h.

Anyway, so we can do all kinds of cool stuff in terms of file I/O. LIE DETECTED. Ok, ok. We can do all kinds of stuff in terms of file I/O. Basically, the strategy is this:

Use fopen() to get a pointer to a file structure of type FILE*. This pointer is what you’ll be passing to many of the other file I/O calls.
Use some of the other file calls, like fscanf(), fgets(), fprintf(), or etc. using the FILE* returned from fopen().
When done, call fclose() with the FILE*. This let’s the operating system know that you’re truly done with the file, no take-backs.

What’s in the FILE*? Well, as you might guess, it points to a struct that contains all kinds of information about the current read and write position in the file, how the file was opened, and other stuff like that. But, honestly, who cares. No one, that’s who. The FILE structure is opaque to you as a programmer; that is, you don’t need to know what’s in it, and you don’t even want to know what’s in it. You just pass it to the other standard I/O functions and they know what to do.

This is actually pretty important: try to not muck around in the FILE structure. It’s not even the same from system to system, and you’ll end up writing some really non-portable code.

One more thing to mention about the standard I/O library: a lot of the functions that operate on files use an “f” prefix on the function name. The same function that is operating on the console will leave the “f” off. For instance, if you want to print to the console, you use printf(), but if you want to print to a file, use fprintf(), see?

Wait a moment! If writing to the console is, deep down, just like writing to a file, since everything in Unix is a file, why are there two functions? Answer: it’s more convenient. But, more importantly, is there a FILE* associated with the console that you can use? Answer: YES!

There are, in fact, three (count ’em!) special FILE*s you have at your disposal merely for just including stdio.h. There is one for input, and two for output.

That hardly seems fair—why does output get two files, and input only get one?

That’s jumping the gun a bit—let’s just look at them:

Stream	Description
`stdin`	Input from the console.
`stdout`	Output to the console.
`stderr`	Output to the console on the error file stream.

So standard input (stdin) is by default just what you type at the keyboard. You can use that in fscanf() if you want, just like this:

    /* this line: */
    scanf("%d", &x);
    
    /* is just like this line: */
    fscanf(stdin, "%d", &x);

And stdout works the same way:

    printf("Hello, world!\n");
    fprintf(stdout, "Hello, world!\n"); /* same as previous line! */

So what is this stderr thing? What happens when you output to that? Well, generally it goes to the console just like stdout, but people use it for error messages, specifically. Why? On many systems you can redirect the output from the program into a file from the command line…and sometimes you’re interested in getting just the error output. So if the program is good and writes all its errors to stderr, a user can redirect just stderr into a file, and just see that. It’s just a nice thing you, as a programmer, can do.

Finally, a lot of these functions return int where you might expect char. This is because the function can return a character or end-of-file (EOF), and EOF is potentially an integer. If you don’t get EOF as a return value, you can safely store the result in a char.

40.1 `remove()`

Delete a file

Synopsis

    #include <stdio.h>
    
    int remove(const char *filename);

Description

Removes the specified file from the filesystem. It just deletes it. Nothing magical. Simply call this function and sacrifice a small chicken and the requested file will be deleted.

Return Value

Returns zero on success, and -1 on error, setting errno.

Example

char *filename = "/home/beej/evidence.txt";

remove(filename);
remove("/disks/d/Windows/system.ini");

40.2 `rename()`

Renames a file and optionally moves it to a new location

Synopsis

    #include <stdio.h>
    
    int rename(const char *old, const char *new);

Description

Renames the file old to name new. Use this function if you’re tired of the old name of the file, and you are ready for a change. Sometimes simply renaming your files makes them feel new again, and could save you money over just getting all new files!

One other cool thing you can do with this function is actually move a file from one directory to another by specifying a different path for the new name.

Return Value

Returns zero on success, and -1 on error, setting errno.

Example

rename("foo", "bar"); // changes the name of the file "foo" to "bar"

// the following moves the file "evidence.txt" from "/tmp" to
// "/home/beej", and also renames it to "nothing.txt":
rename("/tmp/evidence.txt", "/home/beej/nothing.txt");

40.3 `tmpfile()`

Create a temporary file

Synopsis

    #include <stdio.h>
    
    FILE *tmpfile(void);

Description

This is a nifty little function that will create and open a temporary file for you, and will return a FILE* to it that you can use. The file is opened with mode “r+b”, so it’s suitable for reading, writing, and binary data.

By using a little magic, the temp file is automatically deleted when it is close()’d or when your program exits. (Specifically, tmpfile() unlinks the file right after it opens it. If you don’t know what that means, it won’t affect your tmpfile() skill, but hey, be curious! It’s for your own good!)

Return Value

This function returns an open FILE* on success, or NULL on failure.

Example

#include <stdio.h>

int main(void)
{
    FILE *temp;
    char s[128];

    temp = tmpfile();

    fprintf(temp, "What is the frequency, Alexander?\n");

    rewind(temp); // back to the beginning

    fscanf(temp, "%s", s); // read it back out

    fclose(temp); // close (and magically delete)
}

40.4 `tmpnam()`

Generate a unique name for a temporary file

Synopsis

    #include <stdio.h>
    
    char *tmpnam(char *s);

Description

This function takes a good hard look at the existing files on your system, and comes up with a unique name for a new file that is suitable for temporary file usage.

Let’s say you have a program that needs to store off some data for a short time so you create a temporary file for the data, to be deleted when the program is done running. Now imagine that you called this file foo.txt. This is all well and good, except what if a user already has a file called foo.txt in the directory that you ran your program from? You’d overwrite their file, and they’d be unhappy and stalk you forever. And you wouldn’t want that, now would you?

Ok, so you get wise, and you decide to put the file in /tmp so that it won’t overwrite any important content. But wait! What if some other user is running your program at the same time and they both want to use that filename? Or what if some other program has already created that file?

See, all of these scary problems can be completely avoided if you just use tmpnam() to get a safe-ready-to-use filename.

So how do you use it? There are two amazing ways. One, you can declare an array (or malloc() it—whatever) that is big enough to hold the temporary file name. How big is that? Fortunately there has been a macro defined for you, L_tmpnam, which is how big the array must be.

And the second way: just pass NULL for the filename. tmpnam() will store the temporary name in a static array and return a pointer to that. Subsequent calls with a NULL argument will overwrite the static array, so be sure you’re done using it before you call tmpnam() again.

Again, this function just makes a file name for you. It’s up to you to later fopen() the file and use it.

One more note: some compilers warn against using tmpnam() since some systems have better functions (like the Unix function mkstemp().) You might want to check your local documentation to see if there’s a better option. Linux documentation goes so far as to say, “Never use this function. Use mkstemp() instead.”

I, however, am going to be a jerk and not talk about mkstemp() because it’s not in the standard I’m writing about. Nyaah.

The macro TMP_MAX holds the number of unique filenames that can be generated by tmpnam(). Ironically, it is the minimum number of such filenames.

Return Value

Returns a pointer to the temporary file name. This is either a pointer to the string you passed in, or a pointer to internal static storage if you passed in NULL. On error (like it can’t find any temporary name that is unique), tmpnam() returns NULL.

Example

char filename[L_tmpnam];
char *another_filename;

if (tmpnam(filename) != NULL)
    printf("We got a temp file named: \"%s\"\n", filename);
else
    printf("Something went wrong, and we got nothing!\n");

another_filename = tmpnam(NULL);
printf("We got another temp file named: \"%s\"\n", another_filename);
printf("And we didn't error check it because we're too lazy!\n");

On my Linux system, this generates the following output:

    We got a temp file named: "/tmp/filew9PMuZ"
    We got another temp file named: "/tmp/fileOwrgPO"
    And we didn't error check it because we're too lazy!

40.5 `fclose()`

The opposite of fopen()—closes a file when you’re done with it so that it frees system resources.

Synopsis

    #include <stdio.h>
    
    int fclose(FILE *stream);

Description

When you open a file, the system sets aside some resources to maintain information about that open file. Usually it can only open so many files at once. In any case, the Right Thing to do is to close your files when you’re done using them so that the system resources are freed.

Also, you might not find that all the information that you’ve written to the file has actually been written to disk until the file is closed. (You can force this with a call to fflush().)

When your program exits normally, it closes all open files for you. Lots of times, though, you’ll have a long-running program, and it’d be better to close the files before then. In any case, not closing a file you’ve opened makes you look bad. So, remember to fclose() your file when you’re done with it!

Return Value

On success, 0 is returned. Typically no one checks for this. On error EOF is returned. Typically no one checks for this, either.

Example

FILE *fp;

fp = fopen("spoonDB.dat", r"); // (you should error-check this)
sort_spoon_database(fp);
fclose(fp);   // pretty simple, huh.

40.6 `fflush()`

Process all buffered I/O for a stream right now

Synopsis

    #include <stdio.h>
    
    int fflush(FILE *stream);

Description

When you do standard I/O, as mentioned in the section on the setvbuf() function, it is usually stored in a buffer until a line has been entered or the buffer is full or the file is closed. Sometimes, though, you really want the output to happen right this second, and not wait around in the buffer. You can force this to happen by calling fflush().

The advantage to buffering is that the OS doesn’t need to hit the disk every time you call fprintf(). The disadvantage is that if you look at the file on the disk after the fprintf() call, it might not have actually been written to yet. (“I called fputs(), but the file is still zero bytes long! Why?!”) In virtually all circumstances, the advantages of buffering outweigh the disadvantages; for those other circumstances, however, use fflush().

Note that fflush() is only designed to work on output streams according to the spec. What will happen if you try it on an input stream? Use your spooky voice: who knooooows!

Return Value

On success, fflush() returns zero. If there’s an error, it returns EOF and sets the error condition for the stream (see ferror().)

Example

In this example, we’re going to use the carriage return, which is '\r'. This is like newline ('\n'), except that it doesn’t move to the next line. It just returns to the front of the current line.

What we’re going to do is a little text-based status bar like so many command line programs implement. It’ll do a countdown from 10 to 0 printing over itself on the same line.

What is the catch and what does this have to do with fflush()? The catch is that the terminal is most likely “line buffered” (see the section on setvbuf() for more info), meaning that it won’t actually display anything until it prints a newline. But we’re not printing newlines; we’re just printing carriage returns, so we need a way to force the output to occur even though we’re on the same line. Yes, it’s fflush()!

#include <stdio.h>
#include <unistd.h> // for prototype for sleep()

int main(void)
{
    int count;

    for(count = 10; count >= 0; count--) {
        printf("\rSeconds until launch: ");  // lead with a CR
        if (count > 0)
            printf("%2d", count);
        else
            printf("blastoff!\n");

        // force output now!!
        fflush(stdout);

        // the sleep() function is non-standard, but virtually every
        // system implements it--it simply delays for the specificed
        // number of seconds:
        sleep(1);
    }
}

40.7 `fopen()`

Opens a file for reading or writing

Synopsis

    #include <stdio.h>
    
    FILE *fopen(const char *path, const char *mode);

Description

The fopen() opens a file for reading or writing.

Parameter path can be a relative or fully-qualified path and file name to the file in question.

Parameter mode tells fopen() how to open the file (reading, writing, or both), and whether or not it’s a binary file. Possible modes are:

Mode	Description
`r`	Open the file for reading (read-only).
`w`	Open the file for writing (write-only). The file is created if it doesn’t exist.
`r+`	Open the file for reading and writing. The file has to already exist.
`w+`	Open the file for writing and reading. The file is created if it doesn’t already exist.
`a`	Open the file for append. This is just like opening a file for writing, but it positions the file pointer at the end of the file, so the next write appends to the end. The file is created if it doesn’t exist.
`a+`	Open the file for reading and appending. The file is created if it doesn’t exist.

Any of the modes can have the letter “b” appended to the end, as is “wb” (“write binary”), to signify that the file in question is a binary file. (“Binary” in this case generally means that the file contains non-alphanumeric characters that look like garbage to human eyes.) Many systems (like Unix) don’t differentiate between binary and non-binary files, so the “b” is extraneous. But if your data is binary, it doesn’t hurt to throw the “b” in there, and it might help someone who is trying to port your code to another system.

The macro FOPEN_MAX tells you how many streams (at least) you can have open at once.

The macro FILENAME_MAX tells you what the longest valid filename can be. Don’t go crazy, now.

Return Value

fopen() returns a FILE* that can be used in subsequent file-related calls.

If something goes wrong (e.g. you tried to open a file for read that didn’t exist), fopen() will return NULL.

Example

int main(void)
{
    FILE *fp;

    if ((fp = fopen("datafile.dat", "r")) == NULL) {
        printf("Couldn't open datafile.dat for reading\n");
        exit(1);
    }

    // fp is now initialized and can be read from it
}

40.8 `freopen()`

Reopen an existing FILE*, associating it with a new path

Synopsis

    #include <stdio.h>
    
    FILE *freopen(const char *filename, const char *mode, FILE *stream);

Description

Let’s say you have an existing FILE* stream that’s already open, but you want it to suddenly use a different file than the one it’s using. You can use freopen() to “re-open” the stream with a new file.

Why on Earth would you ever want to do that? Well, the most common reason would be if you had a program that normally would read from stdin, but instead you wanted it to read from a file. Instead of changing all your scanf()s to fscanf()s, you could simply reopen stdin on the file you wanted to read from.

Another usage that is allowed on some systems is that you can pass NULL for filename, and specify a new mode for stream. So you could change a file from “r+” (read and write) to just “r” (read), for instance. It’s implementation dependent which modes can be changed.

When you call freopen(), the old stream is closed. Otherwise, the function behaves just like the standard fopen().

Return Value

freopen() returns stream if all goes well.

If something goes wrong (e.g. you tried to open a file for read that didn’t exist), freopen() will return NULL.

Example

#include <stdio.h>

int main(void)
{
    int i, i2;

    scanf("%d", &i); // read i from stdin

    // now change stdin to refer to a file instead of the keyboard
    freopen("someints.txt", "r", stdin);

    scanf("%d", &i2); // now this reads from the file "someints.txt"

    printf("Hello, world!\n"); // print to the screen

    // change stdout to go to a file instead of the terminal:
    freopen("output.txt", "w", stdout);

    printf("This goes to the file \"output.txt\"\n");

    // this is allowed on some systems--you can change the mode of a file:
    freopen(NULL, "wb", stdout); // change to "wb" instead of "w"
}

40.9 `setbuf()`, `setvbuf()`

Configure buffering for standard I/O operations

Synopsis

    #include <stdio.h>
    
    void setbuf(FILE *stream, char *buf);
    
    int setvbuf(FILE *stream, char *buf, int mode, size_t size);

Description

Now brace yourself because this might come as a bit of a surprise to you: when you printf() or fprintf() or use any I/O functions like that, it does not normally work immediately. For the sake of efficiency, and to irritate you, the I/O on a FILE* stream is buffered away safely until certain conditions are met, and only then is the actual I/O performed. The functions setbuf() and setvbuf() allow you to change those conditions and the buffering behavior.

So what are the different buffering behaviors? The biggest is called “full buffering”, wherein all I/O is stored in a big buffer until it is full, and then it is dumped out to disk (or whatever the file is). The next biggest is called “line buffering”; with line buffering, I/O is stored up a line at a time (until a newline ('\n') character is encountered) and then that line is processed. Finally, we have “unbuffered”, which means I/O is processed immediately with every standard I/O call.

You might have seen and wondered why you could call putchar() time and time again and not see any output until you called putchar('\n'); that’s right—stdout is line-buffered!

Since setbuf() is just a simplified version of setvbuf(), we’ll talk about setvbuf() first.

The stream is the FILE* you wish to modify. The standard says you must make your call to setvbuf() before any I/O operation is performed on the stream, or else by then it might be too late.

The next argument, buf allows you to make your own buffer space (using malloc() or just a char array) to use for buffering. If you don’t care to do this, just set buf to NULL.

Now we get to the real meat of the function: mode allows you to choose what kind of buffering you want to use on this stream. Set it to one of the following:

Mode	Description
`_IOFBF`	`stream` will be fully buffered.
`_IOLBF`	`stream` will be line buffered.
`_IONBF`	`stream` will be unbuffered.

Finally, the size argument is the size of the array you passed in for buf…unless you passed NULL for buf, in which case it will resize the existing buffer to the size you specify.

Now what about this lesser function setbuf()? It’s just like calling setvbuf() with some specific parameters, except setbuf() doesn’t return a value. The following example shows the equivalency:

    // these are the same:
    setbuf(stream, buf);
    setvbuf(stream, buf, _IOFBF, BUFSIZ); // fully buffered
    
    // and these are the same:
    setbuf(stream, NULL);
    setvbuf(stream, NULL, _IONBF, BUFSIZ); // unbuffered

Return Value

setvbuf() returns zero on success, and nonzero on failure. setbuf() has no return value.

Example

FILE *fp;
char lineBuf[1024];

fp = fopen("somefile.txt", "r");
setvbuf(fp, lineBuf, _IOLBF, 1024);  // set to line buffering
// ...
fclose(fp);

fp = fopen("another.dat", "rb");
setbuf(fp, NULL); // set to unbuffered
// ...
fclose(fp);

40.10 `printf()`, `fprintf()`, `sprintf()`, `snprintf()`

Print a formatted string to the console or to a file.

Synopsis

    #include <stdio.h>
    
    int printf(const char *format, ...);
    
    int fprintf(FILE *stream, const char *format, ...);
    
    int sprintf(char * restrict s, const char * restrict format, ...);
    
    int snprintf(char * restrict s, size_t n, const char * restrict format, ...);

Description

These functions print formatted output to a variety of destinations.

Function	Output Destination
`printf()`	Print to console (screen by default, typically).
`fprintf()`	Print to a file.
`sprintf()`	Print to a string.
`snprintf()`	Print to a string (safely).

The only differences between these is are the leading parameters that you pass to them before the format string.

Function	What you pass before `format`
`printf()`	Nothing comes before `format`.
`fprintf()`	Pass a `FILE*`.
`sprintf()`	Pass a `char*` to a buffer to print into.
`snprintf()`	Pass a `char*` to the buffer and a maximum buffer length.

The printf() function is legendary as being one of the most flexible outputting systems ever devised. It can also get a bit freaky here or there, most notably in the format string. We’ll take it a step at a time here.

The easiest way to look at the format string is that it will print everything in the string as-is, unless a character has a percent sign (%) in front of it. That’s when the magic happens: the next argument in the printf() argument list is printed in the way described by the percent code. These percent codes are called format specifiers.

Here are the most common format specifiers.

Specifier	Description
`%d`	Print the next argument as a signed decimal number, like `3490`. The argument printed this way should be an `int`, or something that gets promoted to `int`.
`%f`	Print the next argument as a signed floating point number, like `3.14159`. The argument printed this way should be a `double`, or something that gets promoted to a `double`.
`%c`	Print the next argument as a character, like `'B'`. The argument printed this way should be a `char` variant.
`%s`	Print the next argument as a string, like `"Did you remember your mittens?"`. The argument printed this way should be a `char*` or `char[]`.
`%%`	No arguments are converted, and a plain old run-of-the-mill percent sign is printed. This is how you print a ‘%’ using `printf()`.

So those are the basics. I’ll give you some more of the format specifiers in a bit, but let’s get some more breadth before then. There’s actually a lot more that you can specify in there after the percent sign.

For one thing, you can put a field width in there—this is a number that tells printf() how many spaces to put on one side or the other of the value you’re printing. That helps you line things up in nice columns. If the number is negative, the result becomes left-justified instead of right-justified. Example:

    printf("%10d", x);  /* prints X on the right side of the 10-space field */
    printf("%-10d", x); /* prints X on the left side of the 10-space field */

If you don’t know the field width in advance, you can use a little kung-foo to get it from the argument list just before the argument itself. Do this by placing your seat and tray tables in the fully upright position. The seatbelt is fastened by placing the—cough. I seem to have been doing way too much flying lately. Ignoring that useless fact completely, you can specify a dynamic field width by putting a * in for the width. If you are not willing or able to perform this task, please notify a flight attendant and we will reseat you.

    int width = 12;
    int value = 3490;
    
    printf("%*d\n", width, value);

You can also put a “0” in front of the number if you want it to be padded with zeros:

    int x = 17;
    printf("%05d", x);  /* "00017" */

When it comes to floating point, you can also specify how many decimal places to print by making a field width of the form “x.y” where x is the field width (you can leave this off if you want it to be just wide enough) and y is the number of digits past the decimal point to print:

    float f = 3.1415926535;
    
    printf("%.2f", f);  /* "3.14" */
    printf("%7.3f", f); /* "  3.141" <-- 7 spaces across */

Ok, those above are definitely the most common uses of printf(), but let’s get total coverage.

40.10.0.1 Format Specifier Layout

Technically, the layout of the format specifier is these things in this order:

%, followed by…
Optional: zero or more flags, left justify, leading zeros, etc.
Optional: Field width, how wide the output field should be.
Optional: Precision, or how many decimal places to print.
Optional: Length modifier, for printing things bigger than int or double.
Conversion specifier, like d, f, etc.

In short, the whole format specifier is laid out like this:

    %[flags][fieldwidth][.precision][lengthmodifier]conversionspecifier

What could be easier?

40.10.0.2 Conversion Specifiers

Let’s talk conversion specifiers first. Each of the following specifies what type it can print, but it can also print anything that gets promoted to that type. For example, %d can print int, short, and char.

Conversion Specifier	Description
`d`	Print an `int` argument as a decimal number.
`i`	Identical to `d`.
`o`	Print an `unsigned int` in octal (base 8).
`u`	Print an `unsigned int` in decimal.
`x`	Print an `unsigned int` in hexadecimal with lowercase letters.
`X`	Print an `unsigned int` in hexadecimal with uppercase letters.
`f`	Print a `double` in decimal notation. Infinity is printed as `infinity` or `inf`, and NaN is printed as `nan`, any of which could have a leading minus sign.
`F`	Same as `f`, except it prints out `INFINITY`, `INF`, or `NAN` in all caps.
`e`	Print a number in scientific notation, e.g. `1.234e56`. Does infinity and NaN like `f`.
`E`	Just like `e`, except prints the exponent `E` (and infinity and NaN) in uppercase.
`g`	Print small numbers like `f` and large numbers like `e`. See note below.
`G`	Print small numbers like `F` and large numbers like `E`. See note below.
`a`	Print a `double` in hexadecimal form `0xh.hhhhpd` where `h` is a lowercase hex digit and `d` is a decimal exponent of 2. Infinity and NaN in the form of `f`. More below.
`A`	Like `a` except everything’s uppercase.
`c`	Convert `int` argument to `unsigned char` and print as a character.
`s`	Print a string starting at the given `char*`.
`p`	Print a `void*` out as a number, probably the numeric address, possibly in hex.
`n`	Store the number of characters written so far in the given `int*`. Doesn’t print anything. See below.
`%`	Print a literal percent sign.

40.10.0.2.1 Note on `%a` and `%A`

When printing floating point numbers in hex form, there is one number before the decimal point, and the rest of are out to the precision.

    double pi = 3.14159265358979;
    
    printf("%.3a\n", pi);  // 0x1.922p+1

C can choose the leading number in such a way to ensure subsequent digits align to 4-bit boundaries.

If the precision is left out and the macro FLT_RADIX is a power of 2, enough precision is used to represent the number exactly. If FLT_RADIX is not a power of two, enough precision is used to be able to tell any two floating values apart.

If the precision is 0 and the # flag isn’t specified, the decimal point is omitted.

40.10.0.2.2 Note on `%g` and `%G`

The gist of this is to use scientific notation when the number gets too “extreme”, and regular decimal notation otherwise.

The exact behavior for whether these print as %f or %e depends on a number of factors:

If the number’s exponent is greater than or equal to -4 and the precision is greater than the exponent, we use %f. In this case, the precision is converted according to $p=p-(x+1)$, where $p$ is the specified precision and $x$ is the exponent.

Otherwise we use %e, and the precision becomes $p-1$.

Trailing zeros in the decimal portion are removed. And if there are none left, the decimal point is removed, too. All this unless the # flag is specified.

40.10.0.2.3 Note on `%n`

This specifier is cool and different, and rarely needed. It doesn’t actually print anything, but stores the number of characters printed so far in the next pointer argument in the list.

    int numChars;
    float a = 3.14159;
    int b = 3490;
    
    printf("%f %d%n\n", a, b, &numChars);
    printf("The above line contains %d characters.\n", numChars);

The above example will print out the values of a and b, and then store the number of characters printed so far into the variable numChars. The next call to printf() prints out that result.

    3.141590 3490
    The above line contains 13 characters

40.10.0.3 Length Modifiers

You can stick a length modifier in front of each of the conversion specifiers, if you want. most of those format specifiers work on int or double types, but what if you want larger or smaller types? That’s what these are good for.

For example, you could print out a long long int with the ll modifier:

    long long int x = 3490;
    
    printf("%lld\n", x);  // 3490

Length Modifier	Conversion Specifier	Description
`hh`	`d`, `i`, `o`, `u`, `x`, `X`	Convert argument to `char` (signed or unsigned as appropriate) before printing.
`h`	`d`, `i`, `o`, `u`, `x`, `X`	Convert argument to `short int` (signed or unsigned as appropriate) before printing.
`l`	`d`, `i`, `o`, `u`, `x`, `X`	Argument is a `long int` (signed or unsigned as appropriate).
`ll`	`d`, `i`, `o`, `u`, `x`, `X`	Argument is a `long long int` (signed or unsigned as appropriate).
`j`	`d`, `i`, `o`, `u`, `x`, `X`	Argument is a `intmax_t` or `uintmax_t` (as appropriate).
`z`	`d`, `i`, `o`, `u`, `x`, `X`	Argument is a `size_t`.
`t`	`d`, `i`, `o`, `u`, `x`, `X`	Argument is a `ptrdiff_t`.
`L`	`a`, `A`, `e`, `E`, `f`, `F`, `g`, `G`	Argument is a `long double`.
`l`	`c`	Argument is in a `wint_t`, a wide character.
`l`	`s`	Argument is in a `wchar_t*`, a wide character string.
`hh`	`n`	Store result in `signed char*` argument.
`h`	`n`	Store result in `short int*` argument.
`l`	`n`	Store result in `long int*` argument.
`ll`	`n`	Store result in `long long int*` argument.
`j`	`n`	Store result in `intmax_t*` argument.
`z`	`n`	Store result in `size_t*` argument.
`t`	`n`	Store result in `ptrdiff_t*` argument.

40.10.0.4 Precision

In front of the length modifier, you can put a precision, which generally means how many decimal places you want on your floating point numbers.

To do this, you put a decimal point (.) and the decimal places afterward.

For example, we could print π rounded to two decimal places like this:

    double pi = 3.14159265358979;
    
    printf("%.2f\n", pi);  // 3.14

Conversion Specifier	Precision Value Meaning
`d`, `i`, `o`, `u`, `x`, `X`	For integer types, minimum number of digits (will pad with leading zeros)
`a`, `e`, `f`, `A`, `E`, `F`	For floating types, the precision is the number of digits past the decimal.
`g`, `G`	For floating types, the precision is the number of significant digits printed.
`s`	The maximum number of bytes (not multibyte characters!) to be written.

If no number is specified in the precision after the decimal point, the precision is zero.

If an * is specified after the decimal, something amazing happens! It means the int argument to printf() before the number to be printed holds the precision. You can use this if you don’t know the precision at compile time.

    int precision;
    double pi = 3.14159265358979;
    
    printf("Enter precision: "); fflush(stdout);
    scanf("%d", &precision);
    
    printf("%.*f\n", precision, pi);

Which gives:

    Enter precision: 4
    3.1416

40.10.0.5 Field Width

In front of the optional precision, you can indicate a field width. This is a decimal number that indicates how wide the region should be in which the argument is printed. The region is padding with leading (or trailing) spaces to make sure it’s wide enough.

If the field width specified is too small to hold the output, it is ignored.

As a preview, you can give a negative field width to justify the item the other direction.

So let’s print a number in a field of width 10. We’ll put some angle brackets around it so we can see the padding spaces in the output.

    printf("<<%10d>>\n", 3490);   // right justified
    printf("<<%-10d>>\n", 3490);  // left justified

    <<      3490>>
    <<3490      >>

Like with the precision, you can use an asterisk (*) as the field width

    int field_width;
    int val = 3490;
    
    printf("Enter field_width: "); fflush(stdout);
    scanf("%d", &field_width);
    
    printf("<<%*d>>\n", field_width, val);

40.10.0.6 Flags

Before the field width, you can put some optional flags that further control the output of the subsequent fields. We just saw that the - flag can be used to left- or right-justify fields. But there are plenty more!

Flag	Description
`-`	For a field width, left justify in the field (right is default).
`+`	If the number is signed, always prefix a `+` or `-` on the front.
[SPACE]	If the number is signed, prefix a space for positive, or a `-` for negative.
`0`	Pad the right-justified field with leading zeros instead of leading spaces.
`#`	Print using an alternate form. See below.

For example, we could pad a hexadecimal number with leading zeros to a field width of 8 with:

    printf("%08x\n", 0x1234);  // 00001234

The # “alternate form” result depends on the conversion specifier.

Conversion Specifier	Alternate Form (`#`) Meaning
`o`	Increase precision of a non-zero number just enough to get one leading `0` on the octal number.
`x`	Prefix a non-zero number with `0x`.
`X`	Same as `x`, except capital `0X`.
`a`, `e`, `f`	Always print a decimal point, even if nothing follows it.
`A`, `E`, `F`	Identical to `a`, `e`, `f`.
`g`, `G`	Always print a decimal point, even if nothing follows it, and keep trailing zeros.

40.10.0.7 `sprintf()` and `snprintf()` Details

Both sprintf() and snprintf() have the quality that if you pass in NULL as the buffer, nothing is written—but you can still check the return value to see how many characters would have been written.

snprintf() always terminates the string with a NUL character. So if you try to write out more than the maximum specified characters, the universe ends.

Just kidding. If you do, snprintf() will write $n-1$ characters so that it has enough room to write the terminator at the end.

Return Value

Returns the number of characters outputted, or a negative number on error.

Example

int a = 100;
float b = 2.717;
char *c = "beej!";
char d = 'X';
int e = 5;

printf("%d", a); /* "100"      */
printf("%f", b); /* "2.717000" */
printf("%s", c); /* "beej!"    */
printf("%c", d); /* "X"        */
printf("110%%"); /* "110%"     */

printf("%10d\n", a);   /* "       100" */
printf("%-10d\n", a);  /* "100       " */
printf("%*d\n", e, a); /* "  100"      */
printf("%.2f\n", b);   /* "2.71"       */

printf("%hhd\n", c); /* "88" <-- ASCII code for 'X' */

printf("%5d %5.2f %c\n", a, b, d); /* "  100  2.71 X" */

40.11 `scanf()`, `fscanf()`, `sscanf()`

Read formatted string, character, or numeric data from the console or from a file.

Synopsis

    #include <stdio.h>
    
    int scanf(const char *format, ...);
    
    int fscanf(FILE *stream, const char *format, ...);
    
    int sscanf(const char * restrict s, const char * restrict format, ...);

Description

These functions read formatted output from a variety of sources.

Function	Input Source
`scanf()`	Read from the console (keyboard by default, typically).
`fscanf()`	Read from a file.
`sscanf()`	Read from a string.

The only differences between these is are the leading parameters that you pass to them before the format string.

Function	What you pass before `format`
`scanf()`	Nothing comes before `format`.
`fscanf()`	Pass a `FILE*`.
`sscanf()`	Pass a `char*` to a buffer to read from.

The scanf() family of functions reads data from the console or from a FILE stream, parses it, and stores the results away in variables you provide in the argument list.

The format string is very similar to that in printf() in that you can tell it to read a "%d", for instance for an int. But it also has additional capabilities, most notably that it can eat up other characters in the input that you specify in the format string.

But let’s start simple, and look at the most basic usage first before plunging into the depths of the function. We’ll start by reading an int from the keyboard:

    int a;
    
    scanf("%d", &a);

scanf() obviously needs a pointer to the variable if it is going to change the variable itself, so we use the address-of operator to get the pointer.

In this case, scanf() walks down the format string, finds a “%d”, and then knows it needs to read an integer and store it in the next variable in the argument list, a.

Here are some of the other format specifiers you can put in the format string:

Format Specifier	Description
`%d`	Reads an integer to be stored in an `int`. This integer can be signed.
`%u`	Reads an integer to be stored in an `unsigned int`.
`%f`	Reads a floating point number, to be stored in a `float`.
`%s`	Reads a string up to the first whitespace character.
`%c`	Reads a `char`.

And that’s the end of the story!

Ha! Just kidding. If you’ve just arrived from the printf() page, you know there’s a near-infinite amount of additional material.

40.11.0.1 Consuming Other Characters

scanf() will move along the format string matching any characters you include.

For example, you could read a hyphenated date like so:

    scanf("%u-%u-%u", &yyyy, &mm, &dd);

In that case, scanf() will attempt to consume an unsigned decimal number, then a hyphen, then another unsigned number, then another hypen, then another unsigned number.

If it fails to match at any point (e.g. the user entered “foo”), scanf() will bail without consuming the offending characters.

And it will return the number of variables successfully converted. In the example above, if the user entered a valid string, scanf() would return 3, one for each variable successfully read.

40.11.0.2 Problems with `scanf()`

I (and the C FAQ and a lot of people) recommend against using scanf() to read directly from the keyboard. It’s too easy for it to stop consuming characters when the user enters some bad data.

If you have data in a file and you’re confident it’s in good shape, fscanf() can be really useful.

But in the case of the keyboard or file, you can always use fgets() to read a complete line into a buffer, and then use sscanf() to scan things out of the buffer. This gives you the best of both worlds.

40.11.0.3 The Deep Details

Let’s check out what a scanf()

And here are some more codes, except these don’t tend to be used as often. You, of course, may use them as often as you wish!

First, the format string. Like we mentioned, it can hold ordinary characters as well as % format specifiers. And whitespace characters.

Whitespace characters have a special role: a whitespace character will cause scanf() to consume as many whitespace characters as it can up to the next non-whitespace character. You can use this to ignore all leading or trailing whitespace.

Also, all format specifiers except for s, c, and [ automatically consume leading whitespace.

But I know what you’re thinking: the meat of this function is in the format specifiers. What do those look like?

These consist of the following, in sequence:

A % sign
Optional: an * to suppress assignment—more later
Optional: a field width—max characters to read
Optional: length modifier, for specifying longer or shorter types
A conversion specifier, like d or f indicating the type to read

40.11.0.4 The Conversion Specifier

Let’s start with the best and last: the conversion specifier.

This is the part of the format specifier that tells us what type of variable scanf() should be reading into, like %d or %f.

Conversion Specifier	Description
`d`	Matches a decimal `int`. Can have a leading sign.
`i`	Like `d`, except will handle it if you put a leading `0x` (hex) or `0` (octal) on the number.
`o`	Matches an octal (base 8) `unsigned int`. Leading zeros are ignored.
`u`	Matches a decimal `unsigned int`.
`x`	Matches a hex (base 16) `unsigned int`.
`f`	Match a floating point number (or scientific notation, or anything `strtod()` can handle).
`c`	Match a `char`, or mutiple `char`s if a field width is given.
`s`	Match a sequence of non-whitespace `char`s.
`[`	Match a sequence of characters from a set. The set ends with `]`. More below.
`p`	Match a pointer, the opposite of `%p` for `printf()`.
`n`	Store the number of characters written so far in the given `int*`. Doesn’t consume anything.
`%`	Match a literal percent sign.

All of the following are equivalent to the f specifier: a, e, g, A, E, F, G.

And capital X is equivalent to lowercase x.

40.11.0.4.1 The Scanset `%[]` Conversion Specifier

This is about the weirdest format specifier there is. It allows you to specify a set of characters (the scanset) to be stored away (likely in an array of chars). Conversion stops when a character that is not in the set is matched.

For example, %[0-9] means “match all numbers zero through nine.” And %[AD-G34] means “match A, D through G, 3, or 4”.

Now, to convolute matters, you can tell scanf() to match characters that are not in the set by putting a caret (^) directly after the %[ and following it with the set, like this: %[^A-C], which means “match all characters that are not A through C.”

To match a close square bracket, make it the first character in the set, like this: %[]A-C] or %[^]A-C]. (I added the “A-C” just so it was clear that the “]” was first in the set.)

To match a hyphen, make it the last character in the set, e.g. to match A-through-C or hyphen: %[A-C-].

So if we wanted to match all letters except “%”, “^”, “]”, “B”, “C”, “D”, “E”, and “-”, we could use this format string: %[^]%^B-E-].

Got it? Now we can go onto the next func—no wait! There’s more! Yes, still more to know about scanf(). Does it never end? Try to imagine how I feel writing about it!

40.11.0.5 The Length Modifier

So you know that “%d” stores into an int. But how do you store into a long, short, or double?

Well, like in printf(), you can add a modifier before the type specifier to tell scanf() that you have a longer or shorter type. The following is a table of the possible modifiers:

Length Modifier	Conversion Specifier	Description
`hh`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `char` (signed or unsigned as appropriate) before printing.
`h`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `short int` (signed or unsigned as appropriate) before printing.
`l`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `long int` (signed or unsigned as appropriate).
`ll`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `long long int` (signed or unsigned as appropriate).
`j`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `intmax_t` or `uintmax_t` (as appropriate).
`z`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `size_t`.
`t`	`d`, `i`, `o`, `u`, `x`, `X`	Convert input to `ptrdiff_t`.
`L`	`a`, `A`, `e`, `E`, `f`, `F`, `g`, `G`	Convert input to `long double`.
`l`	`c`,`s`,`[`	Convert input to `wchar_t`, a wide character.
`l`	`s`	Argument is in a `wchar_t*`, a wide character string.
`hh`	`n`	Store result in `signed char*` argument.
`h`	`n`	Store result in `short int*` argument.
`l`	`n`	Store result in `long int*` argument.
`ll`	`n`	Store result in `long long int*` argument.
`j`	`n`	Store result in `intmax_t*` argument.
`z`	`n`	Store result in `size_t*` argument.
`t`	`n`	Store result in `ptrdiff_t*` argument.

40.11.0.6 Field Widths

The field width generally allows you to specify a maximum number of characters to consume. If the thing you’re trying to match is shorter than the field width, that input will stop being processed before the field width is reached.

So a string will stop being consumed when whitespace is found, even if fewer than the field width characters are matched.

And a float will stop being consumed at the end of the number, even if fewer characters than the field width are matched.

But %c is an interesting one—it doesn’t stop consuming characters on anything. So it’ll go exactly to the field width. (Or 1 character if no field width is given.)

40.11.0.7 Skip Input with `*`

If you put an * in the format specifier, it tells scanf() do to the conversion specified, but not store it anywhere. It simply discards the data as it reads it. This is what you use if you want scanf() to eat some data but you don’t want to store it anywhere; you don’t give scanf() an argument for this conversion.

    // Read 3 ints, but discard the middle one
    scanf("%d %*d %d", &int1, &int3);

Return Value

scanf() returns the number of items assigned into variables. Since assignment into variables stops when given invalid input for a certain format specifier, this can tell you if you’ve input all your data correctly.

Also, scanf() returns EOF on end-of-file.

Example

int a;
long int b;
unsigned int c;
float d;
double e;
long double f;
char s[100];

scanf("%d", &a);  // store an int
scanf(" %d", &a); // eat any whitespace, then store an int
scanf("%s", s); // store a string
scanf("%Lf", &f); // store a long double

// store an unsigned, read all whitespace, then store a long int:
scanf("%u %ld", &c, &b);

// store an int, read whitespace, read "blendo", read whitespace,
// and store a float:
scanf("%d blendo %f", &a, &d);

// read all whitespace, then store all characters up to a newline
scanf(" %[^\n]", s);

// store a float, read (and ignore) an int, then store a double:
scanf("%f %*d %lf", &d, &e);

// store 10 characters:
scanf("%10c", s);

40.12 `vprintf()`, `vfprintf()`, `vsprintf()`, `vsnprintf()`

printf() variants using variable argument lists (va_list)

Synopsis

    #include <stdio.h>
    #include <stdarg.h>
    
    int vprintf(const char * restrict format, va_list arg);
    
    int vfprintf(FILE * restrict stream, const char * restrict format,
                 va_list arg);
    
    int vsprintf(char * restrict s, const char * restrict format, va_list arg);
    
    int vsnprintf(char * restrict s, size_t n, const char * restrict format,
                  va_list arg);

Description

These are just like the printf() variants except instead of taking an actual variable number of arguments, they take a fixed number—the last of which is a va_list that refers to the variable arguments.

Like with printf(), the different variants send output different places.

Function	Output Destination
`vprintf()`	Print to console (screen by default, typically).
`vfprintf()`	Print to a file.
`vsprintf()`	Print to a string.
`vsnprintf()`	Print to a string (safely).

Both vsprintf() and vsnprintf() have the quality that if you pass in NULL as the buffer, nothing is written—but you can still check the return value to see how many characters would have been written.

vsnprintf() always terminates the string with a NUL character. So if you try to write out more than the maximum specified characters, a massive interstellar war breaks out.

Just kidding. If you do, vsnprintf() will write $n-1$ characters so that it has enough room to write the terminator at the end.

As for why in the heck would you ever want to do this, the most common reason is to create your own specialized versions of printf()-type functions, piggybacking on all that printf() functionality goodness.

See the example for an example, predictably.

Return Value

vprintf() and vfprintf() return the number of characters printed, or a negative value on error.

vsprintf() returns the number of characters printed to the buffer, not counting the NUL terminator, or a negative value if an error occurred.

vnsprintf() returns the number of characters printed to the buffer. Or the number that would have been printed if the buffer had been large enough.

Example

In this example, we make our own version of printf() called logger() that timestamps output. Notice how the calls to logger() have all the bells and whistles of printf().

#include <stdio.h>
#include <stdarg.h>
#include <time.h>

int logger(char *format, ...)
{
    va_list va;
    time_t now_secs = time(NULL);
    struct tm *now = gmtime(&now_secs);

    // Output timestamp in format "YYYY-MM-DD hh:mm:ss : "
    printf("%04d-%02d-%02d %02d:%02d:%02d : ",
        now->tm_year + 1900, now->tm_mon + 1, now->tm_mday,
        now->tm_hour, now->tm_min, now->tm_sec);

    va_start(va, format);
    int result = vprintf(format, va);
    va_end(va);

    printf("\n");

    return result;
}

int main(void)
{
    int x = 12;
    float y = 3.2;

    logger("Hello!");
    logger("x = %d and y = %.2f", x, y);
}

40.13 `vscanf()`, `vfscanf()`, `vsscanf()`

scanf() variants using variable argument lists (va_list)

Synopsis

    #include <stdio.h>
    #include <stdarg.h>
    
    int vscanf(const char * restrict format, va_list arg);
    
    int vfscanf(FILE * restrict stream, const char * restrict format,
                va_list arg);
    
    int vsscanf(const char * restrict s, const char * restrict format,
                va_list arg);

Description

These are just like the scanf() variants except instead of taking an actual variable number of arguments, they take a fixed number—the last of which is a va_list that refers to the variable arguments.

Function	Input Source
`vscanf()`	Read from the console (keyboard by default, typically).
`vfscanf()`	Read from a file.
`vsscanf()`	Read from a string.

Like with the vprintf() functions, this would be a good way to add additional functionality that took advantage of the power scanf() has to offer.

Return Value

Returns the number of items successfully scanned, or EOF on end-of-file or error.

Example

I have to admit I was wracking my brain to think of when you’d ever want to use this. The best example I could find was one on Stack Overflow ¹⁸⁰ that error-checks the return value from scanf() against the expected. A variant of that is shown below.

#include <stdio.h>
#include <stdarg.h>
#include <assert.h>

int error_check_scanf(int expected_count, char *format, ...)
{
    va_list va;

    va_start(va, format);
    int count = vscanf(format, va);
    va_end(va);

    // This line will crash the program if the condition is false:
    assert(count == expected_count);

    return count;
}

int main(void)
{
    int a, b;
    float c;

    error_check_scanf(3, "%d, %d/%f", &a, &b, &c);
    error_check_scanf(2, "%d", &a);
}

40.14 `getc()`, `fgetc()`, `getchar()`

Get a single character from the console or from a file.

Synopsis

    #include <stdio.h>
    
    int getc(FILE *stream);
    
    int fgetc(FILE *stream);
    
    int getchar(void);

Description

All of these functions in one way or another, read a single character from the console or from a FILE. The differences are fairly minor, and here are the descriptions:

getc() returns a character from the specified FILE. From a usage standpoint, it’s equivalent to the same fgetc() call, and fgetc() is a little more common to see. Only the implementation of the two functions differs.

fgetc() returns a character from the specified FILE. From a usage standpoint, it’s equivalent to the same getc() call, except that fgetc() is a little more common to see. Only the implementation of the two functions differs.

Yes, I cheated and used cut-n-paste to do that last paragraph.

getchar() returns a character from stdin. In fact, it’s the same as calling getc(stdin).

Return Value

All three functions return the unsigned char that they read, except it’s cast to an int.

If end-of-file or an error is encountered, all three functions return EOF.

Example

// read all characters from a file, outputting only the letter 'b's
// it finds in the file

#include <stdio.h>

int main(void)
{
    FILE *fp;
    int c;

    fp = fopen("datafile.txt", "r"); // error check this!

    // this while-statement assigns into c, and then checks against EOF:

    while((c = fgetc(fp)) != EOF) {
        if (c == 'b') {
            putchar(c);
        }
    }

    fclose(fp);
}

40.15 `gets()`, `fgets()`

Read a string from console or file

Synopsis

    #include <stdio.h>
    
    char *fgets(char *s, int size, FILE *stream);
    char *gets(char *s);

Description

These are functions that will retrieve a newline-terminated string from the console or a file. In other normal words, it reads a line of text. The behavior is slightly different, and, as such, so is the usage. For instance, here is the usage of gets():

Don’t use gets(). In fact, as of C11, it ceases to exist! This is one of the rare cases of a function being removed from the standard.

Admittedly, rationale would be useful, yes? For one thing, gets() doesn’t allow you to specify the length of the buffer to store the string in. This would allow people to keep entering data past the end of your buffer, and believe me, this would be Bad News.

I was going to add another reason, but that’s basically the primary and only reason not to use gets(). As you might suspect, fgets() allows you to specify a maximum string length.

One difference here between the two functions: gets() will devour and throw away the newline at the end of the line, while fgets() will store it at the end of your string (space permitting).

Here’s an example of using fgets() from the console, making it behave more like gets() (with the exception of the newline inclusion):

    char s[100];
    gets(s);  // don't use this--read a line (from stdin)
    fgets(s, sizeof(s), stdin); // read a line from stdin

In this case, the sizeof() operator gives us the total size of the array in bytes, and since a char is a byte, it conveniently gives us the total size of the array.

Of course, like I keep saying, the string returned from fgets() probably has a newline at the end that you might not want. You can write a short function to chop the newline off—in fact, let’s just roll that into our own version of gets()

    #include <stdio.h>
    #include <string.h>
    
    char *ngets(char *s, int size)
    {
        char *rv = fgets(s, size, stdin);
    
        if (rv == NULL)
            return NULL;
    
        char *p = strchr(s, '\n');  // Find a newline
    
        if (p != NULL)  // if there's a newline
            *p = '\0';  // truncate the string there
    
        return s;
    }

So, in summary, use fgets() to read a line of text from the keyboard or a file, and don’t use gets().

Return Value

Both gets() and fgets() return a pointer to the string passed.

On error or end-of-file, the functions return NULL.

Example

char s[100];

gets(s); // read from standard input (don't use this--use fgets()!)

fgets(s, sizeof(s), stdin); // read 100 bytes from standard input

fp = fopen("datafile.dat", "r"); // (you should error-check this)
fgets(s, 100, fp); // read 100 bytes from the file datafile.dat
fclose(fp);

fgets(s, 20, stdin); // read a maximum of 20 bytes from stdin

40.16 `putc()`, `fputc()`, `putchar()`

Write a single character to the console or to a file.

Synopsis

    #include <stdio.h>
    
    int putc(int c, FILE *stream);
    
    int fputc(int c, FILE *stream);
    
    int putchar(int c);

Description

All three functions output a single character, either to the console or to a FILE.

putc() takes a character argument, and outputs it to the specified FILE. fputc() does exactly the same thing, and differs from putc() in implementation only. Most people use fputc().

putchar() writes the character to the console, and is the same as calling putc(c, stdout).

Return Value

All three functions return the character written on success, or EOF on error.

Example

// print the alphabet

#include <stdio.h>

int main(void)
{
    char i;

    for(i = 'A'; i <= 'Z'; i++)
        putchar(i);

    putchar('\n'); // put a newline at the end to make it pretty
}

40.17 `puts()`, `fputs()`

Write a string to the console or to a file.

Synopsis

    #include <stdio.h>
    
    int puts(const char *s);
    
    int fputs(const char *s, FILE *stream);

Description

Both these functions output a NUL-terminated string. puts() outputs to the console, while fputs() allows you to specify the file for output.

Return Value

Both functions return non-negative on success, or EOF on error.

Example

// read strings from the console and save them in a file

#include <stdio.h>

int main(void)
{
    FILE *fp;
    char s[100];

    fp = fopen("datafile.txt", "w"); // error check this!

    while(fgets(s, sizeof(s), stdin) != NULL) { // read a string
        fputs(s, fp);  // write it to the file we opened
    }

    fclose(fp);
}

40.18 `ungetc()`

Pushes a character back into the input stream.

Synopsis

    #include <stdio.h>
    
    int ungetc(int c, FILE *stream);

Description

You know how getc() reads the next character from a file stream? Well, this is the opposite of that—it pushes a character back into the file stream so that it will show up again on the very next read from the stream, as if you’d never gotten it from getc() in the first place.

Why, in the name of all that is holy would you want to do that? Perhaps you have a stream of data that you’re reading a character at a time, and you won’t know to stop reading until you get a certain character, but you want to be able to read that character again later. You can read the character, see that it’s what you’re supposed to stop on, and then ungetc() it so it’ll show up on the next read.

Yeah, that doesn’t happen very often, but there we are.

Here’s the catch: the standard only guarantees that you’ll be able to push back one character. Some implementations might allow you to push back more, but there’s really no way to tell and still be portable.

Return Value

On success, ungetc() returns the character you passed to it. On failure, it returns EOF.

Example

// read a piece of punctuation, then everything after it up to the next
// piece of punctuation. return the punctuation, and store the rest
// in a string
//
// sample input: !foo#bar*baz
//
// output:
//
// !: foo
// #: bar
// *: baz
//

#include <stdio.h>
#include <ctype.h>

int read_punctstring(FILE *fp, char *s)
{
    int origpunct, c;
    
    origpunct = fgetc(fp);

    if (origpunct == EOF)  // return EOF on end-of-file
        return EOF;

    while (c = fgetc(fp), !ispunct(c) && c != EOF)
        *s++ = c;  // save it in the string

    *s = '\0'; // nul-terminate the string

    // if we read punctuation last, ungetc it so we can fgetc it next
    // time:
    if (ispunct(c))
        ungetc(c, fp);

    return origpunct;
}

int main(void)
{
    char s[128];
    char c;

    while((c = read_punctstring(stdin, s)) != EOF) {
        printf("%c: %s\n", c, s);
    }
}

40.19 `fread()`

Read binary data from a file.

Synopsis

    #include <stdio.h>
    
    size_t fread(void *p, size_t size, size_t nmemb, FILE *stream);

Description

You might remember that you can call fopen() with the “b” flag in the open mode string to open the file in “binary” mode. Files open in not-binary (ASCII or text mode) can be read using standard character-oriented calls like fgetc() or fgets(). Files open in binary mode are typically read using the fread() function.

All this function does is says, “Hey, read this many things where each thing is a certain number of bytes, and store the whole mess of them in memory starting at this pointer.”

This can be very useful, believe me, when you want to do something like store 20 ints in a file.

But wait—can’t you use fprintf() with the “%d” format specifier to save the ints to a text file and store them that way? Yes, sure. That has the advantage that a human can open the file and read the numbers. It has the disadvantage that it’s slower to convert the numbers from ints to text and that the numbers are likely to take more space in the file. (Remember, an int is likely 4 bytes, but the string “12345678” is 8 bytes.)

So storing the binary data can certainly be more compact and faster to read.

Return Value

This function returns the number of items successfully read. If all requested items are read, the return value will be equal to that of the parameter nmemb. If EOF occurs, the return value will be zero.

To make you confused, it will also return zero if there’s an error. You can use the functions feof() or ferror() to tell which one really happened.

Example

// read 10 numbers from a file and store them in an array

int main(void)
{
    int i;
    int n[10]
    FILE *fp;

    fp = fopen("binarynumbers.dat", "rb");
    fread(n, sizeof(int), 10, fp);  // read 10 ints
    fclose(fp);

    // print them out:
    for(i = 0; i < 10; i++)
        printf("n[%d] == %d\n", i, n[i]);
}

40.20 `fwrite()`

Write binary data to a file.

Synopsis

    #include <stdio.h>
    
    size_t fwrite(const void *p, size_t size, size_t nmemb, FILE *stream);

Description

This is the counterpart to the fread() function. It writes blocks of binary data to disk. For a description of what this means, see the entry for fread().

Return Value

fwrite() returns the number of items successfully written, which should hopefully be nmemb that you passed in. It’ll return zero on error.

Example

// save 10 random numbers to a file

int main(void)
{
    int i;
    int r[10];
    FILE *fp;

    // populate the array with random numbers:
    for(i = 0; i < 10; i++) {
        r[i] = rand();
    }

    // save the random numbers (10 ints) to the file
    fp = fopen("binaryfile.dat", "wb");
    fwrite(r, sizeof(int), 10, fp); // write 10 ints
    fclose(fp);
}

40.21 `fgetpos()`, `fsetpos()`

Get the current position in a file, or set the current position in a file. Just like ftell() and fseek() for most systems.

Synopsis

    #include <stdio.h>
    
    int fgetpos(FILE *stream, fpos_t *pos);
    
    int fsetpos(FILE *stream, fpos_t *pos);

Description

These functions are just like ftell() and fseek(), except instead of counting in bytes, they use an opaque data structure to hold positional information about the file. (Opaque, in this case, means you’re not supposed to know what the data type is made up of.)

On virtually every system (and certainly every system that I know of), people don’t use these functions, using ftell() and fseek() instead. These functions exist just in case your system can’t remember file positions as a simple byte offset.

Since the pos variable is opaque, you have to assign to it using the fgetpos() call itself. Then you save the value for later and use it to reset the position using fsetpos().

Return Value

Both functions return zero on success, and -1 on error.

Example

char s[100];
fpos_t pos;

fgets(s, sizeof(s), fp); // read a line from the file

fgetpos(fp, &pos);   // save the position

fgets(s, sizeof(s), fp); // read another line from the file

fsetpos(fp, &pos);   // now restore the position to where we saved

40.22 `fseek()`, `rewind()`

Position the file pointer in anticipition of the next read or write.

Synopsis

    #include <stdio.h>
    
    int fseek(FILE *stream, long offset, int whence);
    
    void rewind(FILE *stream);

Description

When doing reads and writes to a file, the OS keeps track of where you are in the file using a counter generically known as the file pointer. You can reposition the file pointer to a different point in the file using the fseek() call. Think of it as a way to randomly access you file.

The first argument is the file in question, obviously. offset argument is the position that you want to seek to, and whence is what that offset is relative to.

Of course, you probably like to think of the offset as being from the beginning of the file. I mean, “Seek to position 3490, that should be 3490 bytes from the beginning of the file.” Well, it can be, but it doesn’t have to be. Imagine the power you’re wielding here. Try to command your enthusiasm.

You can set the value of whence to one of three things:

`whence`	Description
`SEEK_SET`	`offset` is relative to the beginning of the file. This is probably what you had in mind anyway, and is the most commonly used value for `whence`.
`SEEK_CUR`	`offset` is relative to the current file pointer position. So, in effect, you can say, “Move to my current position plus 30 bytes,” or, “move to my current position minus 20 bytes.”
`SEEK_END`	`offset` is relative to the end of the file. Just like `SEEK_SET` except from the other end of the file. Be sure to use negative values for `offset` if you want to back up from the end of the file, instead of going past the end into oblivion.

Speaking of seeking off the end of the file, can you do it? Sure thing. In fact, you can seek way off the end and then write a character; the file will be expanded to a size big enough to hold a bunch of zeros way out to that character.

Now that the complicated function is out of the way, what’s this rewind() that I briefly mentioned? It repositions the file pointer at the beginning of the file:

    fseek(fp, 0, SEEK_SET); // same as rewind()
    rewind(fp);             // same as fseek(fp, 0, SEEK_SET)

Return Value

For fseek(), on success zero is returned; -1 is returned on failure.

The call to rewind() never fails.

Example

fseek(fp, 100, SEEK_SET); // seek to the 100th byte of the file
fseek(fp, -30, SEEK_CUR); // seek backward 30 bytes from the current pos
fseek(fp, -10, SEEK_END); // seek to the 10th byte before the end of file

fseek(fp, 0, SEEK_SET);   // seek to the beginning of the file
rewind(fp);               // seek to the beginning of the file

40.23 `ftell()`

Tells you where a particular file is about to read from or write to.

Synopsis

    #include <stdio.h>
    
    long ftell(FILE *stream);

Description

This function is the opposite of fseek(). It tells you where in the file the next file operation will occur relative to the beginning of the file.

It’s useful if you want to remember where you are in the file, fseek() somewhere else, and then come back later. You can take the return value from ftell() and feed it back into fseek() (with whence parameter set to SEEK_SET) when you want to return to your previous position.

Return Value

Returns the current offset in the file, or -1 on error.

Example

long pos;

// store the current position in variable "pos":
pos = ftell(fp);

// seek ahead 10 bytes:
fseek(fp, 10, SEEK_CUR);

// do some mysterious writes to the file
do_mysterious_writes_to_file(fp);

// and return to the starting position, stored in "pos":
fseek(fp, pos, SEEK_SET);

40.24 `feof()`, `ferror()`, `clearerr()`

Determine if a file has reached end-of-file or if an error has occurred.

Synopsis

    #include <stdio.h>
    
    int feof(FILE *stream);
    
    int ferror(FILE *stream);
    
    void clearerr(FILE *stream);

Description

Each FILE* that you use to read and write data from and to a file contains flags that the system sets when certain events occur. If you get an error, it sets the error flag; if you reach the end of the file during a read, it sets the EOF flag. Pretty simple really.

The functions feof() and ferror() give you a simple way to test these flags: they’ll return non-zero (true) if they’re set.

Once the flags are set for a particular stream, they stay that way until you call clearerr() to clear them.

Return Value

feof() and ferror() return non-zero (true) if the file has reached EOF or there has been an error, respectively.

Example

// read binary data, checking for eof or error
int main(void)
{
    int a;
    FILE *fp;

    fp = fopen("binaryints.dat", "rb");

    // read single ints at a time, stopping on EOF or error:

    while(fread(&a, sizeof(int), 1, fp), !feof(fp) && !ferror(fp)) {
        printf("I read %d\n", a);
    }

    if (feof(fp))
        printf("End of file was reached.\n");

    if (ferror(fp))
        printf("An error occurred.\n");

    fclose(fp);
}

40.25 `perror()`

Print the last error message to stderr

Synopsis

    #include <stdio.h>
    #include <errno.h>  // only if you want to directly use the "errno" var
    
    void perror(const char *s);

Description

Many functions, when they encounter an error condition for whatever reason, will set a global variable called errno (in <errno.h>) for you. errno is just an interger representing a unique error.

But to you, the user, some number isn’t generally very useful. For this reason, you can call perror() after an error occurs to print what error has actually happened in a nice human-readable string.

And to help you along, you can pass a parameter, s, that will be prepended to the error string for you.

One more clever trick you can do is check the value of the errno (you have to include errno.h to see it) for specific errors and have your code do different things. Perhaps you want to ignore certain errors but not others, for instance.

The standard only defines three values for errno, but your system undoubtedly defines more. The three that are defined are:

`errno`	Description
`EDOM`	Math operation outside domain.
`EILSEQ`	Invalid sequence in multibyte to wide character encoding.
`ERANGE`	Result of operation doesn’t fit in specified type.

The catch is that different systems define different values for errno, so it’s not very portable beyond the above 3. The good news is that at least the values are largely portable between Unix-like systems, at least.

Return Value

Returns nothing at all! Sorry!

Example

fseek() returns -1 on error, and sets errno, so let’s use it. Seeking on stdin makes no sense, so it should generate an error:

#include <stdio.h>
#include <errno.h> // must include this to see "errno" in this example

int main(void)
{
    if (fseek(stdin, 10L, SEEK_SET) < 0)
        perror("fseek");

    fclose(stdin); // stop using this stream

    if (fseek(stdin, 20L, SEEK_CUR) < 0) {

        // specifically check errno to see what kind of
        // error happened...this works on Linux, but your
        // mileage may vary on other systems!

        if (errno == EBADF) {
            perror("fseek again, EBADF");
        } else {
            perror("fseek again");
        }
    }
}

And the output is:

    fseek: Illegal seek
    fseek again, EBADF: Bad file descriptor

41 `<string.h>` String Manipulation

Function	Description
`memchr()`	Find the first occurrence of a character in memory.
`memcmp()`	Compare two regions of memory.
`memcpy()`	Copy a region of memory to another.
`memmove()`	Move a (potentially overlapping) region of memory.
`memset()`	Set a region of memory to a value.
`strcat()`	Concatenate (join) two strings together.
`strchr()`	Find the first occurrence of a character in a string.
`strcmp()`	Compare two strings.
`strcoll()`	Compare two strings accounting for locale.
`strcpy()`	Copy a string.
`strcspn()`	Find length of a string not consisting of a set of characters.
`strerror()`	Return a human-readable error message for a given code.
`strlen()`	Return the length of a string.
`strncat()`	Concatenate (join) two strings, length-limited.
`strncmp()`	Compare two strings, length-limited.
`strncpy()`	Copy two strings, length-limited.
`strpbrk()`	Search a string for one of a set of character.
`strrchr()`	Find the last occurrence of a character in a string.
`strspn()`	Find length of a string consisting of a set of characters.
`strstr()`	Find a substring in a string.
`strtok()`	Tokenize a string.
`strxfrm()`	Prepare a string for comparison as if by `strcoll()`.

As has been mentioned earlier in the guide, a string in C is a sequence of bytes in memory, terminated by a NUL character (‘\0’). The NUL at the end is important, since it lets all these string functions (and printf() and puts() and everything else that deals with a string) know where the end of the string actually is.

Fortunately, when you operate on a string using one of these many functions available to you, they add the NUL terminator on for you, so you actually rarely have to keep track of it yourself. (Sometimes you do, especially if you’re building a string from scratch a character at a time or something.)

In this section you’ll find functions for pulling substrings out of strings, concatenating strings together, getting the length of a string, and so forth and so on.

41.1 `memcpy()`, `memmove()`

Copy bytes of memory from one location to another

Synopsis

    #include <string.h>
    
    void *memcpy(void * restrict s1, const void * restrict s2, size_t n);
    
    void *memmove(void *s1, const void *s2, size_t n);

Description

These functions copy memory—as many bytes as you want! From source to destination!

The main difference between the two is that memcpy() cannot safely copy overlapping memory regions, whereas memmove() can.

On the one hand, I’m not sure why you’d want to ever use memcpy() instead of memmove(), but I’ll bet it’s possibly more performant.

The parameters are in a particular order: destination first, then source. I remember this order because it behaves like an “=” assignment: the destination is on the left.

Return Value

Both functions return whatever you passed in for parameter s1 for your convenience.

Example

char s[100] = "Goats";
char t[100];

memcpy(t, s, 6);       // Copy non-overlapping memory

memmove(s + 2, s, 6);  // Copy overlapping memory

41.2 `strcpy()`, `strncpy()`

Copy a string

Synopsis

    #include <string.h>
    
    char *strcpy(char *dest, char *src);
    
    char *strncpy(char *dest, char *src, size_t n);

Description

These functions copy a string from one address to another, stopping at the NUL terminator on the srcstring.

strncpy() is just like strcpy(), except only the first n characters are actually copied. Beware that if you hit the limit, n before you get a NUL terminator on the src string, your dest string won’t be NUL-terminated. Beware! BEWARE!

(If the src string has fewer than n characters, it works just like strcpy().)

You can terminate the string yourself by sticking the '\0' in there yourself:

    char s[10];
    char foo = "My hovercraft is full of eels."; // more than 10 chars
    
    strncpy(s, foo, 9); // only copy 9 chars into positions 0-8
    s[9] = '\0';        // position 9 gets the terminator

Return Value

Both functions return dest for your convenience, at no extra charge.

Example

char *src = "hockey hockey hockey hockey hockey hockey hockey hockey";
char dest[20];

int len;

strcpy(dest, "I like "); // dest is now "I like "

len = strlen(dest);

// tricky, but let's use some pointer arithmetic and math to append
// as much of src as possible onto the end of dest, -1 on the length to
// leave room for the terminator:
strncpy(dest+len, src, sizeof(dest)-len-1);

// remember that sizeof() returns the size of the array in bytes
// and a char is a byte:
dest[sizeof(dest)-1] = '\0'; // terminate

// dest is now:       v null terminator
// I like hockey hocke 
// 01234567890123456789012345

41.3 `strcat()`, `strncat()`

Concatenate two strings into a single string.

Synopsis

    #include <string.h>
    
    int strcat(const char *dest, const char *src);
    
    int strncat(const char *dest, const char *src, size_t n);

Description

“Concatenate”, for those not in the know, means to “stick together”. These functions take two strings, and stick them together, storing the result in the first string.

These functions don’t take the size of the first string into account when it does the concatenation. What this means in practical terms is that you can try to stick a 2 megabyte string into a 10 byte space. This will lead to unintended consequences, unless you intended to lead to unintended consequences, in which case it will lead to intended unintended consequences.

Technical banter aside, your boss and/or professor will be irate.

If you want to make sure you don’t overrun the first string, be sure to check the lengths of the strings first and use some highly technical subtraction to make sure things fit.

You can actually only concatenate the first n characters of the second string by using strncat() and specifying the maximum number of characters to copy.

Return Value

Both functions return a pointer to the destination string, like most of the string-oriented functions.

Example

char dest[20] = "Hello";
char *src = ", World!";
char numbers[] = "12345678";

printf("dest before strcat: \"%s\"\n", dest); // "Hello"

strcat(dest, src);
printf("dest after strcat:  \"%s\"\n", dest); // "Hello, world!"

strncat(dest, numbers, 3); // strcat first 3 chars of numbers
printf("dest after strncat: \"%s\"\n", dest); // "Hello, world!123"

Notice I mixed and matched pointer and array notation there with src and numbers; this is just fine with string functions.

41.4 `strcmp()`, `strncmp()`, `memcmp()`

Compare two strings or memory regions and return a difference.

Synopsis

    #include <string.h>
    
    int strcmp(const char *s1, const char *s2);
    
    int strncmp(const char *s1, const char *s2, size_t n);
    
    int memcmp(const void *s1, const void *s2, size_t n);

Description

All these functions compare chunks of bytes in memory.

strcmp() and strncmp() operate on NUL-terminated strings, whereas memcmp() will compare the number of bytes you specify, brazenly ignoring any NUL characters it finds along the way.

strcmp() compares the entire string down to the end, while strncmp() only compares the first n characters of the strings.

It’s a little funky what they return. Basically it’s a difference of the strings, so if the strings are the same, it’ll return zero (since the difference is zero). It’ll return non-zero if the strings differ; basically it will find the first mismatched character and return less-than zero if that character in s1 is less than the corresponding character in s2. It’ll return greater-than zero if that character in s1 is greater than that in s2.

So if they return 0, the comparison was equal (i.e. the difference was 0.)

These functions can be used as comparison functions for qsort() if you have an array of char*s you want to sort.

Return Value

Returns zero if the strings or memory are the same, less-than zero if the first different character in s1 is less than that in s2, or greater-than zero if the first difference character in s1 is greater than than in s2.

Example

char *s1 = "Muffin";
char *s2 = "Muffin Sandwich";
char *s3 = "Muffin";

strcmp("Biscuits", "Kittens"); // returns < 0 since 'B' < 'K'
strcmp("Kittens", "Biscuits"); // returns > 0 since 'K' > 'B'

if (strcmp(s1, s2) == 0)
    printf("This won't get printed because the strings differ");

if (strcmp(s1, s3) == 0)
    printf("This will print because s1 and s3 are the same");

// this is a little weird...but if the strings are the same, it'll
// return zero, which can also be thought of as "false". Not-false
// is "true", so (!strcmp()) will be true if the strings are the
// same. yes, it's odd, but you see this all the time in the wild
// so you might as well get used to it:

if (!strcmp(s1, s3))
    printf("The strings are the same!")

if (!strncmp(s1, s2, 6))
    printf("The first 6 characters of s1 and s2 are the same");

41.5 `strcoll()`

Compare two strings accounting for locale

Synopsis

    #include <string.h>
    
    int strcoll(const char *s1, const char *s2);

Description

This is basically strcmp(), except that it handles accented characters better depending on the locale.

For example, my strcmp() reports that the character “é” (with accent) is greater than “f”. But that’s hardly useful for alphabetizing.

By setting the LC_COLLATE locale value (either by name or via LC_ALL), you can have strcoll() sort in a way that’s more meaningful by the current locale. For example, by having “é” appear sanely before “f”.

It’s also a lot slower than strcmp() so use it only if you have to. See strxfrm() for a potential speedup.

Return Value

Like the other string comparison functions, strcoll() returns a negative value if s1 is less than s2, or a positive value if s1 is greater than s2. Or 0 if they are equal.

Example

#include <stdio.h>
#include <string.h>
#include <locale.h>

int main(void)
{
    setlocale(LC_ALL, "");

    // If your source character set doesn't support "é" in a string
    // you can replace it with `\u00e9`, the Unicode code point
    // for "é".

    printf("%d\n", strcmp("é", "f"));   // Reports é > f, yuck.
    printf("%d\n", strcoll("é", "f"));  // Reports é < f, yay!
}

41.6 `strxfrm()`

Transform a string for comparing based on locale

Synopsis

    #include <string.h>
    
    size_t strxfrm(char * restrict s1, const char * restrict s2, size_t n);

Description

This is a strange little function, so bear with me.

Firstly, if you haven’t done so, get familiar with strcoll() because this is closely related to that.

OK! Now that you’re back, you can think of strxfrm() as the first part of the strcoll() internals. Basically, strcoll() has to transform a string into a form that can be compared with strcmp(). And it does this with strxfrm() for both strings every time you call it.

strxform() takes string s2 and transforms it (readies it for strcmp()) storing the result in s1. It writes no more than n bytes, protecting us from terrible buffer overflows.

But hang on—there’s another mode! If you pass NULL for s1 and 0 for n, it will return the number of bytes that the transformed string would have used¹⁸¹. This is useful if you need to allocate some space to hold the transformed string before you strcmp() it against another.

What I’m getting at, not to be too blunt, is that strcoll() is slow compared to strcmp(). It does a lot of extra work running strxfrm() on all its strings.

In fact, we can see how it works by writing our own like this:

int my_strcoll(char *s1, char *s2)
{
    // Use n = 0 to just get the lengths of the transformed strings
    int len1 = strxfrm(NULL, s1, 0) + 1;
    int len2 = strxfrm(NULL, s2, 0) + 1;

    // Allocate enough room for each
    char *d1 = malloc(len1);
    char *d2 = malloc(len2);

    // Transform the strings for comparison
    strxfrm(d1, s1, len1);
    strxfrm(d2, s2, len2);

    // Compare the transformed strings
    int result = strcmp(d1, d2);

    // Free up the transformed strings
    free(d2);
    free(d1);

    return result;
}

You see on lines 12, 13, and 16, above how we transform the two input strings and then call strcmp() on the result.

So why do we have this function? Can’t we just call strcoll() and be done with it?

The idea is that if you have one string that you’re going to be comparing against a whole lot of other ones, maybe you just want to transform that string one time, then use the faster strcmp() saving yourself a bunch of the work we had to do in the function, above.

We’ll do that in the example.

Return Value

Returns the number of bytes in the transformed sequence. If the value is greater than n, the results in s1 are meaningless.

Example

#include <stdio.h>
#include <string.h>
#include <locale.h>
#include <malloc.h>

// Transform a string for comparison, returning a malloc'd
// result
char *get_xfrm_str(char *s)
{
    int len = strxfrm(NULL, s, 0) + 1;
    char *d = malloc(len);

    strxfrm(d, s, len);

    return d;
}

// Does half the work of a regular strcoll() because the second
// string arrives already transformed.
int half_strcoll(char *s1, char *s2_transformed)
{
    char *s1_transformed = get_xfrm_str(s1);

    int result = strcmp(s1_transformed, s2_transformed);

    free(s1_transformed);

    return result;
}

int main(void)
{
    setlocale(LC_ALL, "");

    // Pre-transform the string to compare against
    char *s = get_xfrm_str("éfg");

    // Repeatedly compare against "éfg" 
    printf("%d\n", half_strcoll("fgh", s));  // "fgh" > "éfg"
    printf("%d\n", half_strcoll("àbc", s));  // "àbc" < "éfg"
    printf("%d\n", half_strcoll("ĥij", s));  // "ĥij" > "éfg"
    
    free(s);
}

41.7 `strchr()`, `strrchr()`, `memchr()`

Find a character in a string.

Synopsis

    #include <string.h>
    
    char *strchr(char *str, int c);
    
    char *strrchr(char *str, int c);
    
    void *memchr(const void *s, int c, size_t n);

Description

The functions strchr() and strrchr find the first or last occurrence of a letter in a string, respectively. (The extra “r” in strrchr() stands for “reverse”–it looks starting at the end of the string and working backward.) Each function returns a pointer to the char in question, or NULL if the letter isn’t found in the string.

memchr() is similar, except that instead of stopping on the first NUL character, it continues searching for however many bytes you specify.

Quite straightforward.

One thing you can do if you want to find the next occurrence of the letter after finding the first, is call the function again with the previous return value plus one. (Remember pointer arithmetic?) Or minus one if you’re looking in reverse. Don’t accidentally go off the end of the string!

Return Value

Returns a pointer to the occurrence of the letter in the string, or NULL if the letter is not found.

Example

// "Hello, world!"
//       ^  ^   ^
//       A  B   C

char *str = "Hello, world!";
char *p;

p = strchr(str, ',');       // p now points at position A
p = strrchr(str, 'o');      // p now points at position B

p = memchr(str, '!', 13);   // p now points at position C

// repeatedly find all occurrences of the letter 'B'
char *str = "A BIG BROWN BAT BIT BEEJ";
char *p;

for(p = strchr(str, 'B'); p != NULL; p = strchr(p + 1, 'B')) {
    printf("Found a 'B' here: %s\n", p);
}

// output is:
//
// Found a 'B' here: BIG BROWN BAT BIT BEEJ
// Found a 'B' here: BROWN BAT BIT BEEJ
// Found a 'B' here: BAT BIT BEEJ
// Found a 'B' here: BIT BEEJ
// Found a 'B' here: BEEJ

41.8 `strspn()`, `strcspn()`

Return the length of a string consisting entirely of a set of characters, or of not a set of characters.

Synopsis

    #include <string.h>
    
    size_t strspn(char *str, const char *accept);
    
    size_t strcspn(char *str, const char *reject);

Description

strspn() will tell you the length of a string consisting entirely of the set of characters in accept. That is, it starts walking down str until it finds a character that is not in the set (that is, a character that is not to be accepted), and returns the length of the string so far.

strcspn() works much the same way, except that it walks down str until it finds a character in the reject set (that is, a character that is to be rejected.) It then returns the length of the string so far.

Return Value

The length of the string consisting of all characters in accept (for strspn()), or the length of the string consisting of all characters except reject (for strcspn()).

Example

char str1[] = "a banana";
char str2[] = "the bolivian navy on manuvers in the south pacific";

// how many letters in str1 until we reach something that's not a vowel?
n = strspn(str1, "aeiou");  // n == 1, just "a"

// how many letters in str1 until we reach something that's not a, b,
// or space?
n = strspn(str1, "ab "); // n == 4, "a ba"

// how many letters in str2 before we get a "y"?
n = strcspn(str2, "y"); // n = 16, "the bolivian nav"

41.9 `strpbrk()`

Search a string for one of a set of characters

Synopsis

    #include <string.h>
    
    char *strpbrk(const char *s1, const char *s2);

Description

This function searches string s1 for any of the characters that are found in string s2.

It’s just like how strchr() searches for a specific character in a string, except it will match any of the characters found in s2.

Think of the power!

Return Value

Returns a pointer to the first character matched in s1, or NULL if the string isn’t found.

Example

//  p points here after strpbrk
//              v
char *s1 = "Hello, world!";
char *s2 = "dow!";  // Match any of these chars

char *p = strpbrk(s1, s2);  // p points to the o

41.10 `strstr()`

Find a string in another string.

Synopsis

    #include <string.h>
    
    char *strstr(const char *str, const char *substr);

Description

Let’s say you have a big long string, and you want to find a word, or whatever substring strikes your fancy, inside the first string. Then strstr() is for you! It’ll return a pointer to the substr within the str!

Return Value

You get back a pointer to the occurrence of the substr inside the str, or NULL if the substring can’t be found.

Example

char *str = "The quick brown fox jumped over the lazy dogs.";
char *p;

p = strstr(str, "lazy");
printf("%s\n", p); // "lazy dogs."

// p is NULL after this, since the string "wombat" isn't in str:
p = strstr(str, "wombat");

41.11 `strtok()`

Tokenize a string.

Synopsis

    #include <string.h>
    
    char *strtok(char *str, const char *delim);

Description

If you have a string that has a bunch of separators in it, and you want to break that string up into individual pieces, this function can do it for you.

The usage is a little bit weird, but at least whenever you see the function in the wild, it’s consistently weird.

Basically, the first time you call it, you pass the string, str that you want to break up in as the first argument. For each subsequent call to get more tokens out of the string, you pass NULL. This is a little weird, but strtok() remembers the string you originally passed in, and continues to strip tokens off for you.

Note that it does this by actually putting a NUL terminator after the token, and then returning a pointer to the start of the token. So the original string you pass in is destroyed, as it were. If you need to preserve the string, be sure to pass a copy of it to strtok() so the original isn’t destroyed.

Return Value

A pointer to the next token. If you’re out of tokens, NULL is returned.

Example

// break up the string into a series of space or
// punctuation-separated words
char *str = "Where is my bacon, dude?";
char *token;

// Note that the following if-do-while construct is very very
// very very very common to see when using strtok().

// grab the first token (making sure there is a first token!)
if ((token = strtok(str, ".,?! ")) != NULL) {
    do {
        printf("Word: \"%s\"\n", token);

        // now, the while continuation condition grabs the
        // next token (by passing NULL as the first param)
        // and continues if the token's not NULL:
    } while ((token = strtok(NULL, ".,?! ")) != NULL);
}

// output is:
//
// Word: "Where"
// Word: "is"
// Word: "my"
// Word: "bacon"
// Word: "dude"
//

41.12 `memset()`

Set a region of memory to a certain value

Synopsis

    #include <string.h>
    
    void *memset(void *s, int c, size_t n);

Description

This function is what you use to set a region of memory to a particular value, namely c converted into unsigned char.

The most common usage is to zero out an array or struct.

Return Value

memset() returns whatever you passed in as s for happy convenience.

Example

struct banana {
    float ripeness;
    char *peel_color;
    int grams;
};

struct banana b;

memset(&b, 0, sizeof b);

b.ripeness == 0.0;     // True
b.peel_color == NULL;  // True
b.grams == 0;          // True

41.13 `strerror()`

Get a string version of an error number

Synopsis

    #include <string.h>
    
    char *strerror(int errnum);

Description

This function ties closely into perror() (which prints a human-readable error message corresponding to errno). But instead of printing, strerror() returns a pointer to the locale-specific error message string.

So if you ever need that string back for some reason (e.g. you’re going to fprintf() it to a file or something), this function will give it to you. All you need to do is pass in errno as an argument. (Recall that errno gets set as an error status by a variety of functions.)

You can actually pass in any integer for errnum you want. The function will return some message, even if the number doesn’t correspond to any known value for errno.

The values of errno and the strings returned by strerror() are system-dependent.

Return Value

A string error message corresponding to the given error number.

You are not allowed to modify the returned string.

Example

#include <stdio.h>
#include <string.h>
#include <errno.h>

int main(void)
{
    FILE *fp = fopen("NONEXISTENT_FILE.TXT", "r");

    if (fp == NULL) {
        char *errmsg = strerror(errno);
        printf("Error %d opening file: %s\n", errno, errmsg);
    }
}

Output:

    Error 2 opening file: No such file or directory

41.14 `strlen()`

Returns the length of a string.

Synopsis

    #include <string.h>
    
    size_t strlen(const char *s);

Description

This function returns the length of the passed null-terminated string (not counting the NUL character at the end). It does this by walking down the string and counting the bytes until the NUL character, so it’s a little time consuming. If you have to get the length of the same string repeatedly, save it off in a variable somewhere.

Return Value

Returns the number of bytes in the string. Note that this might be different than the number of characters in a multibyte string.

Example

char *s = "Hello, world!"; // 13 characters

// prints "The string is 13 characters long.":

printf("The string is %d characters long.\n", strlen(s));

42 `<math.h>` Mathematics

Many of the following functions have float and long double variants as described below (e.g. pow(), powf(), powl()). The float and long double variants are omitted from the following table to keep your eyeballs from melting out.

Function	Description
`acos()`	Calculate the arc cosine of a number.
`acosh()`	Compute arc hyperbolic cosine.
`asin()`	Calculate the arc sine of a number.
`asinh()`	Compute arc hyperbolic sine.
`atan()`, `atan2()`	Calculate the arc tangent of a number.
`atanh()`	Compute the arc hyperbolic tangent.
`cbrt()`	Compute the cube root.
`ceil()`	Ceiling—return the next whole number not smaller than the given number.
`copysign()`	Copy the sign of one value into another.
`cos()`	Calculate the cosine of a number.
`cosh()`	Compute the hyperbolic cosine.
`erf()`	Compute the error function of the given value.
`erfc()`	Compute the complementary error function of a value.
`exp()`	Compute $e$ raised to a power.
`exp2()`	Compute 2 to a power.
`expm1()`	Compute $e^x-1$.
`fabs()`	Compute the absolute value.
`fdim()`	Return the positive difference between two numbers clamped at 0.
`floor()`	Compute the largest whole number not larger than the given value.
`fma()`	Floating (AKA “Fast”) multiply and add.
`fmax()`, `fmin()`	Return the maximum or minimum of two numbers.
`fmod()`	Compute the floating point remainder.
`fpclassify()`	Return the classification of a given floating point number.
`frexp()`	Break a number into its fraction part and exponent (as a power of 2).
`hypot()`	Compute the length of the hypotenuse of a triangle.
`ilogb()`	Return the exponent of a floating point number.
`isfinite()`	True if the number is not infinite or NaN.
`isgreater()`	True if one argument is greater than another.
`isgreatereequal()`	True if one argument is greater than or equal to another.
`isinf()`	True if the number is infinite.
`isless()`	True if one argument is less than another.
`islesseequal()`	True if one argument is less than or equal to another.
`islessgreater()`	Test if a floating point number is less than or greater than another.
`isnan()`	True if the number is Not-a-Number.
`isnormal()`	True if the number is normal.
`isunordered()`	Macro returns true if either floating point argument is NaN.
`ldexp()`	Multiply a number by an integral power of 2.
`lgamma()`	Compute the natural logarithm of the absolute value of $\Gamma(x)$.
`log()`	Compute the natural logarithm.
`log10()`	Compute the log-base-10 of a number.
`log2()`	Compute the base-2 logarithm of a number.
`logb()`	Extract the exponent of a number given `FLT_RADIX`.
`logp1()`	Compute the natural logarithm of a number plus 1.
`lrint()`	Returns `x` rounded in the current rounding direction as an integer.
`lround()`, `llround()`	Round a number in the good old-fashioned way, returning an integer.
`modf()`	Extract the integral and fractional parts of a number.
`nan()`	Return `NAN`.
`nearbyint()`	Rounds a value in the current rounding direction.
`nextafter()`	Get the next (or previous) representable floating point value.
`nextoward()`	Get the next (or previous) representable floating point value.
`pow()`	Compute a value raised to a power.
`remainder()`	Compute the remainder IEC 60559-style.
`remquo()`	Compute the remainder and (some of the) quotient.
`rint()`	Rounds a value in the current rounding direction.
`round()`	Round a number in the good old-fashioned way.
`scalbn()`, `scalbln()`	Efficiently compute $x\times r^n$, where $r$ is `FLT_RADIX`.
`signbit()`	Return the sign of a number.
`sin()`	Calculate the sine of a number.
`sqrt()`	Calculate the square root of a number.
`tan()`	Calculate the tangent of a number.
`tanh()`	Compute the hyperbolic tangent.
`tgamma()`	Compute the gamma function, $\Gamma(x)$.
`trunc()`	Truncate the fractional part off a floating point value.

It’s your favorite subject: Mathematics! Hello, I’m Doctor Math, and I’ll be making math FUN and EASY!

[vomiting sounds]

Ok, I know math isn’t the grandest thing for some of you out there, but these are merely functions that quickly and easily do math you either know, want, or just don’t care about. That pretty much covers it.

42.1 Math Function Idioms

Many of these math functions exist in three forms, each corresponding to the argument and/or return types the function uses, float, double, or long double.

The alternate form for float is made by appending f to the end of the function name.

The alternate form for long double is made by appending l to the end of the function name.

For example, the pow() function, which computes $x^y$, exists in these forms:

    double      pow(double x, double y);             // double
    float       powf(float x, float y);              // float
    long double powl(long double x, long double y);  // long double

Remember that parameters are given values as if you assigned into them. So if you pass a double into powf(), it’ll choose the closest float it can to hold the double. If the double doesn’t fit, undefined behavior happens.

42.2 Math Types

We have two exciting new types in <math.h>:

float_t
double_t

The float_t type is at least as accurate as a float, and the double_t type is at least as accurate as a double.

The idea with these types is they can represent the most efficient way of storing numbers for maximum speed.

Their actual types vary by implementation, but can be determined by the value of the FLT_EVAL_METHOD macro.

`FLT_EVAL_METHOD`	`float_t` type	`double_t` type
`0`	`float`	`double`
`1`	`double`	`double`
`2`	`long double`	`long double`
Other	Implementation-defined	Implementation-defined

For all defined values of FLT_EVAL_METHOD, float_t is the least-precise type used for all floating calculations.

42.3 Math Macros

There are actually a number of these defined, but we’ll cover most of them in their relevant reference sections, below.

But here are a couple:

NAN represents Not-A-Number.

Defined in <float.h> is FLT_RADIX: the number base used by floating point numbers. This is commonly 2, but could be anything.

42.4 Math Errors

As we know, nothing can ever go wrong with math… except everything!

So there are just a couple errors that might occur when using some of these functions.

Range errors mean that some result is beyond what can be stored in the result type.
Domain errors mean that you’ve passed in an argument that doesn’t have a defined result for this function.
Pole errors mean that the limit of the function as $x$ approaches the given argument is infinite.
Overflow errors are when the result is really large, but can’t be stored without incurring large roundoff error.
Underflow errors are like overflow errors, except with very small numbers.

Now, the C math library can do a couple things when these errors occur:

Set errno to some value, or…
Raise a floating point exception.

Your system might vary on what happens. You can check it by looking at the value of the variable math_errhandling. It will be equivalent to one of the following¹⁸²:

`math_errhandling`	Description
`MATH_ERRNO`	The system uses `errno` for math errors.
`MATH_ERREXCEPT`	The system uses exceptions for math errors.
`MATH_ERRNO \| MATH_ERREXCEPT`	The system does both! (That’s a bitwise-OR!)

You are not allowed to change math_errhandling.

For a fuller description on how exceptions work and their meanings, see the <fenv.h> section.

42.5 Math Pragmas

In case you don’t remember, you can brush up on pragmas back in the C Preprocessor section.

But in a nutshell, they offer various ways to control the compiler’s behavior.

In this case, we have a pragma FP_CONTRACT that can be turned off and on.

What does it mean?

First of all, keep in mind that any operation in an expression can cause rounding error. So each step of the expression can introduce more rounding error.

But what if the compiler knows a double secret way of taking the expression you wrote and converting it to a single instruction that reduced the number of steps such that the intermediate rounding error didn’t occur?

Could it use it? I mean, the results would be different than if you let the rounding error settle each step of the way…

Because the results would be different, you can tell the compiler if you want to allow it to do this or not.

If you want to allow it:

    #pragma STDC FP_CONTRACT ON

and to disallow it:

    #pragma STDC FP_CONTRACT OFF

If you do this at global scope, it stays at whatever state you set it to until you change it.

If you do it at block scope, it reverts to the value outside the block when the block ends.

The initial value of the FP_CONTRACT pragma varies from system to system.

42.6 `fpclassify()`

Return the classification of a given floating point number.

Synopsis

    #include <math.h>
    
    int fpclassify(any_floating_type x);

Description

What kind of entity does this floating point number represent? What are the options?

We’re used to floating point numbers being regular old things like 3.14 or 3490.0001.

But floating point numbers can also represent things like infinity. Or Not-A-Number (NAN). This function will let you know which type of floating point number the argument is.

This is a macro, so you can use it with float, double, long double or anything similar.

Return Value

Returns one of these macros depending on the argument’s classification:

Classification	Description
`FP_INFINITE`	Number is infinite.
`FP_NAN`	Number is Not-A-Number (NAN).
`FP_NORMAL`	Just a regular number.
`FP_SUBNORMAL`	Number is a sub-normal number.
`FP_ZERO`	Number is zero.

A discussion of subnormal numbers is beyond the scope of the guide, and is something that most devs go their whole lives without dealing with. In a nutshell, it’s a way to represent really small numbers that might normally round down to zero. If you want to know more, see the Wikipedia page on denormal numbers ¹⁸³.

Example

Print various number classifications.

#include <stdio.h>
#include <math.h>

const char *get_classification(double n)
{
    switch (fpclassify(n)) {
        case FP_INFINITE: return "infinity";
        case FP_NAN: return "not a number";
        case FP_NORMAL: return "normal";
        case FP_SUBNORMAL: return "subnormal";
        case FP_ZERO: return "zero";
    }

    return "unknown";
}
 
int main(void)
{
    printf("    1.23: %s\n", get_classification(1.23));
    printf("     0.0: %s\n", get_classification(0.0));
    printf("sqrt(-1): %s\n", get_classification(sqrt(-1)));
    printf("1/tan(0): %s\n", get_classification(1/tan(0)));
    printf("  1e-310: %s\n", get_classification(1e-310));  // very small!
}

Output¹⁸⁴:

        1.23: normal
         0.0: zero
    sqrt(-1): not a number
    1/tan(0): infinity
      1e-310: subnormal

42.7 `isfinite()`, `isinf()`, `isnan()`, `isnormal()`

Return true if a number matches a classification.

Synopsis

    #include <math.h>
    
    int isfinite(any_floating_type x);
    
    int isinf(any_floating_type x);
    
    int isnan(any_floating_type x);
    
    int isnormal(any_floating_type x);

Description

These are helper macros to fpclassify(). Bring macros, they work on any floating point type.

Macro	Description
`isfinite()`	True if the number is not infinite or NaN.
`isinf()`	True if the number is infinite.
`isnan()`	True if the number is Not-a-Number.
`isnormal()`	True if the number is normal.

For more superficial discussion on normal and subnormal numbers, see fpclassify().

Return Value

Returns non-zero for true, and zero for false.

Example

#include <stdio.h>
#include <math.h>

int main(void)
{
    printf("  isfinite(1.23): %d\n", isfinite(1.23));    // 1
    printf(" isinf(1/tan(0)): %d\n", isinf(1/tan(0)));   // 1
    printf(" isnan(sqrt(-1)): %d\n", isnan(sqrt(-1)));   // 1
    printf("isnormal(1e-310): %d\n", isnormal(1e-310));  // 0
}

42.8 `signbit()`

Return the sign of a number.

Synopsis

    #include <math.h>
    
    int signbit(any_floating_type x);

Description

This macro takes any floating point number and returns a value indicating the sign of the number, positive or negative.

Return Value

Returns 1 if the sign is negative, otherwise 0.

Example

printf("%d\n", signbit(3490.0));  // 0
printf("%d\n", signbit(-37.0));   // 1

42.9 `acos()`, `acosf()`, `acosl()`

Calculate the arc cosine of a number.

Synopsis

    #include <math.h>
    
    double acos(double x);
    float acosf(float x);
    long double acosl(long double x);

Description

Calculates the arc cosine of a number in radians. (That is, the value whose cosine is x.) The number must be in the range -1.0 to 1.0.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

Returns the arc cosine of x, unless x is out of range. In that case, errno will be set to EDOM and the return value will be NaN. The variants return different types.

Example

double acosx;
long double ldacosx;

acosx = acos(0.2);
ldacosx = acosl((long double)0.3);

42.10 `asin()`, `asinf()`, `asinl()`

Calculate the arc sine of a number.

Synopsis

    #include <math.h>
    
    double asin(double x);
    float asinf(float x);
    long double asinl(long double x);

Description

Calculates the arc sine of a number in radians. (That is, the value whose sine is x.) The number must be in the range -1.0 to 1.0.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

Returns the arc sine of x, unless x is out of range. In that case, errno will be set to EDOM and the return value will be NaN. The variants return different types.

Example

double asinx;
long double ldasinx;

asinx = asin(0.2);
ldasinx = asinl((long double)0.3);

42.11 `atan()`, `atanf()`, `atanl()`, `atan2()`, `atan2f()`, `atan2l()`

Calculate the arc tangent of a number.

Synopsis

    #include <math.h>
    
    double atan(double x);
    float atanf(float x);
    long double atanl(long double x);
    
    double atan2(double y, double x);
    float atan2f(float y, float x);
    long double atan2l(long double y, long double x);

Description

Calculates the arc tangent of a number in radians. (That is, the value whose tangent is x.)

The atan2() variants are pretty much the same as using atan() with y/x as the argument…except that atan2() will use those values to determine the correct quadrant of the result.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

The atan() functions return the arc tangent of x, which will be between PI/2 and -PI/2. The atan2() functions return an angle between PI and -PI.

Example

double atanx;
long double ldatanx;

atanx = atan(0.2);
ldatanx = atanl((long double)0.3);

atanx = atan2(0.2);
ldatanx = atan2l((long double)0.3);

42.12 `cos()`, `cosf()`, `cosl()`

Calculate the cosine of a number.

Synopsis

    #include <math.h>
    
    double cos(double x)
    float cosf(float x)
    long double cosl(long double x)

Description

Calculates the cosine of the value x, where x is in radians.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

Returns the cosine of x. The variants return different types.

Example

double sinx;
long double ldsinx;

sinx = sin(3490.0); // round and round we go!
ldsinx = sinl((long double)3.490);

42.13 `sin()`, `sinf()`, `sinl()`

Calculate the sine of a number.

Synopsis

    #include <math.h>
    
    double sin(double x);
    float sinf(float x);
    long double sinl(long double x);

Description

Calculates the sine of the value x, where x is in radians.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

Returns the sine of x. The variants return different types.

Example

double sinx;
long double ldsinx;

sinx = sin(3490.0); // round and round we go!
ldsinx = sinl((long double)3.490);

42.14 `tan()`, `tanf()`, `tanl()`

Calculate the tangent of a number.

Synopsis

    #include <math.h>
    
    double tan(double x)
    float tanf(float x)
    long double tanl(long double x)

Description

Calculates the tangent of the value x, where x is in radians.

For those of you who don’t remember, radians are another way of measuring an angle, just like degrees. To convert from degrees to radians or the other way around, use the following code:

    degrees = radians * 180.0f / M_PI;
    radians = degrees * M_PI / 180;

Return Value

Returns the tangent of x. The variants return different types.

Example

double tanx;
long double ldtanx;

tanx = tan(3490.0); // round and round we go!
ldtanx = tanl((long double)3.490);

42.15 `acosh()`, `acoshf()`, `acoshl()`

Compute arc hyperbolic cosine.

Synopsis

    #include <math.h>
    
    double acosh(double x);
    
    float acoshf(float x);
    
    long double acoshl(long double x);

Description

Trig lovers can rejoice! C has arc hyperbolic cosine!

These functions return the nonnegative acosh of x, which must be greater than or equal to 1.

Return Value

Returns the arc hyperbolic cosince in the range $[0,+\infty]$.

Example

printf("acosh 1.8 = %f\n", acosh(1.8));  // 1.192911

42.16 `asinh()`, `asinhf()`, `asinhl()`

Compute arc hyperbolic sine.

Synopsis

    #include <math.h>
    
    double asinh(double x);
    
    float asinhf(float x);
    
    long double asinhl(long double x);

Description

Trig lovers can rejoice! C has arc hyperbolic sine!

These functions return the asinh of x.

Return Value

Returns the arc hyperbolic sine.

Example

printf("acosh 1.8 = %f\n", asinh(1.8));  // 1.350441

42.17 `atanh()`, `atanhf()`, `atanhl()`

Compute the arc hyperbolic tangent.

Synopsis

    #include <math.h>
    
    double atanh(double x);
    
    float atanhf(float x);
    
    long double atanhl(long double x);

Description

These functions compute the arc hyperbolic tangent of x, which must be in the range $[-1,+1]$. Passing exactly $-1$ or $+1$ might result in a pole error.

Return Value

Returns the arc hyperbolic tangent of x.

Example

printf("atanh 0.5 = %f\n", atanh(0.5));  // 0.549306

42.18 `cosh()`, `coshf()`, `coshl()`

Compute the hyperbolic cosine.

Synopsis

    #include <math.h>
    
    double cosh(double x);
    
    float coshf(float x);
    
    long double coshl(long double x);

Description

These functions predictably compute the hyperbolic cosine of x. A range error might occur if x is too large.

Return Value

Returns the hyperbolic cosine of x.

Example

printf("cosh 0.5 = %f\n", cosh(0.5));  // 1.127626

42.19 `tanh()`, `tanhf()`, `tanhl()`

Compute the hyperbolic tangent.

Synopsis

    #include <math.h>
    
    double tanh(double x);
    
    float tanhf(float x);
    
    long double tanhl(long double x);

Description

These functions predictably compute the hyperbolic tangent of x.

Mercifully, this is the last trig-related man page I’m going to write.

Return Value

Returns the hyperbolic tangent of x.

Example

printf("tanh 0.5 = %f\n", tanh(0.5));  // 0.462117

42.20 `exp()`, `expf()`, `expl()`

Compute $e$ raised to a power.

Synopsis

    #include <math.h>
    
    double exp(double x);
    
    float expf(float x);
    
    long double expl(long double x);

Description

Compute $e^x$ where $e$ is Euler’s number ¹⁸⁵.

The number $e$ is named after Leonard Euler, born April 15, 1707, who is responsible, among other things, for making this reference page longer than it needed to be.

Return Value

Returns $e^x$.

Example

printf("exp(1) = %f\n", exp(1));  // 2.718282
printf("exp(2) = %f\n", exp(2));  // 7.389056

42.21 `exp2()`, `exp2f()`, `exp2l()`

Compute 2 to a power.

Synopsis

    #include <math.h>
    
    double exp2(double x);
    
    float exp2f(float x);
    
    long double exp2l(long double x);

Description

These functions raise 2 to a power. Very exciting, since computers are all about twos-to-powers!

These are likely to be faster than using pow() to do the same thing.

They support fractional exponents, as well.

A range error occurs if x is too large.

Return Value

exp2() returns $2^x$.

Example

printf("2^3 = %f\n", exp2(3));      // 2^3 = 8.000000
printf("2^8 = %f\n", exp2(8));      // 2^8 = 256.000000
printf("2^0.5 = %f\n", exp2(0.5));  // 2^0.5 = 1.414214

42.22 `expm1()`, `expm1f()`, `expm1l()`

Compute $e^x-1$.

Synopsis

    #include <math.h>
    
    double expm1(double x);
    
    float expm1f(float x);
    
    long double expm1l(long double x);

Description

This is just like exp() except—plot twist!–it computes that result minus one.

For more discussion about what $e$ is, see the exp() man page.

If x is giant, a range error might occur.

For small values of x near zero, expm1(x) might be more accurate than computing exp(x)-1.

Return Value

Returns $e^x-1$.

Example

printf("%f\n", expm1(2.34));  // 9.381237

42.23 `frexp()`, `frexpf()`, `frexpl()`

Break a number into its fraction part and exponent (as a power of 2).

Synopsis

    #include <math.h>
    
    double frexp(double value, int *exp);
    
    float frexpf(float value, int *exp);
    
    long double frexpl(long double value, int *exp);

Description

If you have a floating point number, you can break it into its fractional part and exponent part (as a power of 2).

For example, if you have the number $1234.56$, this can be represented as a multiple of a power of 2 like so:

$1234.56=0.6028125\times2^{11}$

And you can use this function to get the $0.6028125$ and $11$ parts of that equation.

As for why, I have a simple answer: I don’t know. I can’t find a use. K&R2 and everyone else I can find just says how to use it, but not why you might want to.

The C99 Rationale document says:

The functions frexp, ldexp, and modf are primitives used by the remainder of the library.

There was some sentiment for dropping them for the same reasons that ecvt, fcvt, and gcvt were dropped, but their adherents rescued them for general use. Their use is problematic: on non-binary architectures, ldexp may lose precision and frexp may be inefficient.

So there you have it. If you need it.

Return Value

frexp() returns the fractional part of value in the range 0.5 (inclusive) to 1 (exclusive), or 0. And it stores the exponent power-of-2 in the variable pointed to by exp.

If you pass in zero, the return value and the variable exp points to are both zero.

Example

double frac;
int expt;

frac = frexp(1234.56, &expt);
printf("1234.56 = %.7f x 2^%d\n", frac, expt);

Output:

    1234.56 = 0.6028125 x 2^11

42.24 `ilogb()`, `ilogbf()`, `ilogbl()`

Return the exponent of a floating point number.

Synopsis

    #include <math.h>
    
    int ilogb(double x);
    
    int ilogbf(float x);
    
    int ilogbl(long double x);

Description

This gives you the exponent of the given number… it’s a little weird, because the exponent depends on the value of FLT_RADIX. Now, this is very often 2—but no guarantees!

It actually returns $\log_r|x|$ where $r$ is FLT_RADIX.

Domain or range errors might occur for invalid values of x, or for return values that are outside the range of the return type.

Return Value

The exponent of the absolute value of the given number, depending on FLT_RADIX.

Specifically $\log_r|x|$ where $r$ is FLT_RADIX.

If you pass in 0, it’ll return FP_ILOGB0.

If you pass in infinity, it’ll return INT_MAX.

If you pass in NaN, it’ll return FP_ILOGBNAN.

The spec goes on to say that the value of FP_ILOGB0 will be either INT_MIN or -INT_MAX. And the value of FP_ILOGBNAN shall be either INT_MAX or INT_MIN, if that’s useful in any way.

Example

printf("%d\n", ilogb(257));  // 8
printf("%d\n", ilogb(256));  // 8
printf("%d\n", ilogb(255));  // 7

42.25 `ldexp()`, `ldexpf()`, `ldexpl()`

Multiply a number by an integral power of 2.

Synopsis

    #include <math.h>
    
    double ldexp(double x, int exp);
    
    float ldexpf(float x, int exp);
    
    long double ldexpl(long double x, int exp);

Description

These functions multiply the given number x by 2 raised to the exp power.

Return Value

Returns $x\times2^{exp}$.

Example

printf("1 x 2^10 = %f\n", ldexp(1, 10));
printf("5.67 x 2^7 = %f\n", ldexp(5.67, 7));

Output:

    1 x 2^10 = 1024.000000
    5.67 x 2^7 = 725.760000

42.26 `log()`, `logf()`, `logl()`

Compute the natural logarithm.

Synopsis

    #include <math.h>
    
    double log(double x);
    
    float logf(float x);
    
    long double logl(long double x);

Description

Natural logarithms! And there was much rejoycing.

These compute the base-$e$ logarithm of a number, $\log_ex$, $\ln x$.

In other words, for a given $x$, solves $x=e^y$ for $y$.

Return Value

The base-$e$ logarithm of the given value, $\log_ex$, $\ln x$.

Example

const double e = 2.718281828459045;

printf("%f\n", log(3490.2));  // 8.157714
printf("%f\n", log(e));       // 1.000000

42.27 `log10()`, `log10f()`, `log10l()`

Compute the log-base-10 of a number.

Synopsis

    #include <math.h>
    
    double log10(double x);
    
    float log10f(float x);
    
    long double log10l(long double x);

Description

Just when you thought you might have to use Laws of Logarithms to compute this, here’s a function coming out of the blue to save you.

These compute the base-$10$ logarithm of a number, $\log_{10}x$.

In other words, for a given $x$, solves $x=10^y$ for $y$.

Return Value

Returns the log base-10 of x, $\log_{10}x$.

Example

printf("%f\n", log10(3490.2));   // 3.542850
printf("%f\n", log10(10));       // 1.000000

42.28 `logp1()`, `logp1f()`, `logp1l()`

Compute the natural logarithm of a number plus 1.

Synopsis

    #include <math.h>
    
    double log1p(double x);
    
    float log1pf(float x);
    
    long double log1pl(long double x);

Description

This computes $\log_e(1 + x)$, $\ln(1+x)$.

This works just like calling:

    log(1 + x)

except it could be more accurate for small values of x.

So if your x is small magnitude, use this.

Return Value

Returns $\log_e(1 + x)$, $\ln(1+x)$.

Example

Compute some big and small logarithm values to see the difference between log1p() and log():

printf("Big log1p()  : %.*Lf\n", LDBL_DECIMAL_DIG-1, log1pl(9));
printf("Big log()    : %.*Lf\n", LDBL_DECIMAL_DIG-1, logl(1 + 9));

printf("Small log1p(): %.*Lf\n", LDBL_DECIMAL_DIG-1, log1pl(0.01));
printf("Small log()  : %.*Lf\n", LDBL_DECIMAL_DIG-1, logl(1 + 0.01));

Output on my system:

    Big log1p()  : 2.30258509299404568403
    Big log()    : 2.30258509299404568403
    Small log1p(): 0.00995033085316808305
    Small log()  : 0.00995033085316809164

42.29 `log2()`, `log2f()`, `log2l()`

Compute the base-2 logarithm of a number.

Synopsis

    #include <math.h>
    
    double log2(double x);
    
    float log2f(float x);
    
    long double log2l(long double x);

Description

Wow! Were you thinking we were done with the logarithm functions? We’re only getting started!

This one computes $\log_2 x$. That is, computes $y$ that satisfies $x=2^y$.

Love me those powers of 2!

Return Value

Returns the base-2 logarithm of the given value, $\log_2 x$.

Example

printf("%f\n", log2(3490.2));  // 11.769094
printf("%f\n", log2(256));     // 8.000000

42.30 `logb()`, `logbf()`, `logbl()`

Extract the exponent of a number given FLT_RADIX.

Synopsis

    #include <math.h>
    
    double logb(double x);
    
    float logbf(float x);
    
    long double logbl(long double x);

Description

This function returns the whole number portion of the exponent of the number with radix FLT_RADIX, namely the whole number portion $\log_r|x|$ where $r$ is FLT_RADIX. Fractional numbers are truncated.

If the number is subnormal ¹⁸⁶, logb() treats it as if it were normalized.

If x is 0, there could be a domain error or pole error.

Return Value

This function returns the whole number portion of $\log_r|x|$ where $r$ is FLT_RADIX.

Example

printf("FLT_RADIX = %d\n", FLT_RADIX);
printf("%f\n", logb(3490.2));
printf("%f\n", logb(256));

Output:

    FLT_RADIX = 2
    11.000000
    8.000000

42.31 `modf()`, `modff()`, `modfl()`

Extract the integral and fractional parts of a number.

Synopsis

    #include <math.h>
    
    double modf(double value, double *iptr);
    
    float modff(float value, float *iptr);
    
    long double modfl(long double value, long double *iptr);

Description

If you have a floating point number, like 123.456, this function will extract the integral part (123.0) and the fractional part (0.456). It’s total coincidence that this is exactly the plot for the latest Jason Statham action spectacular.

Both the integral part and fractional parts keep the sign of the passed in value.

The integral part is stored in the address pointed to by iptr.

See the note in frexp() regarding why this is in the library.

Return Value

These functions return the fractional part of the number. The integral part is stored in the address pointed to by iptr. Both the integral and fractional parts preserve the sign of the passed-in value.

Example

#include <stdio.h>
#include <math.h>

void print_parts(double x)
{
    double i, f;

    f = modf(x, &i);

    printf("Entire number  : %f\n", x);
    printf("Integral part  : %f\n", i);
    printf("Fractional part: %f\n\n", f);
}

int main(void)
{
    print_parts(123.456);
    print_parts(-123.456);
}

Output:

    Entire number  : 123.456000
    Integral part  : 123.000000
    Fractional part: 0.456000
    
    Entire number  : -123.456000
    Integral part  : -123.000000
    Fractional part: -0.456000

42.32 `scalbn()`, `scalbnf()`, `scalbnl()` `scalbln()`, `scalblnf()`, `scalblnl()`

Efficiently compute $x\times r^n$, where $r$ is FLT_RADIX.

Synopsis

    #include <math.h>
    
    double scalbn(double x, int n);
    
    float scalbnf(float x, int n);
    
    long double scalbnl(long double x, int n);
    
    double scalbln(double x, long int n);
    
    float scalblnf(float x, long int n);
    
    long double scalblnl(long double x, long int n);

Description

These functions efficiently compute $x\times r^n$, where $r$ is FLT_RADIX.

If FLT_RADIX happens to be 2 (no guarantees!), then this works like exp2().

The name of this function should have an obvious meaning to you. Clearly they all start with the prefix “scalb” which means…

…OK, I confess! I have no idea what it means. My searches are futile!

But let’s look at the suffixes:

Suffix	Meaning
`n`	`scalbn()`—exponent `n` is an `int`
`nf`	`scalbnf()`—`float` version of `scalbn()`
`nl`	`scalbnl()`—`long double` version of `scalbn()`
`ln`	`scalbln()`—exponent `n` is a `long int`
`lnf`	`scalblnf()`—`float` version of `scalbln()`
`lnl`	`scalblnl()`—`long double` version of `scalbln()`

So while I’m still in the dark about “scalb”, at least I have that part down.

A range error might occur for large values.

Return Value

Returns $x\times r^n$, where $r$ is FLT_RADIX.

Example

#include <stdio.h>
#include <math.h>
#include <float.h>

int main(void)
{
    printf("FLT_RADIX = %d\n\n", FLT_RADIX);
    printf("scalbn(3, 8)       = %f\n", scalbn(2, 8));
    printf("scalbn(10.2, 20.7) = %f\n", scalbn(10.2, 20.7));
}

Output on my system:

    FLT_RADIX = 2
    
    scalbn(3, 8)       = 512.000000
    scalbn(10.2, 20.7) = 10695475.200000

42.33 `cbrt()`, `cbrtf()`, `cbrtl()`

Compute the cube root.

Synopsis

    #include <math.h>
    
    double cbrt(double x);
    
    float cbrtf(float x);
    
    long double cbrtl(long double x);

Description

Computes the cube root of x, $x^{1/3}$, $\sqrt[3]{x}$.

Return Value

Returns the cube root of x, $x^{1/3}$, $\sqrt[3]{x}$.

Example

printf("cbrt(1729.03) = %f\n", cbrt(1729.03));

Output:

    cbrt(1729.03) = 12.002384

42.34 `fabs()`, `fabsf()`, `fabsl()`

Compute the absolute value.

Synopsis

    #include <math.h>
    
    double fabs(double x);
    
    float fabsf(float x);
    
    long double fabsl(long double x);

Description

These functions straightforwardly return the absolute value of x, that is $|x|$.

If you’re rusty on your absolute values, all it means is that the result will be positive, even if x is negative. It’s just strips negative signs off.

Return Value

Returns the absolute value of x, $|x|$.

Example

printf("fabs(3490.0)  = %f\n", fabs(3490.0));  // 3490.000000
printf("fabs(-3490.0) = %f\n", fabs(3490.0));  // 3490.000000

42.35 `hypot()`, `hypotf()`, `hypotl()`

Compute the length of the hypotenuse of a triangle.

Synopsis

    #include <math.h>
    
    double hypot(double x, double y);
    
    float hypotf(float x, float y);
    
    long double hypotl(long double x, long double y);

Description

Pythagorean Theorem ¹⁸⁷ fans rejoice! This is the function you’ve been waiting for!

If you know the lengths of the two sides of a right triangle, x and y, you can compute the length of the hypotenuse (the longest, diagonal side) with this function.

In particular, it computes the square root of the sum of the squares of the sides: $\sqrt{x^2 + y^2}$.

Return Value

Returns the lenght of the hypotenuse of a right triangle with side lengths x and y: $\sqrt{x^2 + y^2}$.

Example

printf("%f\n", hypot(3, 4));  // 5.000000

42.36 `pow()`, `powf()`, `powl()`

Compute a value raised to a power.

Synopsis

    #include <math.h>
    
    double pow(double x, double y);
    
    float powf(float x, float y);
    
    long double powl(long double x, long double y);

Description

Computes x raised to the yth power: $x^y$.

These arguments can be fractional.

Return Value

Returns x raised to the yth power: $x^y$.

A domain error can occur if:

x is a finite negative number and y is a finite non-integer
x is zero and y is zero.

A domain error or pole error can occur if x is zero and y is negative.

A range error can occur for large values.

Example

printf("%f\n", pow(3, 4));    // 3^4    = 81.000000
printf("%f\n", pow(2, 0.5));  // sqrt 2 = 1.414214

42.37 `sqrt()`

Calculate the square root of a number.

Synopsis

    #include <math.h>
    
    double sqrt(double x);
    
    float sqrtf(float x);
    
    long double sqrtl(long double x);

Description

Computes the square root of a number: $\sqrt{x}$. To those of you who don’t know what a square root is, I’m not going to explain. Suffice it to say, the square root of a number delivers a value that when squared (multiplied by itself) results in the original number.

Ok, fine—I did explain it after all, but only because I wanted to show off. It’s not like I’m giving you examples or anything, such as the square root of nine is three, because when you multiply three by three you get nine, or anything like that. No examples. I hate examples!

And I suppose you wanted some actual practical information here as well. You can see the usual trio of functions here—they all compute square root, but they take different types as arguments. Pretty straightforward, really.

A domain error occurs if x is negative.

Return Value

Returns (and I know this must be something of a surprise to you) the square root of x: $\sqrt{x}$.

Example

// example usage of sqrt()

float something = 10;

double x1 = 8.2, y1 = -5.4;
double x2 = 3.8, y2 = 34.9;
double dx, dy;

printf("square root of 10 is %.2f\n", sqrtf(something));

dx = x2 - x1;
dy = y2 - y1;
printf("distance between points (x1, y1) and (x2, y2): %.2f\n",
    sqrt(dx*dx + dy*dy));

And the output is:

    square root of 10 is 3.16
    distance between points (x1, y1) and (x2, y2): 40.54

42.38 `erf()`, `erff()`, `erfl()`

Compute the error function of the given value.

Synopsis

    #include <math.h>
    
    double erfc(double x);
    
    float erfcf(float x);
    
    long double erfcl(long double x);

Description

These functions compute the error function ¹⁸⁸ of a value.

Return Value

Returns the error function of x:

${\displaystyle \frac{2}{\sqrt\pi} \int_0^x e^{-t^2}\,dt}$

Example

for (float i = -2; i <= 2; i += 0.5)
    printf("% .1f: %f\n", i, erf(i));

Output:

    -2.0: -0.995322
    -1.5: -0.966105
    -1.0: -0.842701
    -0.5: -0.520500
     0.0: 0.000000
     0.5: 0.520500
     1.0: 0.842701
     1.5: 0.966105
     2.0: 0.995322

42.39 `erfc()`, `erfcf()`, `erfcl()`

Compute the complementary error function of a value.

Synopsis

    #include <math.h>
    
    double erfc(double x);
    
    float erfcf(float x);
    
    long double erfcl(long double x);

Description

These functions compute the complementary error function ¹⁸⁹ of a value.

This is the same as:

    1 - erf(x)

A range error can occur if x is too large.

Return Value

Returns 1 - erf(x), namely:

${\displaystyle \frac{2}{\sqrt\pi} \int_x^{\infty} e^{-t^2}\,dt}$

Example

for (float i = -2; i <= 2; i += 0.5)
    printf("% .1f: %f\n", i, erfc(i));

Output:

    -2.0: 1.995322
    -1.5: 1.966105
    -1.0: 1.842701
    -0.5: 1.520500
     0.0: 1.000000
     0.5: 0.479500
     1.0: 0.157299
     1.5: 0.033895
     2.0: 0.004678

42.40 `lgamma()`, `lgammaf()`, `lgammal()`

Compute the natural logarithm of the absolute value of $\Gamma(x)$.

Synopsis

    #include <math.h>
    
    double lgamma(double x);
    
    float lgammaf(float x);
    
    long double lgammal(long double x);

Description

Compute the natural log of the absolute value of gamma ¹⁹⁰ x, $\log_e|\Gamma(x)|$.

A range error can occur if x is too large.

A pole error can occur is x is non-positive.

Return Value

Returns $\log_e|\Gamma(x)|$.

Example

for (float i = 0.5; i <= 4; i += 0.5)
    printf("%.1f: %f\n", i, lgamma(i));

Output:

    0.5: 0.572365
    1.0: 0.000000
    1.5: -0.120782
    2.0: 0.000000
    2.5: 0.284683
    3.0: 0.693147
    3.5: 1.200974
    4.0: 1.791759

42.41 `tgamma()`, `tgammaf()`, `tgammal()`

Compute the gamma function, $\Gamma(x)$.

Synopsis

    #include <math.h>
    
    double tgamma(double x);
    
    float tgammaf(float x);
    
    long double tgammal(long double x);

Description

Computes the gamma function ¹⁹¹ of x, $\Gamma(x)$.

A domain or pole error might occur if x is non-positive.

A range error might occur if x is too large or too small.

Return Value

Returns the gamma function of x, $\Gamma(x)$.

Example

for (float i = 0.5; i <= 4; i += 0.5)
    printf("%.1f: %f\n", i, tgamma(i));

Output:

    0.5: 1.772454
    1.0: 1.000000
    1.5: 0.886227
    2.0: 1.000000
    2.5: 1.329340
    3.0: 2.000000
    3.5: 3.323351
    4.0: 6.000000

42.42 `ceil()`, `ceilf()`, `ceill()`

Ceiling—return the next whole number not smaller than the given number.

Synopsis

    #include <math.h>
    
    double ceil(double x);
    
    float ceilf(float x);
    
    long double ceill(long double x);

Description

Returns the ceiling of the x: $\lceil{x}\rceil$.

This is the next whole number not smaller than x.

Beware this minor dragon: it’s not just “rounding up”. Well, it is for positive numbers, but negative numbers effectively round toward zero. (Because the ceiling function is headed for the next largest whole number and $-4$ is larger than $-5$.)

Return Value

Returns the next largest whole number larger than x.

Example

Notice for the negative numbers it heads toward zero, i.e. toward the next largest whole number—just like the positives head toward the next largest whole number.

printf("%f\n", ceil(4.0));   //  4.000000
printf("%f\n", ceil(4.1));   //  5.000000
printf("%f\n", ceil(-2.0));  // -2.000000
printf("%f\n", ceil(-2.1));  // -2.000000
printf("%f\n", ceil(-3.1));  // -3.000000

42.43 `floor()`, `floorf()`, `floorl()`

Compute the largest whole number not larger than the given value.

Synopsis

    #include <math.h>
    double floor(double x);
    float floorf(float x);
    long double floorl(long double x);

Description

Returns the floor of the value: $\lfloor{x}\rfloor$. This is the opposite of ceil().

This is the largest whole number that is not greater than x.

For positive numbers, this is like rounding down: 4.5 becomes 4.0.

For negative numbers, it’s like rounding up: -3.6 becomes -4.0.

In both cases, those results are the largest whole number not bigger than the given number.

Return Value

Returns the largest whole number not greater than x: $\lfloor{x}\rfloor$.

Example

Note how the negative numbers effectively round away from zero, unlike the positives.

printf("%f\n", floor(4.0));   //  4.000000
printf("%f\n", floor(4.1));   //  4.000000
printf("%f\n", floor(-2.0));  // -2.000000
printf("%f\n", floor(-2.1));  // -3.000000
printf("%f\n", floor(-3.1));  // -4.000000

42.44 `nearbyint()`, `nearbyint()`, `nearbyint()`

Rounds a value in the current rounding direction.

Synopsis

    #include <math.h>
    
    double nearbyint(double x);
    
    float nearbyintf(float x);
    
    long double nearbyintl(long double x);

Description

This function rounds x to the nearest integer in the current rounding direction.

The rounding direction can be set with fesetround() in <fenv.h>.

nearbyint() won’t raise the “inexact” floating point exception.

Return Value

Returns x rounded in the current rounding direction.

Example

#include <stdio.h>
#include <math.h>
#include <fenv.h>

int main(void)
{
    #pragma STDC FENV_ACCESS ON        // If supported

    fesetround(FE_TONEAREST);          // round to nearest

    printf("%f\n", nearbyint(3.14));   // 3.000000
    printf("%f\n", nearbyint(3.74));   // 4.000000

    fesetround(FE_TOWARDZERO);         // round toward zero

    printf("%f\n", nearbyint(1.99));   // 1.000000
    printf("%f\n", nearbyint(-1.99));  // -1.000000
}

42.45 `rint()`, `rintf()`, `rintl()`

Rounds a value in the current rounding direction.

Synopsis

    #include <math.h>
    
    double rint(double x);
    
    float rintf(float x);
    
    long double rintl(long double x);

Description

This works just like nearbyint() except that is can raise the “inexact” floating point exception.

Return Value

Returns x rounded in the current rounding direction.

Example

#include <stdio.h>
#include <math.h>
#include <fenv.h>

int main(void)
{
    #pragma STDC FENV_ACCESS ON

    fesetround(FE_TONEAREST);

    printf("%f\n", rint(3.14));   // 3.000000
    printf("%f\n", rint(3.74));   // 4.000000

    fesetround(FE_TOWARDZERO);

    printf("%f\n", rint(1.99));   // 1.000000
    printf("%f\n", rint(-1.99));  // -1.000000
}

42.46 `lrint()`, `lrintf()`, `lrintl()` `llrint()`, `llrintf()`, `llrintl()`

Returns x rounded in the current rounding direction as an integer.

Synopsis

    #include <math.h>
    
    long int lrint(double x);
    long int lrintf(float x);
    long int lrintl(long double x);
    
    long long int llrint(double x);
    long long int llrintf(float x);
    long long int llrintl(long double x);

Description

Round a floating point number in the current rounding direction, but this time return an integer intead of a float. You know, just to mix it up.

These come in two variants:

lrint()—returns long int
llrint()—returns long long int

If the result doesn’t fit in the return type, a domain or range error might occur.

Return Value

The value of x rounded to an integer in the current rounding direction.

Example

#include <stdio.h>
#include <math.h>
#include <fenv.h>

int main(void)
{
    #pragma STDC FENV_ACCESS ON

    fesetround(FE_TONEAREST);

    printf("%ld\n", lrint(3.14));   // 3
    printf("%ld\n", lrint(3.74));   // 4

    fesetround(FE_TOWARDZERO);

    printf("%ld\n", lrint(1.99));   // 1
    printf("%ld\n", lrint(-1.99));  // -1
}

42.47 `round()`, `roundf()`, `roundl()`

Round a number in the good old-fashioned way.

Synopsis

    #include <math.h>
    
    double round(double x);
    
    float roundf(float x);
    
    long double roundl(long double x);

Description

Rounds a number to the nearest whole value.

In case of halfsies, rounds away from zero (i.e. “round up” in magnitude).

The current rounding direction’s Jedi mind tricks don’t work on this function.

Return Value

The rounded value of x.

Example

printf("%f\n", round(3.14));   // 3.000000
printf("%f\n", round(3.5));    // 4.000000

printf("%f\n", round(-1.5));   // -2.000000
printf("%f\n", round(-1.14));  // -1.000000

42.48 `lround()`, `lroundf()`, `lroundl()` `llround()`, `llroundf()`, `llroundl()`

Round a number in the good old-fashioned way, returning an integer.

Synopsis

    #include <math.h>
    
    long int lround(double x);
    long int lroundf(float x);
    long int lroundl(long double x);
    
    long long int llround(double x);
    long long int llroundf(float x);
    long long int llroundl(long double x);

Description

These are just like round() except they return integers.

Halfway values round away from zero, e.g. $1.5$ rounds to $2$ and $-1.5$ rounds to $-2$.

The functions are grouped by return type:

lround()—returns a long int
llround()—returns a long long int

If the rounded value can’t find in the return time, a domain or range error can occur.

Return Value

Returns the rounded value of x as an integer.

Example

printf("%ld\n", lround(3.14));   // 3
printf("%ld\n", lround(3.5));    // 4

printf("%ld\n", lround(-1.5));   // -2
printf("%ld\n", lround(-1.14));  // -1

42.49 `trunc()`, `truncf()`, `truncl()`

Truncate the fractional part off a floating point value.

Synopsis

    #include <math.h>
    
    double trunc(double x);
    
    float truncf(float x);
    
    long double truncl(long double x);

Description

These functions just drop the fractional part of a floating point number. Boom.

In other words, they always round toward zero.

Return Value

Returns the truncated floating point number.

Example

printf("%f\n", trunc(3.14));   // 3.000000
printf("%f\n", trunc(3.8));    // 3.000000

printf("%f\n", trunc(-1.5));   // -1.000000
printf("%f\n", trunc(-1.14));  // -1.000000

42.50 `fmod()`, `fmodf()`, `fmodl()`

Compute the floating point remainder.

Synopsis

    #include <math.h>
    
    double fmod(double x, double y);
    
    float fmodf(float x, float y);
    
    long double fmodl(long double x, long double y);

Description

Returns the remainder of $\frac{x}{y}$. The result will have the same sign as x.

Under the hood, the computation performed is:

    x - trunc(x / y) * y

But it might be easier just to think of the remainder.

Return Value

Returns the remainder of $\frac{x}{y}$ with the same sign as x.

Example

printf("%f\n", fmod(-9.2, 5.1));  // -4.100000
printf("%f\n", fmod(9.2, 5.1));   //  4.100000

42.51 `remainder()`, `remainderf()`, `remainderl()`

Compute the remainder IEC 60559-style.

Synopsis

    #include <math.h>
    
    double remainder(double x, double y);
    
    float remainderf(float x, float y);
    
    long double remainderl(long double x, long double y);

Description

This is similar to fmod(), but not quite the same. fmod() is probably what you’re after if you’re expecting remainders to wrap around like an odometer.

The C spec quotes IEC 60559 on how this works:

When $y\neq0$, the remainder $r=x$ REM $y$ is defined regardless of the rounding mode by the mathematical relation $r=x-ny$, where $n$ is the integer nearest the exact value of $x/y$; whenever $|n-x/y|=1/2$, then $n$ is even. If $r=0$, its sign shall be that of $x$.

Hope that clears it up!

OK, maybe not. Here’s the upshot:

You know how if you fmod() something by, say 2.0 you get a result that is somewhere between 0.0 and 2.0? And how if you just increase the number that you’re modding by 2.0, you can see the result climb up to 2.0 and then wrap around to 0.0 like your car’s odometer?

remainder() works just like that, except if y is 2.0, it wraps from -1.0 to 1.0 instead of from 0.0 to 2.0.

In other words, the range of the function runs from -y/2 to y/2. Contrasted to fmod() that runs from 0.0 to y, remainder()’s output is just shifted down half a y.

And zero-remainder-anything is 0.

Except if y is zero, the function might return zero or a domain error might occur.

Return Value

The IEC 60559 result of x-remainder-y.

Example

printf("%f\n", remainder(3.7, 4));  // -0.300000
printf("%f\n", remainder(4.3, 4));  //  0.300000

42.52 `remquo()`, `remquof()`, `remquol()`

Compute the remainder and (some of the) quotient.

Synopsis

    #include <math.h>
    
    double remquo(double x, double y, int *quo);
    
    float remquof(float x, float y, int *quo);
    
    long double remquol(long double x, long double y, int *quo);

Description

This is a funky little thing.

First of all, the return value is the remainder, the same as the remainder() function, so check that out.

And the quotient comes back in the quo pointer.

Or at least some of it does. You’ll get at least 3 bits worth of the quotient.

But why?

So a couple things.

One is that the quotient of some very large floating point numbers can easily be far too gigantic to fit in even a long long unsigned int. So some of it might very well need to be lopped off, anyway.

But at 3 bits? How’s that even useful? That only gets you from 0 to 7!

The C99 Rationale document states:

The remquo functions are intended for implementing argument reductions which can exploit a few low-order bits of the quotient. Note that $x$ may be so large in magnitude relative to $y$ that an exact representation of the quotient is not practical.

So… implementing argument reductions… which can exploit a few low-order bits… Ooookay.

CPPReference has this to say ¹⁹² on the matter, which is spoken so well, I will quote wholesale:

This function is useful when implementing periodic functions with the period exactly representable as a floating-point value: when calculating $\sin(πx)$ for a very large x, calling sin directly may result in a large error, but if the function argument is first reduced with remquo, the low-order bits of the quotient may be used to determine the sign and the octant of the result within the period, while the remainder may be used to calculate the value with high precision.

And there you have it. If you have another example that works for you… congratulations! :)

Return Value

Returns the same as remainder: The IEC 60559 result of x-remainder-y.

In addition, at least the lowest 3 bits of the quotient will be stored in quo with the same sign as x/y.

Example

There’s a great cos() example at CPPReference ¹⁹³ that covers a genuine use case.

But instead of stealing it, I’ll just post a simple example here and you can visit their site for a real one.

int quo;
double rem;

rem = remquo(12.75, 2.25, &quo);

printf("%d remainder %f\n", quo, rem);  // 6 remainder -0.750000

42.53 `copysign()`, `copysignf()`, `copysignl()`

Copy the sign of one value into another.

Synopsis

    #include <math.h>
    
    double copysign(double x, double y);
    
    float copysignf(float x, float y);
    
    long double copysignl(long double x, long double y);

Description

These functions return a number that has the magnitude of x and the sign of y. You can use them to coerce the sign to that of another value.

Neither x nor y are modified, of course. The return value holds the result.

Return Value

Returns a value with the magnitude of x and the sign of y.

Example

double x = 34.9;
double y = -999.9;
double z = 123.4;

printf("%f\n", copysign(x, y)); // -34.900000
printf("%f\n", copysign(x, z)); //  34.900000

42.54 `nan()`, `nanf()`, `nanl()`

Return NAN.

Synopsis

    #include <math.h>
    
    double nan(const char *tagp);
    
    float nanf(const char *tagp);
    
    long double nanl(const char *tagp);

Description

These functions return a quiet NaN¹⁹⁴. It is produced as if calling strtod() with "NAN" (or a variant thereof) as an argument.

tagp points to a string which could be several things, including empty. The contents of the string determine which variant of NaN might get returned depending on the implementation.

Which version of NaN? Did you even know it was possible to get this far into the weeds with something that wasn’t a number?

Case 1 in which you pass in an empty string, in which case these are the same:

    nan("");
    
    strtod("NAN()", NULL);

Case 2 in which the string contains only digits 0-9, letters a-z, letters A-Z, and/or underscore:

    nan("goats");
    
    strtod("NAN(goats)", NULL);

And Case 3, in which the string contains anything else and is ignored:

    nan("!");
    
    strtod("NAN", NULL);

As for what strtod() does with those values in parens, see the [strtod()] reference page. Spoiler: it’s implementation-defined.

Return Value

Returns the requested quiet NaN, or 0 if such things aren’t supported by your system.

Example

printf("%f\n", nan(""));       // nan
printf("%f\n", nan("goats"));  // nan
printf("%f\n", nan("!"));      // nan

42.55 `nextafter()`, `nextafterf()`, `nextafterl()`

Get the next (or previous) representable floating point value.

Synopsis

    #include <math.h>
    
    double nextafter(double x, double y);
    
    float nextafterf(float x, float y);
    
    long double nextafterl(long double x, long double y);

Description

As you probably know, floating point numbers can’t represent every possible real number. There are limits.

And, as such, there exists a “next” and “previous” number after or before any floating point number.

These functions return the next (or previous) representable number. That is, no floating point numbers exist between the given number and the next one.

The way it figures it out is it works from x in the direction of y, answering the question of "what is the next representable number from x as we head toward y.

Return Value

Returns the next representable floating point value from x in the direction of y.

If x equals y, returns y. And also x, I suppose.

Example

printf("%.*f\n", DBL_DECIMAL_DIG, nextafter(0.5, 1.0));
printf("%.*f\n", DBL_DECIMAL_DIG, nextafter(0.349, 0.0));

Output on my system:

    0.50000000000000011
    0.34899999999999992

42.56 `nextoward()`, `nextowardf()`, `nextowardl()`

Get the next (or previous) representable floating point value.

Synopsis

    include <math.h>
    
    double nexttoward(double x, long double y);
    
    float nexttowardf(float x, long double y);
    
    long double nexttowardl(long double x, long double y);

Description

These functions are the same as nextafter() except the second parameter is always long double.

Return Value

Returns the same as nextafter() except if x equals y, returns y cast to the function’s return type.

Example

printf("%.*f\n", DBL_DECIMAL_DIG, nexttoward(0.5, 1.0));
printf("%.*f\n", DBL_DECIMAL_DIG, nexttoward(0.349, 0.0));

Output on my system:

    0.50000000000000011
    0.34899999999999992

42.57 `fdim()`, `fdimf()`, `fdiml()`

Return the positive difference between two numbers clamped at 0.

Synopsis

    #include <math.h>
    
    double fdim(double x, double y);
    
    float fdimf(float x, float y);
    
    long double fdiml(long double x, long double y);

Description

The positive difference between x and y is the difference… except if the difference is less than 0, it’s clamped to 0.

These functions might throw a range error.

Return Value

Returns the difference of x-y if the difference is greater than 0. Otherwise it returns 0.

Example

printf("%f\n", fdim(10.0, 3.0));   // 7.000000
printf("%f\n", fdim(3.0, 10.0));   // 0.000000, clamped

42.58 `fmax()`, `fmax()`, `fmax()`, `fmin()`, `fmin()`, `fmin()`

Return the maximum or minimum of two numbers.

Synopsis

    #include <math.h>
    
    double fmax(double x, double y);
    
    float fmaxf(float x, float y);
    
    long double fmaxl(long double x, long double y);
    
    double fmin(double x, double y);
    
    float fminf(float x, float y);
    
    long double fminl(long double x, long double y);

Description

Straightforwardly, these functions return the minimum or maximum of two given numbers.

If one of the numbers is NaN, the functions return the non-NaN number. If both arguments are NaN, the functions return NaN.

Return Value

Returns the minimum or maximum values, with NaN handled as mentioned above.

Example

printf("%f\n", fmin(10.0, 3.0));   //  3.000000
printf("%f\n", fmax(3.0, 10.0));   // 10.000000

42.59 `fma()`, `fmaf()`, `fmal()`

Floating (AKA “Fast”) multiply and add.

Synopsis

    #include <math.h>
    
    double fma(double x, double y, double z);
    
    float fmaf(float x, float y, float z);
    
    long double fmal(long double x, long double y, long double z);

Description

This performs the operation $(x\times{y})+z$, but does so in a nifty way. It does the computation as if it had infinite precision, and then rounds the final result to the final data type according to the current rounding mode.

Contrast to if you’d do the math yourself, where it would have rounded each step of the way, potentially.

Also some architectures have a CPU instruction to do exactly this calculation, so it can do it super quick. (If it doesn’t, it’s considerably slower.)

You can tell if your CPU supports the fast version by checking that the macro FP_FAST_FMA is set to 1. (The float and long variants of fma() can be tested with FP_FAST_FMAF and FP_FAST_FMAL, respectively.)

These functions might cause a range error to occur.

Return Value

Returns (x * y) + z.

Example

printf("%f\n", fma(1.0, 2.0, 3.0));  // 5.000000

42.60 `isgreater()`, `isgreaterequal()`, `isless()`, `islessequal()`

Floating point comparison macros.

Synopsis

    #include <math.h>
    
    int isgreater(any_floating_type x, any_floating_type y);
    
    int isgreaterequal(any_floating_type x, any_floating_type y);
    
    int isless(any_floating_type x, any_floating_type y);
    
    int islessequal(any_floating_type x, any_floating_type y);

Description

These macros compare floating point numbers. Being macros, we can pass in any floating point type.

You might think you can already do that with just regular comparison operators—and you’d be right!

One one exception: the comparison operators raise the “invalid” floating exception if one or more of the operands is NaN. These macros do not.

Note that you must only pass floating point types into these functions. Passing an integer or any other type is undefined behavior.

Return Value

isgreater() returns the result of x > y.

isgreaterequal() returns the result of x >= y.

isless() returns the result of x < y.

islessequal() returns the result of x <= y.

Example

printf("%d\n", isgreater(10.0, 3.0));        // 1
printf("%d\n", isgreaterequal(10.0, 10.0));  // 1
printf("%d\n", isless(10.0, 3.0));           // 0
printf("%d\n", islessequal(10.0, 3.0));      // 0

42.61 `islessgreater()`

Test if a floating point number is less than or greater than another.

Synopsis

    #include <math.h>
    
    int islessgreater(any_floating_type x, any_floating_type y);

Description

This macro is similar to isgreater() and all those, except it made the section name too long if I included it up there. So it gets its own spot.

This returns true if $x < y$ or $x > y$.

Even though it’s a macro, we can rest assured that x and y are only evaluated once.

And even if x or y are NaN, this will not throw an “invalid” exception, unlike the normal comparison operators.

If you pass in a non-floating type, the behavior is undefined.

Return Value

Returns (x < y) || (x > y).

Example

printf("%d\n", islessgreater(10.0, 3.0));   // 1
printf("%d\n", islessgreater(10.0, 30.0));  // 1
printf("%d\n", islessgreater(10.0, 10.0));  // 0

42.62 `isunordered()`

Macro returns true if either floating point argument is NaN.

Synopsis

    #include <math.h>
    
    int isunordered(any_floating_type x, any_floating_type y);

Description

The spec writes:

The isunordered macro determines whether its arguments are unordered.

See? Told you C was easy!

It does also elaborate that the arguments are unordered if one or both of them are NaN.

Return Value

This macro returns true if one or both of the arguments are NaN.

Example

printf("%d\n", isunordered(1.0, 2.0));       // 0
printf("%d\n", isunordered(1.0, sqrt(-1)));  // 1
printf("%d\n", isunordered(NAN, 30.0));      // 1
printf("%d\n", isunordered(NAN, NAN));       // 1
}

43 `<stdlib.h>` Standard Library Functions

Some of the following functions have variants that handle different types: atoi(), strtod(), strtol(), abs(), and div(). Only a single one is listed here for brevity.

Function	Description
`_Exit()`	Exit the currently-running program and don’t look back
`abort()`	Abruptly end program execution
`abs()`	Compute the absolute value of an integer
`aligned_alloc()`	Allocate specifically-aligned memory
`at_quick_exit()`	Set up handlers to run when the program quickly exits
`atexit()`	Set up handlers to run when the program exits
`atof()`	Convert a string to a floating point value
`atoi()`	Convert an integer in a string into a integer type
`bsearch()`	Binary Search (maybe) an array of objects
`calloc()`	Allocate and clear memory for arbitrary use
`div()`	Compute the quotient and remainder of two numbers
`exit()`	Exit the currently-running program
`free()`	Free a memory region
`getenv()`	Get the value of an environment variable
`malloc()`	Allocate memory for arbitrary use
`mblen()`	Return the number of bytes in a multibyte character
`mbstowcs()`	Convert a multibyte string to a wide character string
`mbtowc()`	Convert a multibyte character to a wide character
`qsort()`	Quicksort (maybe) some data
`quick_exit()`	Exit the currently-running program quickly
`rand()`	Return a pseudorandom number
`realloc()`	Resize a previously allocated stretch of memory
`srand()`	Seed the built-in pseudorandom number generator
`strtod()`	Convert a string to a floating point number
`strtol()`	Convert a string to an integer
`system()`	Run an external program
`wcstombs()`	Convert a wide character string to a multibyte string
`wctomb()`	Convert a wide character to a multibyte character

The <stdlib.h> header has all kinds of—dare I say—miscellaneous functions bundled into it. This functionality includes:

Conversions from numbers to strings
Conversions from strings to numbers
Pseudorandom number generation
Dynamic memory allocation
Various ways to exit the program
Ability to run external programs
Binary search (or some fast search)
Quicksort (or some fast sort)
Integer arithmetic functions
Multibyte and wide character and string conversions

So, you know… a little of everything.

43.1 `<stdlib.h>` Types and Macros

A couple new types and macros are introduced, though some of these might also be defined elsewhere:

Type	Description
`size_t`	Returned from `sizeof` and used elsewhere
`wchar_t`	For wide character operations
`div_t`	For the `div()` function
`ldiv_t`	For the `ldiv()` function
`lldiv_t`	for the `lldiv()` function

And some macros:

Type	Description
`NULL`	Our good pointer friend
`EXIT_SUCCESS`	Good exit status when things go well
`EXIT_FAILURE`	Good exit status when things go poorly
`RAND_MAX`	The maximum value that can be returned by the `rand()` function
`MB_CUR_MAX`	Maximum number of bytes in a multibyte character in the current locale

And there you have it. Just a lot of fun, useful functions in here. Let’s check ’em out!

43.2 `atof()`

Convert a string to a floating point value

Synopsis

    #include <stdlib.h>
    
    double atof(const char *nptr);

Description

This stood for “ASCII-To-Floating” back in the day ¹⁹⁵, but no one would dare to use such coarse language now.

But the gist is the same: we’re going to convert a string with numbers and (optionally) a decimal point into a floating point value. Leading whitespace is ignored, and translation stops at the first invalid character.

If the result doesn’t fit in a double, behavior is undefined.

It generally works as if you’d called strtod():

    strtod(nptr, NULL)

So check out that reference page for more info.

In fact, strtod() is just better and you should probably use that.

Return Value

Returns the string converted to a double.

Example

    double x = atof("3.141593");
    printf("%f\n", x);  // 3.141593

43.3 `atoi()`, `atol()`, `atoll()`

Convert an integer in a string into a integer type

Synopsis

    #include <stdlib.h>
    
    int atoi(const char *nptr);
    
    long int atol(const char *nptr);
    
    long long int atoll(const char *nptr);

Description

Back in the day, atoi() stood for “ASCII-To_Integer”¹⁹⁶ but now the spec makes no mention of that.

These functions take a string with a number in them and convert it to an integer of the specified return type. Leading whitespace is ignored. Translation stops at the first invalid character.

If the result doesn’t fit in the return type, behavior is undefined.

It generally works as if you’d called strtol() family of functions:

    atoi(nptr)                 // is basically the same as...
    (int)strtol(nptr, NULL, 10)
    
    atol(nptr)                 // is basically the same as...
    strtol(nptr, NULL, 10)
    
    atoll(nptr)                // is basically the same as...
    strtoll(nptr, NULL, 10)

Again, the strtol() functions are generally better, so I recommend them instead of these.

Return Value

Returns an integer result corresponding to the return type.

Example

int x = atof("3490");
printf("%d\n", x);  // 3490

43.4 `strtod()`, `strtof()`, `strtold()`

Convert a string to a floating point number

Synopsis

    #include <stdlib.h>
    
    double strtod(const char * restrict nptr, char ** restrict endptr);
    
    float strtof(const char * restrict nptr, char ** restrict endptr);
    
    long double strtold(const char * restrict nptr, char ** restrict endptr);

Description

These are some neat functions that convert strings to floating point numbers (or even NaN or Infinity) and provide some error checking, besides.

Firstly, leading whitespace is skipped.

Then the functions attempt to convert characters into the floating point result. Finally, when an invalid character (or NUL character) is reached, they set endptr to point to the invalid character.

Set endptr to NULL if you don’t care about where the first invalid character is.

If you didn’t set endptr to NULL, it will point to a NUL character if the translation didn’t find any bad characters. That is:

    if (*endptr == '\0') {
        printf("What a perfectly-formed number!\n");
    } else {
        printf("I found badness in your number: \"%s\"\n", endptr);
    }

But guess what! You can also translate strings into special values, like NaN and Infinity!

If nptr points to a string containing INF or INFINITY (upper or lowercase), the value for Infinity will be returned.

If nptr points to a string containing NAN, then (a quiet, non-signalling) NaN will be returned. You can tag the NAN with a sequence of characters from the set 0-9, a-z, A-Z, and _ by enclosing them in parens:

    NAN(foobar_3490)

What your compiler does with this is implementation-defined, but it can be used to specify different kinds of NaN.

You can also specify a number in hexadecimal with a power-of-two exponent ($2^x$) if you lead with 0x (or 0X). For the exponent, use a p followed by a base 10 exponent. (You can’t use e because that’s a valid hex digit!)

Example:

    0xabc.123p15

Which computes to $0xabc.123\times2^{15}$.

You can put in FLT_DECIMAL_DIG, DBL_DECIMAL_DIG, or LDBL_DECIMAL_DIG digits and get a correctly-rounded result for the type.

Return Value

Returns the converted number. If there was no number, returns 0. endptr is set to point to the first invalid character, or the NUL terminator if all characters were consumed.

If there’s an overflow, HUGE_VAL, HUGE_VALF, or HUGE_VALL is returned, signed like the input, and errno is set to ERANGE.

If there’s an underflow, it returns the smallest number closest to zero with the input sign. errno may be set to ERANGE.

Example

char *inp = "   123.4567beej";
char *badchar;

double val = strtod(inp, &badchar);

printf("Converted string to %f\n", val);
printf("Encountered bad characters: %s\n", badchar);

val = strtod("987.654321beej", NULL);
printf("Ignoring bad chars: %f\n", val);

val = strtod("11.2233", &badchar);

if (*badchar == '\0')
    printf("No bad chars: %f\n", val);
else
    printf("Found bad chars: %f, %s\n", val, badchar);

Output:

    Converted string to 123.456700
    Encountered bad characters: beej
    Ignoring bad chars: 987.654321
    No bad chars: 11.223300

43.5 `strtol()`, `strtoll()`, `strtoul()`, `strtoull()`

Convert a string to an integer

Synopsis

    #include <stdlib.h>
    
    long int strtol(const char * restrict nptr,
                    char ** restrict endptr, int base);
    
    long long int strtoll(const char * restrict nptr,
                          char ** restrict endptr, int base);
    
    unsigned long int strtoul(const char * restrict nptr,
                              char ** restrict endptr, int base);
    
    unsigned long long int strtoull(const char * restrict nptr,
                                    char ** restrict endptr, int base);

Description

These convert a string to an integer like atoi(), but they have a few more bells and whistles.

Most notable, they can tell you where conversion started going wrong, i.e. where invalid characters, if any, appear. Leading spaces are ignored. A + or - sign may precede the number.

The basic idea is that if things go well, these functions will return the integer values contained in the strings. And if you pass in the char** typed endptr, it’ll set it to point at the NUL at the end of the string.

If things don’t go well, they’ll set endptr to point at the first character where things have gone awry. That is, if you’re converting a value 103z2! in base 10, they’ll send endptr to point at the z because that’s the first non-numeric character.

You can pass in NULL for endptr if you don’t care to do any of that kind of error checking.

Wait—did I just say we could set the number base for the conversion? Yes! Yes, I did. Now number bases ¹⁹⁷ are out of scope for this document, but certainly some of the more well-known are binary (base 2), octal (base 8), decimal (base 10), and hexadecimal (base 16).

You can specify the number base for the conversion as the third parameter. Bases from 2 to 36 are supported, with case-insensitive digits running from 0 to Z.

If you specify a base of 0, the function will make an effort to determine it. It’ll default to base 10 except for a couple cases:

If the number has a leading 0, it will be octal (base 8)
If the number has a leading 0x or 0X, it will be hex (base 16)

The locale might affect the behavior of these functions.

Return Value

Returns the converted value.

endptr, if not NULL is set to the first invalid character, or to the beginning of the string if no conversion was performed, or to the string terminal NUL if all characters were valid.

If there’s overflow, one of these values will be returned: LONG_MIN, LONG_MAX, LLONG_MIN, LLONG_MAX, ULONG_MAX, ULLONG_MAX. And errno is set to ERANGE.

Example

// All output in decimal (base 10)

printf("%ld\n", strtol("123", NULL, 0));      // 123
printf("%ld\n", strtol("123", NULL, 10));     // 123
printf("%ld\n", strtol("101010", NULL, 2));   // binary, 42
printf("%ld\n", strtol("123", NULL, 8));      // octal, 83
printf("%ld\n", strtol("123", NULL, 16));     // hex, 291

printf("%ld\n", strtol("0123", NULL, 0));     // octal, 83
printf("%ld\n", strtol("0x123", NULL, 0));    // hex, 291

char *badchar;
long int x = strtol("   1234beej", &badchar, 0);

printf("Value is %ld\n", x);               // Value is 1234
printf("Bad chars at \"%s\"\n", badchar);  // Bad chars at "beej"

Output:

    123
    123
    42
    83
    291
    83
    291
    Value is 1234
    Bad chars at "beej"

43.6 `rand()`

Return a pseudorandom number

Synopsis

    #include <stdlib.h>
    
    int rand(void);

Description

This gives us back a pseudorandom number in the range 0 to RAND_MAX, inclusive. (RAND_MAX will be at least $32767$.)

If you want to force this to a certain range, the classic way to do this is to force it with the modulo operator %, although this introduces biases ¹⁹⁸ if RAND_MAX+1 is not a multiple of the number you’re modding by. Dealing with this is out of scope for this guide.

If you want to to make a floating point number between 0 and 1 inclusive, you can divide the result by RAND_MAX. Or RAND_MAX+1 if you don’t want to include 1. But of course, there are out-of-scope problems with this, as well ¹⁹⁹.

In short, rand() is a great way to get potentially poor random numbers with ease. Probably good enough for the game you’re writing.

The spec elaborates:

There are no guarantees as to the quality of the random sequence produced and some implementations are known to produce sequences with distressingly non-random low-order bits. Applications with particular requirements should use a generator that is known to be sufficient for their needs.

Your system probably has a good random number generator on it if you need a stronger source. Linux users have getrandom(), for example, and Windows has CryptGenRandom().

For other more demanding random number work, you might find a library like the GNU Scientific Library ²⁰⁰ of use.

You can explicitly seed the random number generator with srand().

Return Value

Returns a random number in the range 0 to RAND_MAX, inclusive.

Example

Note that all of these examples don’t produce perfectly uniform distributions. But good enough for the untrained eye, and really common in general use when mediocre random number quality is acceptable.

printf("RAND_MAX = %d\n", RAND_MAX);

printf("0 to 9: %d\n", rand() % 10);

printf("10 to 44: %d\n", rand() % 35 + 10);
printf("0 to 0.99999: %f\n", rand() / ((float)RAND_MAX + 1));
printf("10.5 to 15.7: %f\n", 10.5 + 5.2 * rand() / (float)RAND_MAX);

Output on my system:

    RAND_MAX = 2147483647
    0 to 9: 3
    10 to 44: 21
    0 to 0.99999: 0.783099
    10.5 to 15.7: 14.651888

43.7 `srand()`

Seed the built-in pseudorandom number generator

Synopsis

    #include <stdlib.h>
    
    void srand(unsigned int seed);

Description

The dirty little secret of pseudorandom number generation is that they’re completely deterministic. There’s nothing random about them. They just look random.

If you use rand() and run your program several times, you might notice something fishy: they produce the same random numbers over and over again.

To mix it up, we need to give the pseudorandom number generator a new “starting point”, if you will. We call that the seed. It’s just a number, but it is used as the basic for subsequent number generation. Give a different seed, and you’ll get a different sequence of random numbers. Give the same seed, and you’ll get the same sequence of random numbers corresponding to it²⁰¹.

But if you can’t hardcode the seed (because that would give you the same sequence every time), how are you supposed to do this?

It’s really common to use the number of seconds since January 1, 1970 to seed the generator. This sounds pretty arbitrary except for the fact that it’s exactly the value returned by the library call time(NULL). We’ll do that in the example.

If you don’t call srand(), it’s as if you called srand(1).

Return Value

Returns nothing!

Example

#include <stdio.h>
#include <stdlib.h>
#include <time.h>    // for the time() call

int main(void)
{
    srand(time(NULL));

    for (int i = 0; i < 5; i++)
        printf("%d\n", rand() % 32);
}

Output:

Output from a subsequent run:

43.8 `aligned_alloc()`

Allocate specifically-aligned memory

Synopsis

    #include <stdlib.h>
    
    void *aligned_alloc(size_t alignment, size_t size);

Description

Maybe you wanted malloc() or calloc() instead of this. But if you’re sure you don’t, read on!

Normally you don’t have to think about this, since malloc() and realloc() both provide memory regions that are suitably aligned ²⁰² for use with any data type.

But if you need a more specific alignment, you can specify it with this function.

When you’re done using the memory region, be sure to free it with a call to free().

Don’t pass in 0 for the size. It probably won’t do anything you want.

In case you’re wondering, all dynamically-allocated memory is automatically freed by the system when the program ends. That said, it’s considered to be Good Form to explicitly free() everything you allocate. This way other programmers don’t think you were being sloppy.

Return Value

Returns a pointer to the newly-allocated memory, aligned as specified. Returns NULL if something goes wrong.

Example

#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>

int main(void)
{
    int *p = aligned_alloc(256, 10 * sizeof(int));

    // Just for fun, let's convert to intptr_t and mod with 256
    // to make sure we're actually aligned on a 256-byte boundary.
    //
    // This is probably some kind of implementation-defined
    // behavior, but I'll bet it works.

    intptr_t ip = (intptr_t)p;

    printf("%ld\n", ip % 256);   // 0!

    // Free it up
    free(p);
}

43.9 `calloc()`, `malloc()`

Allocate memory for arbitrary use

Synopsis

    #include <stdlib.h>
    
    void *calloc(size_t nmemb, size_t size);
    
    void *malloc(size_t size);

Description

Both of these functions allocate memory for general-purpose use. It will be aligned such that it’s useable for storing any data type.

malloc() allocates exactly the specified number of bytes of memory in a contiguous block. The memory might be full of garbage data. (You can clear it with memset(), if you wish.)

calloc() is different in that it allocates space for nmemb objects of size bytes each. (You can do the same with malloc(), but you have to do the multiplication yourself.)

calloc() has an additional feature: it clears all the memory to 0.

So if you’re planning to zero the memory anyway, calloc() is probably the way to go. If you’re not, you can avoid that overhead by calling malloc().

When you’re done using the memory region, free it with a call to free().

Don’t pass in 0 for the size. It probably won’t do anything you want.

Return Value

Both functions return a pointer to the shiny, newly-allocated memory. Or NULL if something’s gone awry.

Example

Comparison of malloc() and calloc() for allocating 5 ints:

// Allocate space for 5 ints
int *p = malloc(5 * sizeof(int));

p[0] = 12;
p[1] = 30;

// Allocate space for 5 ints
// (Also clear that memory to 0)
int *q = calloc(5, sizeof(int));

q[0] = 12;
q[1] = 30;

// All done
free(p);
free(q);

43.10 `free()`

Free a memory region

Synopsis

    #include <stdlib.h>
    
    void free(void *ptr);

Description

You know that pointer you got back from malloc(), calloc(), or aligned_alloc()? You pass that pointer to free() to free the memory associated with it.

If you don’t do this, the memory will stay allocated FOREVER AND EVER! (Well, until your program exits, anyway.)

Fun fact: free(NULL) does nothing. You can safely call that. Sometimes it’s convenient.

Don’t free() a pointer that’s already been free()d. Don’t free() a pointer that you didn’t get back from one of the allocation functions. It would be Bad²⁰³.

Return Value

Returns nothing!

Example

// Allocate space for 5 ints
int *p = malloc(5 * sizeof(int));

p[0] = 12;
p[1] = 30;

// Free that space
free(p);

43.11 `realloc()`

Resize a previously allocated stretch of memory

Synopsis

    #include <stdlib.h>
    
    void *realloc(void *ptr, size_t size);

Description

This takes a pointer to some memory previously allocated with malloc() or calloc() and resizes it to the new size.

If the new size is smaller than the old size, any data larger than the new size is discarded.

If the new size is larger than the old size, the new larger part is uninitialized. (You can clear it with memset().)

Important note: the memory might move! If you resize, the system might need to relocate the memory to a larger continguous chunk. If this happens, realloc() will copy the old data to the new location for you.

Because of this, it’s important to save the returned value to your pointer to update it to the new location if things move. (Also, be sure to error-check so that you don’t overwrite your old pointer with NULL, leaking the memory.)

You can also relloc() memory allocated with aligned_alloc(), but it will potentially lose its alignment if the block is moved.

Return Value

Returns a pointer to the resized memory region. This might be equivalent to the ptr passed in, or it might be some other location.

Example

// Allocate space for 5 ints
int *p = malloc(5 * sizeof(int));

p[0] = 12;
p[1] = 30;

// Reallocate for 10 bytes
int *new_p = realloc(p, 10 * sizeof(int));

if (new_p == NULL) {
    printf("Error reallocing\n");
} else {
    p = new_p;  // It's good; let's keep it
    p[7] = 99;
}

// All done
free(p);

43.12 `abort()`

Abruptly end program execution

Synopsis

    #include <stdlib.h>
    
    _Noreturn void abort(void);

Description

This ends program execution abnormally and immediately. Use this in rare, unexpected circumstances.

Open streams might not be flushed. Temporary files created might not be removed. Exit handlers are not called.

A non-zero exit status is returned to the environment.

On some systems, abort() might dump core ²⁰⁴, but this is outside the scope of the spec.

You can cause the equivalent of an abort() by calling raise(SIGABRT), but I don’t know why you’d do that.

The only portable way to stop an abort() call midway is to use signal() to catch SIGABRT and then exit() in the signal handler.

Return Value

This function never returns.

Example

if (bad_thing) {
    printf("This should never have happened!\n");
    fflush(stdout);  // Make sure the message goes out
    abort();
}

On my system, this outputs:

    This should never have happened!
    zsh: abort (core dumped)  ./foo

43.13 `atexit()`, `at_quick_exit()`

Set up handlers to run when the program exits

Synopsis

    #include <stdlib.h>
    
    int atexit(void (*func)(void));
    
    int at_quick_exit(void (*func)(void));

Description

When the program does a normal exit with exit() or returns from main(), it looks for previously-registered handlers to call on the way out. These handlers are registered with the atexit() call.

Think of it like, “Hey, when you’re about to exit, do these extra things.”

For the quick_exit() call, you can use the at_quick_exit() function to register handlers for that²⁰⁵. There’s no crossover in handlers from exit() to quick_exit(), i.e. for a call to one, none of the other’s handlers will fire.

You can register multiple handlers to fire—at least 32 handlers are supported by both exit() and quick_exit().

The argument func to the functions looks a little weird—it’s a pointer to a function to call. Basically just put the function name to call in there (without parentheses after). See the example, below.

If you call atexit() from inside your atexit() handler (or equivalent in your at_quick_exit() handler), it’s unspecified if it will get called. So get them all registered before you exit.

When exiting, the functions will be called in the reverse order they were registered.

Return Value

These functions return 0 on success, or nonzero on failure.

Example

atexit():

#include <stdio.h>
#include <stdlib.h>

void exit_handler_1(void)
{
    printf("Exit handler 1 called!\n");
}

void exit_handler_2(void)
{
    printf("Exit handler 2 called!\n");
}

int main(void)
{
    atexit(exit_handler_1);
    atexit(exit_handler_2);

    exit(0);
}

For the output:

    Exit handler 2 called!
    Exit handler 1 called!

And a similar example with quick_exit():

#include <stdio.h>
#include <stdlib.h>

void exit_handler_1(void)
{
    printf("Exit handler 1 called!\n");
}

void exit_handler_2(void)
{
    printf("Exit handler 2 called!\n");
}

int main(void)
{
    at_quick_exit(exit_handler_1);
    at_quick_exit(exit_handler_2);

    quick_exit(0);
}

43.14 `exit()`, `quick_exit()`, `_Exit()`

Exit the currently-running program

Synopsis

    #include <stdlib.h>
    
    _Noreturn void exit(int status);
    
    _Noreturn void quick_exit(int status);
    
    _Noreturn void _Exit(int status);

Description

All these functions cause the program to exit, with various levels of cleanup performed.

exit() does the most cleanup and is the most normal exit.

quick_exit() is the second most.

_Exit() unceremoniously drops everything and ragequits on the spot.

Calling either of exit() or quick_exit() causes their respective atexit() or at_quick_exit() handlers to be called in the reverse order in which they were registered.

exit() will flush all streams and delete all temporary files.

quick_exit() or _Exit() might not perform that nicety.

_Exit() doesn’t call any of the at-exit handlers, either.

For all functions, the exit status is returned to the environment.

Defined exit statuses are:

Status	Description
`EXIT_SUCCESS`	Typically returned when good things happen
`0`	Same as `EXIT_SUCCESS`
`EXIT_FAILURE`	Oh noes! Definitely failure!
Any positive value	Generally indicates another failure of some kind

Return Value

None of these functions ever return.

Example

exit(EXIT_SUCCESS);

quick_exit(EXIT_FAILURE);

_Exit(2);

43.15 `getenv()`

Get the value of an environment variable

Synopsis

    #include <stdlib.h>
    
    char *getenv(const char *name);

Description

The environment often provides variables that are set before the program run that you can access at runtime.

Of course the exact details are system dependent, but these variables are key/value pairs, and you can get the value by passing the key to getenv() as the name parameter.

You’re not allowed to overwrite the string that’s returned.

This is pretty limited in the standard, but your OS often provides better functionality. See the Environment Variables section for more details.

Return Value

Returns a pointer to the environment variable value, or NULL if the variable doesn’t exist.

Example

printf("PATH is %s\n", getenv("PATH"));

Output (truncated in my case):

    PATH is /usr/bin:/usr/local/bin:/usr/sbin:/home/beej/.cargo/bin [...]

43.16 `system()`

Run an external program

Synopsis

    #include <stdlib.h>
    
    int system(const char *string);

Description

This will run an external program and then return to the caller.

The manner in which it runs the program is system-defined, but typically you can pass something to it just like you’d run on the command line, searching the PATH, etc.

Not all systems have this capability, but you can test for it by passing NULL to system() and seeing if it returns 0 (no command processor is available) or non-zero (a command processor is available! Yay!)

If you’re getting user input and passing it to the system() call, be extremely careful to escape all special shell characters (everything that’s not alphanumeric) with a backslash to keep a villain from running something you don’t want them to.

Return Value

If NULL is passed, returns nonzero if a command processor is available (i.e. system() will work at all).

Otherwise returns an implementation-defined value.

Example

printf("Here's a directory listing:\n\n");

system("ls -l");   // Run this command and return

printf("\nAll done!\n");

Output:

    Here's a directory listing:
    
    total 92
    drwxr-xr-x 3 beej beej  4096 Oct 14 21:38 bin
    drwxr-xr-x 2 beej beej  4096 Dec 20 20:07 examples
    -rwxr-xr-x 1 beej beej 16656 Feb 23 21:49 foo
    -rw-rw-rw- 1 beej beej   155 Feb 23 21:49 foo.c
    -rw-r--r-- 1 beej beej  1350 Jan 27 22:11 Makefile
    -rw-r--r-- 1 beej beej  4644 Jan 18 09:12 README.md
    drwxr-xr-x 3 beej beej  4096 Feb 23 20:21 src
    drwxr-xr-x 6 beej beej  4096 Feb 21 20:24 stage
    drwxr-xr-x 2 beej beej  4096 Sep 27 20:54 translations
    drwxr-xr-x 2 beej beej  4096 Sep 27 20:54 website
    
    All done!

43.17 `bsearch()`

Binary Search (maybe) an array of objects

Synopsis

    #include <stdlib.h>
    
    void *bsearch(const void *key, const void *base,
                  size_t nmemb, size_t size,
                  int (*compar)(const void *, const void *));

Description

This crazy-looking function searches an array for a value.

It probably is a binary search or some fast, efficient search. But the spec doesn’t really say.

However, the array must be sorted! So binary search seems likely.

key is a pointer to the value to find.
base is a pointer to the start of the array—the array must be sorted!
nmemb is the number of elements in the array.
size is the sizeof each element in the array.
compar is a pointer to a function that will compare the key against other values.

The comparison function takes the key as the first argument and the value to compare against as the second. It should return a negative number if the key is less than the value, 0 if the key equals the value, and a positive number if the key is greater than the value.

This is commonly computed by taking the difference between the key and the value to be compared. If subtraction is supported.

The return value from the strcmp() function can be used for comparing strings.

Again, the array must be sorted according to the order of the comparison function before running bsearch(). Luckily for you, you can just call qsort() with the same comparison function to get this done.

It’s a general-purpose function—it’ll search any type of array for anything. The catch is you have to write the comparison function.

And that’s not as scary as it looks. Jump down to the example

Return Value

The function returns a pointer to the found value, or NULL if it can’t be found.

Example

#include <stdio.h>
#include <stdlib.h>

int compar(const void *key, const void *value)
{
    const int *k = key, *v = value;  // Need ints, not voids

    return *k - *v;
}

int main(void)
{
    int a[9] = {2, 6, 9, 12, 13, 18, 20, 32, 47};

    int *r, key;

    key = 12;  // 12 is in there
    r = bsearch(&key, a, 9, sizeof(int), compar);
    printf("Found %d\n", *r);

    key = 30;  // Won't find a 30
    r = bsearch(&key, a, 9, sizeof(int), compar);
    if (r == NULL)
        printf("Didn't find 30\n");

    // Searching with an unnamed key, pointer to 32
    r = bsearch(&(int){32}, a, 9, sizeof(int), compar);
    printf("Found %d\n", *r);  // Found it
}

Output:

    Found 12
    Didn't find 30
    Found 32

43.18 `qsort()`

Quicksort (maybe) some data

Synopsis

    #include <stdlib.h>
    
    void qsort(void *base, size_t nmemb, size_t size,
               int (*compar)(const void *, const void *));

Description

This function will quicksort (or some other sort, probably speedy) an array of data in-place²⁰⁶.

Like bsearch(), it’s data-agnostic. Any data for which you can define a relative ordering can be sorted, whether ints, structs, or anything else.

Also like bsearch(), you have to give a comparison function to do the actual compare.

base is a pointer to the start of the array to be sorted.
nmemb is the number of elements in the array.
size is the sizeof each element.
compar is a pointer to the comparison function.

The comparison function takes pointers to two elements of the array as arguments and compares them. It should return a negative number if the first argument is less than the second, 0 if they are equal, and a positive number if the first argument is greater than the second.

This is commonly computed by taking the difference between the first argument and the second. If subtraction is supported.

The return value from the strcmp() function can provide sort order for strings.

If you have to sort a struct, just subtract the specific field you want to sort by.

This comparison function can be used by bsearch() to do searches after the list is sorted.

To reverse the sort, subtract the second argument from the first, i.e. negate the return value from compar().

Return Value

Returns nothing!

Example

#include <stdio.h>
#include <stdlib.h>

int compar(const void *elem0, const void *elem1)
{
    const int *x = elem0, *y = elem1;  // Need ints, not voids

    return *x - *y;
}

int main(void)
{
    int a[9] = {14, 2, 3, 17, 10, 8, 6, 1, 13};

    // Sort the list

    qsort(a, 9, sizeof(int), compar);

    // Print sorted list

    for (int i = 0; i < 9; i++)
        printf("%d ", a[i]);

    putchar('\n');

    // Use the same compar() function to binary search
    // for 17 (passed in as an unnamed object)

    int *r = bsearch(&(int){17}, a, 9, sizeof(int), compar);
    printf("Found %d!\n", *r);
}

Output:

    1 2 3 6 8 10 13 14 17
    Found 17!

43.19 `abs()`, `labs()`, `llabs()`

Compute the absolute value of an integer

Synopsis

    #include <stdlib.h>
    
    int abs(int j);
    
    long int labs(long int j);
    
    long long int llabs(long long int j);

Description

Compute the absolute value of j. If you don’t remember, that’s how far from zero j is.

In other words, if j is negative, return it as a positive. If it’s positive, return it as a positive. Always be positive. Enjoy life.

If the result cannot be represented, the behavior is undefined. Be especially aware of the upper half of unsigned numbers.

Return Value

Returns the absolute value of j, $|j|$.

Example

printf("|-2| = %d\n", abs(-2));
printf("|4|  = %d\n", abs(4));

Output:

    |-2| = 2
    |4|  = 4

43.20 `div()`, `ldiv()`, `lldiv()`

Compute the quotient and remainder of two numbers

Synopsis

    #include <stdlib.h>
    
    div_t div(int numer, int denom);
    
    ldiv_t ldiv(long int numer, long int denom);
    
    lldiv_t lldiv(long long int numer, long long int denom);

Description

These functions get you the quotient and remainder of a pair of numbers in one go.

They return a structure that has two fields, quot, and rem, the types of which match types of numer and denom. Note how each function returns a different variant of div_t.

These div_t variants are equivalent to the following:

    typedef struct {
        int quot, rem;
    } div_t;
    
    typedef struct {
        long int quot, rem;
    } ldiv_t;
    
    typedef struct {
        long long int quot, rem;
    } lldiv_t;

Why use these instead of the division operator?

The C99 Rationale says:

Because C89 had implementation-defined semantics for division of signed integers when negative operands were involved, div and ldiv, and lldiv in C99, were invented to provide well-specified semantics for signed integer division and remainder operations. The semantics were adopted to be the same as in Fortran. Since these functions return both the quotient and the remainder, they also serve as a convenient way of efficiently modeling underlying hardware that computes both results as part of the same operation. Table 7.2 summarizes the semantics of these functions.

Indeed, K&R2 (C89) says:

The direction of truncation for / and the sign of the result for % are machine-dependent for negative operands […]

The Rationale then goes on to spell out what the signs of the quotient and remainder will be given the signs of a numerator and denominator when using the div() functions:

`numer`	`denom`	`quot`	`rem`
$+$	$+$	$+$	$+$
$-$	$+$	$-$	$-$
$+$	$-$	$-$	$+$
$-$	$-$	$+$	$-$

Return Value

A div_t, ldiv_t, or lldiv_t structure with the quot and rem fields loaded with the quotient and remainder of the operation of numer/denom.

Example

div_t d = div(64, -7);

printf("64 / -7 = %d\n", d.quot);
printf("64 %% -7 = %d\n", d.rem);

Output:

    64 / -7 = -9
    64 % -7 = 1

43.21 `mblen()`

Return the number of bytes in a multibyte character

Synopsis

    #include <stdlib.h>
    
    int mblen(const char *s, size_t n);

Description

If you have a multibyte character in a string, this will tell you how many bytes long it is.

n is the maximum number of bytes mblen() will scan before giving up.

If s is a NULL pointer, tests if this encoding has state dependency, as noted in the return value, below. It also resets the state, if there is one.

The behavior of this function is influenced by the locale.

Return Value

Returns the number of bytes used to encode this character, or -1 if there is no valid multibyte character in the next n bytes.

Or, if s is NULL, returns true if this encoding has state dependency.

Example

For the example, I used my extended character set to put Unicode characters in the source. If this doesn’t work for you, use the \uXXXX escape.

#include <stdio.h>
#include <stdlib.h>
#include <locale.h>

int main(void)
{
    setlocale(LC_ALL, "");

    printf("State dependency: %d\n", mblen(NULL, 0));
    printf("Bytes for €: %d\n", mblen("€", 5));
    printf("Bytes for \u00e9: %d\n", mblen("\u00e9", 5));  // \u00e9 == é
    printf("Bytes for &: %d\n", mblen("&", 5));
}

Output (in my case, the encoding is UTF-8, but your mileage may vary):

    State dependency: 0
    Bytes for €: 3
    Bytes for é: 2
    Bytes for &: 1

43.22 `mbtowc()`

Convert a multibyte character to a wide character

Synopsis

    #include <stdlib.h>
    
    int mbtowc(wchar_t * restrict pwc, const char * restrict s, size_t n);

Description

If you have a multibyte character, this function will convert it to a wide character and stored at the address pointed to by pwc. Up to n bytes of the multibyte character will be analyzed.

If pwc is NULL, the resulting character will not be stored. (Useful for just getting the return value.)

If s is a NULL pointer, tests if this encoding has state dependency, as noted in the return value, below. It also resets the state, if there is one.

The behavior of this function is influenced by the locale.

Return Value

Returns the number of bytes used in the encoded wide character, or -1 if there is no valid multibyte character in the next n bytes.

Returns 0 if s points to the NUL character.

Or, if s is NULL, returns true if this encoding has state dependency.

Example

#include <stdio.h>
#include <stdlib.h>
#include <locale.h>
#include <wchar.h>

int main(void)
{
    setlocale(LC_ALL, "");

    printf("State dependency: %d\n", mbtowc(NULL, NULL, 0));

    wchar_t wc;
    int bytes;

    bytes = mbtowc(&wc, "€", 5);

    printf("€ takes %d bytes as wide char L'%lc'\n", bytes, wc);
}

Output on my system:

    State dependency: 0
    € takes 3 bytes as wide char L'€'

43.23 `wctomb()`

Convert a wide character to a multibyte character

Synopsis

    #include <stdlib.h>
    
    int wctomb(char *s, wchar_t wc);

Description

If you have your hands on a wide character, you can use this to make it multibyte.

The wide character wc is stored as a multibyte character in the string pointed to by s. The buffer s points to should be at least MB_CUR_MAX characters long. Note that MB_CUR_MAX changes with locale.

If wc is a NUL wide character, a NUL is stored in s after the bytes needed to reset the shift state (if any).

If s is a NULL pointer, tests if this encoding has state dependency, as noted in the return value, below. It also resets the state, if there is one.

The behavior of this function is influenced by the locale.

Return Value

Returns the number of bytes used in the encoded multibyte character, or -1 if wc does not correspond to any valid multibyte character.

Or, if s is NULL, returns true if this encoding has state dependency.

Example

#include <stdio.h>
#include <stdlib.h>
#include <locale.h>
#include <wchar.h>

int main(void)
{
    setlocale(LC_ALL, "");

    printf("State dependency: %d\n", mbtowc(NULL, NULL, 0));

    int bytes;
    char mb[MB_CUR_MAX + 1];

    bytes = wctomb(mb, L'€');
    mb[bytes] = '\0';

    printf("L'€' takes %d bytes as multibyte char '%s'\n", bytes, mb);
}

Output on my system:

    State dependency: 0
    L'€' takes 3 bytes as multibyte char '€'

43.24 `mbstowcs()`

Convert a multibyte string to a wide character string

Synopsis

    #include <stdlib.h>
    
    size_t mbstowcs(wchar_t * restrict pwcs, const char * restrict s, size_t n);

Description

If you have a multibyte string (AKA a regular string), you can convert it wto a wide character string with this function.

At most n wide characters are written to the destination pwcs from the source s.

A NUL character is stored as a wide NUL character.

Non-portable POSIX extension: if you’re using a POSIX-complaint library, this function allows pwcs to be NULL if you’re only interested in the return value. Most notably, this will give you the number of characters in a multibyte string (as opposed to strlen() which counts the bytes.)

Return Value

Returns the number of wide characters written to the destination pwcs.

If an invalid multibyte character was found, returns (size_t)(-1).

If the return value is n, it means the result was not NUL-terminated.

Example

This source uses an extended character set. If your compiler doesn’t support it, you’ll have to replace them with \u escapes.

#include <stdio.h>
#include <stdlib.h>
#include <locale.h>
#include <string.h>

int main(void)
{
    setlocale(LC_ALL, "");

    wchar_t wcs[128];
    char *s = "€200 for this spoon?";  // 20 characters

    size_t char_count, byte_count;

    char_count = mbstowcs(wcs, s, 128);
    byte_count = strlen(s);

    printf("Wide string: L\"%ls\"\n", wcs);
    printf("Char count : %zu\n", char_count);    // 20
    printf("Byte count : %zu\n\n", byte_count);  // 22 on my system

    // POSIX Extension that allows you to pass NULL for
    // the destination so you can just use the return
    // value (which is the character count of the string, 
    // if no errors have occurred)

    s = "§¶°±π€•";  // 7 characters

    char_count = mbstowcs(NULL, s, 0);  // POSIX-only, nonportable
    byte_count = strlen(s);

    printf("Multibyte str: \"%s\"\n", s);
    printf("Char count   : %zu\n", char_count);  // 7
    printf("Byte count   : %zu\n", byte_count);  // 16 on my system
}

Output on my system (byte count will depend on your encoding):

    Wide string: L"€200 for this spoon?"
    Char count : 20
    Byte count : 22
    
    Multibyte str: "§¶°±π€•"
    Char count   : 7
    Byte count   : 16

43.25 `wcstombs()`

Convert a wide character string to a multibyte string

Synopsis

    #include <stdlib.h>
    
    size_t wcstombs(char * restrict s, const wchar_t * restrict pwcs, size_t n);

Description

If you have a wide character string and you want it as multibyte string, this is the function for you!

It’ll take the wide characters pointed to by pwcs and convert them to multibyte characters stored in s. No more than n bytes will be written to s.

Non-portable POSIX extension: if you’re using a POSIX-complaint library, this function allows s to be NULL if you’re only interested in the return value. Most notably, this will give you the number of bytes needed to encode the wide characters in a multibyte string.

Return Value

Returns the number of bytes written to s, or (size_t)(-1) if one of the characters can’t be encoded into a multibyte string.

If the return value is n, it means the result was not NUL-terminated.

Example

This source uses an extended character set. If your compiler doesn’t support it, you’ll have to replace them with \u escapes.

#include <stdio.h>
#include <stdlib.h>
#include <locale.h>
#include <string.h>

int main(void)
{
    setlocale(LC_ALL, "");

    char mbs[128];
    wchar_t *wcs = L"€200 for this spoon?";  // 20 characters

    size_t byte_count;

    byte_count = wcstombs(mbs, wcs, 128);

    printf("Wide string: L\"%ls\"\n", wcs);
    printf("Multibyte  : \"%s\"\n", mbs);
    printf("Byte count : %zu\n\n", byte_count);  // 22 on my system

    // POSIX Extension that allows you to pass NULL for
    // the destination so you can just use the return
    // value (which is the character count of the string, 
    // if no errors have occurred)

    wcs = L"§¶°±π€•";  // 7 characters

    byte_count = wcstombs(NULL, wcs, 0);  // POSIX-only, nonportable

    printf("Wide string: L\"%ls\"\n", wcs);
    printf("Byte count : %zu\n", byte_count);  // 16 on my system
}

Output on my system (byte count will depend on your encoding):

    Wide string: L"€200 for this spoon?"
    Multibyte  : "€200 for this spoon?"
    Byte count : 22
    
    Wide string: L"§¶°±π€•"
    Byte count : 16

44 `<time.h>` Date and Time Functions

Function	Description
`clock()`	How much processor time has been used by this process
`difftime()`	Compute the difference between two times
`mktime()`	Convert a `struct tm` into a `time_t`
`time()`	Get the current calendar time
`timespec_get()`	Get a higher resolution time, probably now
`asctime()`	Return a human-readable version of a `struct tm`
`ctime()`	Return a human-readable version of a `time_t`
`gmtime()`	Convert a calendar time into a UTC broken-down time
`localtime()`	Convert a calendar time into a broken-down local time
`strftime()`	Formatted date and time output

When it comes to time and C, there are two main types to look for:

time_t holds a calendar time. This is an potentially opaque numeric type that represents an absolute time that can be converted to UTC²⁰⁷ or local time.
struct tm holds a broken-down time. This has things like the day of the week, the day of the month, the hour, the minute, the second, etc.

On POSIX systems and Windows, time_t is an integer and represents the number of seconds that have elapsed since January 1, 1970 at 00:00 UTC.

A struct tm contains the following fields:

    struct tm {
        int tm_sec;    // seconds after the minute -- [0, 60]
        int tm_min;    // minutes after the hour -- [0, 59]
        int tm_hour;   // hours since midnight -- [0, 23]
        int tm_mday;   // day of the month -- [1, 31]
        int tm_mon;    // months since January -- [0, 11]
        int tm_year;   // years since 1900
        int tm_wday;   // days since Sunday -- [0, 6]
        int tm_yday;   // days since January 1 -- [0, 365]
        int tm_isdst;  // Daylight Saving Time flag
    };

You can convert between the two with mktime(), gmtime(), and localtime().

You can print time information to strings with ctime(), asctime(), and strftime().

44.1 Thread Safety Warning

asctime(), ctime(): These two functions return a pointer to a static memory region. They both might return the same pointer. If you need thread safety, you’ll need a mutex across them. If you need both results at once, strcpy() one of them out.

All these problems with asctime() and ctime() can be avoided by using the more flexible and thread-safe strftime() function instead.

localtime(), gmtime(): These other two functions also return a pointer to a static memory region. They both might return the same pointer. If you need thread safety, you’ll need a mutex across them. If you need both results at once, copy the struct to another.

44.2 `clock()`

How much processor time has been used by this process

Synopsis

    #include <time.h>
    
    clock_t clock(void);

Description

Your processor is juggling a lot of things right now. Just because a process has been alive for 20 minutes doesn’t mean that it used 20 minutes of “CPU time”.

Most of the time your average process spends asleep, and that doesn’t count toward the CPU time spent.

This function returns an opaque type representing the number of “clock ticks”²⁰⁸ the process has spent in operation.

You can get the number of seconds out of that by dividing by the macro CLOCKS_PER_SEC. This is an integer, so you will have to cast part of the expression to a floating type to get a fractional time.

Note that this is not the “wall clock time” of the program. If you want to get that loosely use time() and difftime() (which might only offer 1-second resolution) or timespec_get() (which might only also offer low resolution, but at least it might go to nanosecond level).

Return Value

Returns the amount of CPU time spent by this process. This comes back in a form that can be divided by CLOCKS_PER_SEC to determine the time in seconds.

Example

#include <stdio.h>
#include <time.h>

// Deliberately naive Fibonacci
long long int fib(long long int n) {
    if (n <= 1) return n;

    return fib(n-1) + fib(n-2);
}

int main(void)
{
    printf("The 42nd Fibonacci Number is %lld\n", fib(42));

    printf("CPU time: %f\n", clock() / (double)CLOCKS_PER_SEC);
}

Output on my system:

    The 42nd Fibonacci Number is 267914296
    CPU time: 1.863078

44.3 `difftime()`

Compute the difference between two times

Synopsis

    #include <time.h>
    
    double difftime(time_t time1, time_t time0);

Description

Since the time_t type is technically opaque, you can’t just straight-up subtract to get the difference between two of them²⁰⁹. Use this function to do it.

There is no guarantee as to the resolution of this difference, but it’s probably to the second.

Return Value

Returns the difference between two time_ts in seconds.

Example

#include <stdio.h>
#include <time.h>

int main(void)
{
    // April 12, 1982 and change
    struct tm time_a = { .tm_year=82, .tm_mon=3, .tm_mday=12,
        .tm_hour=4, .tm_min=00, .tm_sec=04, .tm_isdst=-1,
    };

    // November 15, 2020 and change
    struct tm time_b = { .tm_year=120, .tm_mon=10, .tm_mday=15,
        .tm_hour=16, .tm_min=27, .tm_sec=00, .tm_isdst=-1,
    };

    time_t cal_a = mktime(&time_a);
    time_t cal_b = mktime(&time_b);

    double diff = difftime(cal_b, cal_a);

    double years = diff / 60 / 60 / 24 / 365.2425;  // close enough

    printf("%f seconds (%f years) between events\n", diff, years);
}

Output:

    1217996816.000000 seconds (38.596783 years) between events

44.4 `mktime()`

Convert a struct tm into a time_t

Synopsis

    #include <time.h>
    
    time_t mktime(struct tm *timeptr);

Description

If you have a local date and time and want it converted to a time_t (so that you can difftime() it or whatever), you can convert it with this function.

Basically you fill out the fields in your struct tm in local time and mktime() will convert those to the UTC time_t equivalent.

A couple notes:

Don’t bother filling out tm_wday or tm_yday. mktime() will fill these out for you.
You can set tm_isdst to 0 to indicate your time isn’t Daylight Saving Time (DST), 1 to indicate it is, and -1 to have mktime() fill it in according to your locale’s preference.

If you need input in UTC, see the non-standard functions timegm()²¹⁰ for Unix-likes and _mkgmtime()²¹¹ for Windows.

Return Value

Returns the local time in the struct tm as a time_t calendar time.

Returns (time_t)(-1) on error.

Example

In the following example, we have mktime() tell us if that time was DST or not.

#include <stdio.h>
#include <time.h>

int main(void)
{
    struct tm broken_down_time = {
        .tm_year=82,   // years since 1900
        .tm_mon=3,     // months since January -- [0, 11]
        .tm_mday=12,   // day of the month -- [1, 31]
        .tm_hour=4,    // hours since midnight -- [0, 23]
        .tm_min=00,    // minutes after the hour -- [0, 59]
        .tm_sec=04,    // seconds after the minute -- [0, 60]
        .tm_isdst=-1,  // Daylight Saving Time flag
    };

    time_t calendar_time = mktime(&broken_down_time);

    char *days[] = {"Sunday", "Monday", "Tuesday",
        "Wednesday", "Furzeday", "Friday", "Saturday"};

    // This will print what was in broken_down_time
    printf("Local time : %s", asctime(localtime(&calendar_time)));
    printf("Is DST     : %d\n", broken_down_time.tm_isdst);
    printf("Day of week: %s\n\n", days[broken_down_time.tm_wday]);

    // This will print UTC for the local time, above
    printf("UTC        : %s", asctime(gmtime(&calendar_time)));
}

Output (for me in Pacific Time—UTC is 8 hours ahead):

    Local time : Mon Apr 12 04:00:04 1982
    Is DST     : 0
    Day of week: Monday
    
    UTC        : Mon Apr 12 12:00:04 1982

44.5 `time()`

Get the current calendar time

Synopsis

    #include <time.h>
    
    time_t time(time_t *timer);

Description

Returns the current calendar time right now. I mean, now. No, now!

If timer is not NULL, it gets loaded with the current time, as well.

This can be converted into a struct tm with localtime() or gmtime(), or printed directly with ctime().

Return Value

Returns the current calendar time. Also loads timer with the current time if it’s not NULL.

Or returns (time_t)(-1) if the time isn’t available because you’ve fallen out of the space-time continuum and/or the system doesn’t support times.

Example

time_t now = time(NULL);

printf("The local time is %s", ctime(&now));

Example output:

    The local time is Mon Mar  1 18:45:14 2021

44.6 `timespec_get()`

Get a higher resolution time, probably now

Synopsis

    #include <time.h>
    
    int timespec_get(struct timespec *ts, int base);

Description

This function loads the current time UTC (unless directed otherwise) into the given struct timespec, ts.

That structure has two fields:

    struct timespec {
        time_t tv_sec;   // Whole seconds
        long   tv_nsec;  // Nanoseconds, 0-999999999
    }

Nanoseconds are billionths of a second. You can divide by 1000000000.0 to convert to seconds.

The base parameter has only one defined value, by the spec: TIME_UTC. So portably make it that. This will load ts with the current time in seconds since a system-defined Epoch ²¹², often January 1, 1970 at 00:00 UTC.

Your implementation might define other values for base.

Return Value

When base is TIME_UTC, loads ts with the current UTC time.

On success, returns base, valid values for which will always be non-zero. On error, returns 0.

Example

struct timespec ts;

timespec_get(&ts, TIME_UTC);

printf("%ld s, %ld ns\n", ts.tv_sec, ts.tv_nsec);

double float_time = ts.tv_sec + ts.tv_nsec/1000000000.0;
printf("%f seconds since epoch\n", float_time);

Example output:

    1614654187 s, 825540756 ns
    1614654187.825541 seconds since epoch

Here’s a helper function to add values to a struct timespec that handles negative values and nanosecond overflow.

    #include <stdlib.h>
    
    // Add delta seconds and delta nanoseconds to ts.
    // Negative values are allowed. Each component is added individually.
    //
    // Subtract 1.5 seconds from the current value:
    //
    // timespec_add(&ts, -1, -500000000L);
    
    struct timespec *timespec_add(struct timespec *ts, long dsec, long dnsec)
    {
        long sec = (long)ts->tv_sec + dsec;
        long nsec = ts->tv_nsec + dnsec;
    
        ldiv_t qr = ldiv(nsec, 1000000000L);
    
        if (qr.rem < 0) {
            nsec = 1000000000L + qr.rem;
            sec += qr.quot - 1;
        } else {
            nsec = qr.rem;
            sec += qr.quot;
        }
    
        ts->tv_sec = sec;
        ts->tv_nsec = nsec;
    
        return ts;
    }

And here are some functions to convert from long double to struct timespec and back, just in case you like thinking in decimals. This is more limited in significant figures than using the integer values.

    #include <math.h>
    
    // Convert a struct timespec into a long double
    long double timespec_to_ld(struct timespec *ts)
    {
        return ts->tv_sec + ts->tv_nsec / 1000000000.0;
    }
    
    // Convert a long double to a struct timespec
    struct timespec ld_to_timespec(long double t)
    {
        long double f;
        struct timespec ts;
        ts.tv_nsec = modfl(t, &f) * 1000000000L;
        ts.tv_sec = f;
    
        return ts;
    }

44.7 `asctime()`

Return a human-readable version of a struct tm

Synopsis

    #include <time.h>
    
    char *asctime(const struct tm *timeptr)

Description

This takes a time in a struct tm and returns a string with that date in the form:

    Sun Sep 16 01:03:52 1973

with a newline included at the end, rather unhelpfully. (strftime() will give you more flexibility.)

It’s just like ctime(), except it takes a struct tm instead of a time_t.

WARNING: This function returns a pointer to a static char* region that isn’t thread-safe and might be shared with the ctime() function. If you need thread safety, use strftime() or use a mutex that covers ctime() and asctime().

Behavior is undefined for:

Years less than 1000
Years greater than 9999
Any members of timeptr are out of range

Return Value

Returns a pointer to the human-readable date string.

Example

time_t now = time(NULL);

printf("Local: %s", asctime(localtime(&now)));
printf("UTC  : %s", asctime(gmtime(&now)));

Sample output:

    Local: Mon Mar  1 21:17:34 2021
    UTC  : Tue Mar  2 05:17:34 2021

44.8 `ctime()`

Return a human-readable version of a time_t

Synopsis

    #include <time.h>
    
    char *ctime(const time_t *timer);

Description

This takes a time in a time_t and returns a string with the local time and date in the form:

    Sun Sep 16 01:03:52 1973

with a newline included at the end, rather unhelpfully. (strftime() will give you more flexibility.)

It’s just like asctime(), except it takes a time_t instead of a struct tm.

WARNING: This function returns a pointer to a static char* region that isn’t thread-safe and might be shared with the asctime() function. If you need thread safety, use strftime() or use a mutex that covers ctime() and asctime().

Behavior is undefined for:

Years less than 1000
Years greater than 9999
Any members of timeptr are out of range

Return Value

A pointer to the human-readable local time and data string.

Example

time_t now = time(NULL);

printf("Local: %s", ctime(&now));

Sample output:

    Local: Mon Mar  1 21:32:23 2021

44.9 `gmtime()`

Convert a calendar time into a UTC broken-down time

Synopsis

    #include <time.h>
    
    struct tm *gmtime(const time_t *timer);

Description

If you have a time_t, you can run it through this function to get a struct tm back full of the corresponding broken-down UTC time information.

This is just like localtime(), except it does UTC instead of local time.

Once you have that struct tm, you can feed it to strftime() to print it out.

WARNING: This function returns a pointer to a static struct tm* region that isn’t thread-safe and might be shared with the localtime() function. If you need thread safety use a mutex that covers gmtime() and localtime().

Return Value

Returns a pointer to the broken-down UTC time, or NULL if it can’t be obtained.

Example

time_t now = time(NULL);

printf("UTC  : %s", asctime(gmtime(&now)));
printf("Local: %s", asctime(localtime(&now)));

Sample output:

    UTC  : Tue Mar  2 05:40:05 2021
    Local: Mon Mar  1 21:40:05 2021

44.10 `localtime()`

Convert a calendar time into a broken-down local time

Synopsis

    #include <time.h>
    
    struct tm *localtime(const time_t *timer);

Description

If you have a time_t, you can run it through this function to get a struct tm back full of the corresponding broken-down local time information.

This is just like gmtime(), except it does local time instead of UTC.

Once you have that struct tm, you can feed it to strftime() to print it out.

WARNING: This function returns a pointer to a static struct tm* region that isn’t thread-safe and might be shared with the gmtime() function. If you need thread safety use a mutex that covers gmtime() and localtime().

Return Value

Returns a pointer to the broken-down local time, or NULL if it can’t be obtained.

Example

time_t now = time(NULL);

printf("Local: %s", asctime(localtime(&now)));
printf("UTC  : %s", asctime(gmtime(&now)));

Sample output:

    Local: Mon Mar  1 21:40:05 2021
    UTC  : Tue Mar  2 05:40:05 2021

44.11 `strftime()`

Formatted date and time output

Synopsis

    #include <time.h>
    
    size_t strftime(char * restrict s, size_t maxsize,
                    const char * restrict format,
                    const struct tm * restrict timeptr);

Description

This is the sprintf() of date and time functions. It’ll take a struct tm and produce a string in just about whatever form you desire, for example:

    2021-03-01
    Monday, March 1 at 9:54 PM
    It's Monday!

It’s a super flexible version of asctime(). And thread-safe, besides, since it doesn’t rely on a static buffer to hold the results.

Basically what you do is give it a destination, s, and its max size in bytes in maxsize. Also, provide a format string that’s analogous to printf()’s format string, but with different format specifiers. And lastly, a struct tm with the broken-down time information to use for printing.

The format string works like this, for example:

    "It's %A, %B %d!"

Which produces:

    It's Monday, March 1!

The %A is the full day-of-week name, the %B is the full month name, and the %d is the day of the month. strftime() substitutes the right thing to produce the result. Brilliant!

So what are all the format specifiers? Glad you asked!

I’m going to be lazy and just drop this table in right from the spec.

Specifier	Description
`%a`	Locale’s abbreviated weekday name. [`tm_wday`]
`%A`	Locale’s full weekday name. [`tm_wday`]
`%b`	Locale’s abbreviated month name. [`tm_mon`]
`%B`	Locale’s full month name. [`tm_mon`]
`%c`	Locale’s appropriate date and time representation.
`%C`	Year divided by 100 and truncated to an integer, as a decimal number (00–99). [`tm_year`]
`%d`	Day of the month as a decimal number (01–31). [`tm_mday`]
`%D`	Equivalent to `"%m/%d/%y"`. [`tm_mon`, `tm_mday`, `tm_year`]
`%e`	Day of the month as a decimal number (1–31); a single digit is preceded by a space. [`tm_mday`]
`%F`	Equivalent to “%Y-%m-%d” (the ISO 8601 date format). [`tm_year`, `tm_mon`, `tm_mday`]
`%g`	Last 2 digits of the week-based year (see below) as a decimal number (00–99). [`tm_year`, `tm_wday`, `tm_yday`]
`%G`	Week-based year (see below) as a decimal number (e.g., 1997). [`tm_year`, `tm_wday`, `tm_yday`]
`%h`	Equivalent to “%b”. [`tm_mon`]
`%H`	Hour (24-hour clock) as a decimal number (00–23). [`tm_hour`]
`%I`	Hour (12-hour clock) as a decimal number (01–12). [`tm_hour`]
`%j`	Day of the year as a decimal number (001–366). [`tm_yday`]
`%m`	Month as a decimal number (01–12).
`%M`	Minute as a decimal number (00–59). [`tm_min`]
`%n`	A new-line character.
`%p`	Locale’s equivalent of the AM/PM designations associated with a 12-hour clock. [`tm_hour`]
`%r`	Locale’s 12-hour clock time. [`tm_hour`, `tm_min`, `tm_sec`]
`%R`	Equivalent to `"%H:%M"`. [`tm_hour`, `tm_min`]
`%S`	Second as a decimal number (00–60). [`tm_sec`]
`%t`	A horizontal-tab character.
`%T`	Equivalent to `"%H:%M:%S"` (the ISO 8601 time format). [`tm_hour`, `tm_min`, `tm_sec`]
`%u`	ISO 8601 weekday as a decimal number (1–7), where Monday is 1. [`tm_wday`]
`%U`	Week number of the year (the first Sunday as the first day of week 1) as a decimal number (00–53). [`tm_year`, `tm_wday`, `tm_yday`]
`%V`	ISO 8601 week number (see below) as a decimal number (01–53). [`tm_year`, `tm_wday`, `tm_yday`]
`%w`	Weekday as a decimal number (0–6), where Sunday is 0.
`%W`	Week number of the year (the first Monday as the first day of week 1) as a decimal number (00–53). [`tm_year`, `tm_wday`, `tm_yday`]
`%x`	Locale’s appropriate date representation.
`%X`	Locale’s appropriate time representation.
`%y`	Last 2 digits of the year as a decimal number (00–99). [`tm_year`]
`%Y`	Year as a decimal number (e.g., 1997). [`tm_year`]
`%z`	Offset from UTC in the ISO 8601 format `"-0430"` (meaning 4 hours 30 minutes behind UTC, west of Greenwich), or by no characters if no time zone is determinable. [`tm_isdst`]
`%Z`	Locale’s time zone name or abbreviation, or by no characters if no time zone is determinable. [`tm_isdst`]
`%%`	A plain ol’ %

Phew. That’s love.

%G, %g, and %v are a little funky in that they use something called the ISO 8601 week-based year. I’d never heard of it. But, again stealing from the spec, these are the rules:

%g, %G, and %V give values according to the ISO 8601 week-based year. In this system, weeks begin on a Monday and week 1 of the year is the week that includes January 4th, which is also the week that includes the first Thursday of the year, and is also the first week that contains at least four days in the year. If the first Monday of January is the 2nd, 3rd, or 4th, the preceding days are part of the last week of the preceding year; thus, for Saturday 2nd January 1999, %G is replaced by 1998 and %V is replaced by 53. If December 29th, 30th, or 31st is a Monday, it and any following days are part of week 1 of the following year. Thus, for Tuesday 30th December 1997, %G is replaced by 1998 and %V is replaced by 01.

Learn something new every day! If you want to know more, Wikipedia has a page on it ²¹³.

If you’re in the “C” locale, the specifiers produce the following (again, stolen from the spec):

Specifier	Description
`%a`	The first three characters of `%A`.
`%A`	One of `Sunday`, `Monday`, … , `Saturday`.
`%b`	The first three characters of `%B`.
`%B`	One of `January`, `February`, … , `December`.
`%c`	Equivalent to `%a %b %e %T %Y`.
`%p`	One of `AM` or `PM`.
`%r`	Equivalent to `%I:%M:%S %p`.
`%x`	Equivalent to `%m/%d/%y`.
`%X`	Equivalent to `%T`.
`%Z`	Implementation-defined.

There are additional variants of the format specifiers that indicate you want to use a locale’s alternative format. These don’t exist for all locales. It’s one of the format specifies above, with either an E or O prefix:

    %Ec %EC %Ex %EX %Ey %EY %Od %Oe %OH %OI
    %Om %OM %OS %Ou %OU %OV %Ow %OW %Oy

The E and O prefixes are ignored in the “C” locale.

Return Value

Returns the total number of bytes put into the result string, not including the NUL terminator.

If the result doesn’t fit in the string, zero is returned and the value in s is indeterminate.

Example

#include <stdio.h>
#include <time.h>

int main(void)
{
    char s[128];
    time_t now = time(NULL);

    // %c: print date as per current locale
    strftime(s, sizeof s, "%c", localtime(&now));
    puts(s);   // Sun Feb 28 22:29:00 2021

    // %A: full weekday name
    // %B: full month name
    // %d: day of the month
    strftime(s, sizeof s, "%A, %B %d", localtime(&now));
    puts(s);   // Sunday, February 28

    // %I: hour (12 hour clock)
    // %M: minute
    // %S: second
    // %p: AM or PM
    strftime(s, sizeof s, "It's %I:%M:%S %p", localtime(&now));
    puts(s);   // It's 10:29:00 PM

    // %F: ISO 8601 yyyy-mm-dd
    // %T: ISO 8601 hh:mm:ss
    // %z: ISO 8601 timezone offset
    strftime(s, sizeof s, "ISO 8601: %FT%T%z", localtime(&now));
    puts(s);   // ISO 8601: 2021-02-28T22:29:00-0800
}

45 `<ctype.h>` Character Classification and Conversion

Function	Description
`isalnum()`	Tests if a character is alphabetic or is a digit
`isalpha()`	Returns true if a character is alphabetic
`isblank()`	Tests if a character is word-separating whitespace
`iscntrl()`	Test if a character is a control character
`isdigit()`	Tests if a character is a digit
`isgraph()`	Tests if the character is printable and not a space
`islower()`	Tests if a character is lowercase
`isprint()`	Tests if a character is printable
`ispunct()`	Test if a character is punctuation
`isspace()`	Test if a character is whitespace
`isupper()`	Tests if a character is uppercase
`isxdigit()`	Tests if a character is a hexadecimal digit
`tolower()`	Convert a letter to lowercase
`toupper()`	Convert a letter to uppercase

This collection of macros is good for testing characters to see if they’re of a certain class, such as alphabetic, numeric, control characters, etc.

Surprisingly, they take int arguments instead of some kind of char. This is so you can feed EOF in for convenience if you have an integer representation of that. If not EOF, the value passed in has to be representable in an unsigned char. Otherwise it’s (dun dun DUUNNNN) undefined behavior. So you can forget about passing in your UTF-8 multibyte characters.

Also, the behavior of these functions varies based on locale.

In many of the pages in this section, I give some examples. These are from the “C” locale, and might vary if you’ve set a different locale.

Note that wide characters have their own set of classification functions, so don’t try to use these on wchar_ts. Or else!

45.1 `isalnum()`

Tests if a character is alphabetic or is a digit

Synopsis

    #include <ctype.h>
    
    int isalnum(int c);

Description

Tests if a character is alphabetic (A-Z or a-z) or a digit (0-9).

Is equivalent to:

    isalpha(c) || isdigit(c)

Return Value

Returns true if a character is alphabetic (A-Z or a-z) or a digit (0-9).

Example

//             testing this char
//                      v
printf("%s\n", isalnum('a')? "yes": "no");  // yes
printf("%s\n", isalnum('B')? "yes": "no");  // yes
printf("%s\n", isalnum('5')? "yes": "no");  // yes
printf("%s\n", isalnum('?')? "yes": "no");  // no

45.2 `isalpha()`

Returns true if a character is alphabetic

Synopsis

    #include <ctype.h>
    
    int isalpha(int c);

Description

Returns true for alphabetic characters (A-Z or a-z).

Technically (and in the “C” locale) equivalent to:

    isupper(c) || islower(c)

Extra super technically, because I know you’re dying for this to be extra unnecessarily complex, it can also include some locale-specific characters for which this is true:

    !iscntrl(c) && !isdigit(c) && !ispunct(c) && !isspace(c)

and this is true:

    isupper(c) || islower(c)

Return Value

Returns true for alphabetic characters (A-Z or a-z).

Or for any of the other crazy stuff in the description, above.

Example

//             testing this char
//                      v
printf("%s\n", isalpha('a')? "yes": "no");  // yes
printf("%s\n", isalpha('B')? "yes": "no");  // yes
printf("%s\n", isalpha('5')? "yes": "no");  // no
printf("%s\n", isalpha('?')? "yes": "no");  // no

45.3 `isblank()`

Tests if a character is word-separating whitespace

Synopsis

    #include <ctype.h>
    
    int isblank(int c);

Description

True if the character is a whitespace character used to separate words in a single line.

For example, space (' ') or horizontal tab ('\t'). Other locales might define other blank characters.

Return Value

Returns true if the character is a whitespace character used to separate words in a single line.

Example

//             testing this char
//                      v
printf("%s\n", isblank(' ')? "yes": "no");   // yes
printf("%s\n", isblank('\t')? "yes": "no");  // yes
printf("%s\n", isblank('\n')? "yes": "no");  // no
printf("%s\n", isblank('a')? "yes": "no");   // no
printf("%s\n", isblank('?')? "yes": "no");   // no

45.4 `iscntrl()`

Test if a character is a control character

Synopsis

    #include <ctype.h>
    
    int iscntrl(int c);

Description

A control character is a local-specific non-printing character.

For the “C” locale, this means control characters are in the range 0x00 to 0x1F (the character right before SPACE) and 0x7F (the DEL character).

Basically if it’s not an ASCII (or Unicode less than 128) printable character, it’s a control character in the “C” locale.

Return Value

Returns true if c is a control character.

Example

//             testing this char
//                      v
printf("%s\n", iscntrl('\t')? "yes": "no");  // yes (tab)
printf("%s\n", iscntrl('\n')? "yes": "no");  // yes (newline)
printf("%s\n", iscntrl('\r')? "yes": "no");  // yes (return)
printf("%s\n", iscntrl('\a')? "yes": "no");  // yes (bell)
printf("%s\n", iscntrl(' ')? "yes": "no");   // no
printf("%s\n", iscntrl('a')? "yes": "no");   // no
printf("%s\n", iscntrl('?')? "yes": "no");   // no

45.5 `isdigit()`

Tests if a character is a digit

Synopsis

    #include <ctype.h>
    
    int isdigit(int c);

Description

Tests if c is a digit in the range 0-9.

Return Value

Returns true if the character is a digit, unsurprisingly.

Example

//             testing this char
//                      v
printf("%s\n", isdigit('0')? "yes": "no");   // yes
printf("%s\n", isdigit('5')? "yes": "no");   // yes
printf("%s\n", isdigit('a')? "yes": "no");   // no
printf("%s\n", isdigit('B')? "yes": "no");   // no
printf("%s\n", isdigit('?')? "yes": "no");   // no

45.6 `isgraph()`

Tests if the character is printable and not a space

Synopsis

    #include <ctype.h>
    
    int isgraph(int c);

Description

Tests if c is any printable character that isn’t a space (' ').

Return Value

Returns true if c is any printable character that isn’t a space (' ').

Example

//             testing this char
//                      v
printf("%s\n", isgraph('0')? "yes": "no");   // yes
printf("%s\n", isgraph('a')? "yes": "no");   // yes
printf("%s\n", isgraph('B')? "yes": "no");   // yes
printf("%s\n", isgraph('?')? "yes": "no");   // yes
printf("%s\n", isgraph(' ')? "yes": "no");   // no
printf("%s\n", isgraph('\n')? "yes": "no");  // no

45.7 `islower()`

Tests if a character is lowercase

Synopsis

    #include <ctype.h>
    
    int islower(int c);

Description

Tests if a character is lowercase, in the range a-z.

In other locales, there could be other lowercase characters. In all cases, to be lowercase, the following must be true:

    !iscntrl(c) && !isdigit(c) && !ispunct(c) && !isspace(c)

Return Value

Returns true if the character is lowercase.

Example

//             testing this char
//                      v
printf("%s\n", islower('c')? "yes": "no");   // yes
printf("%s\n", islower('0')? "yes": "no");   // no
printf("%s\n", islower('B')? "yes": "no");   // no
printf("%s\n", islower('?')? "yes": "no");   // no
printf("%s\n", islower(' ')? "yes": "no");   // no

45.8 `isprint()`

Tests if a character is printable

Synopsis

    #include <ctype.h>
    
    int isprint(int c);

Description

Tests if a character is printable, including space (' '). So like isgraph(), except space isn’t left out in the cold.

Return Value

Returns true if the character is printable, including space (' ').

Example

//             testing this char
//                      v
printf("%s\n", isprint('c')? "yes": "no");   // yes
printf("%s\n", isprint('0')? "yes": "no");   // yes
printf("%s\n", isprint(' ')? "yes": "no");   // yes
printf("%s\n", isprint('\r')? "yes": "no");  // no

45.9 `ispunct()`

Test if a character is punctuation

Synopsis

    #include <ctype.h>
    
    int ispunct(int c);

Description

Tests if a character is punctuation.

In the “C” locale, this means:

    !isspace(c) && !isalnum(c)

In other locales, there could be other punctuation characters (but they also can’t be space or alphanumeric).

Return Value

True if the character is punctuation.

Example

//             testing this char
//                      v
printf("%s\n", ispunct(',')? "yes": "no");   // yes
printf("%s\n", ispunct('!')? "yes": "no");   // yes
printf("%s\n", ispunct('c')? "yes": "no");   // no
printf("%s\n", ispunct('0')? "yes": "no");   // no
printf("%s\n", ispunct(' ')? "yes": "no");   // no
printf("%s\n", ispunct('\n')? "yes": "no");  // no

45.10 `isspace()`

Test if a character is whitespace

Synopsis

    #include <ctype.h>
    
    int isspace(int c);

Description

Tests if c is a whitespace character. These are:

Space (' ')
Formfeed ('\f')
Newline ('\n')
Carriage Return ('\r')
Horizontal Tab ('\t')
Vertical Tab ('\v')

Other locales might specify other whitespace characters. isalnum() is false for all whitespace characters.

Return Value

True if the character is whitespace.

Example

//             testing this char
//                      v
printf("%s\n", isspace(' ')? "yes": "no");   // yes
printf("%s\n", isspace('\n')? "yes": "no");  // yes
printf("%s\n", isspace('\t')? "yes": "no");  // yes
printf("%s\n", isspace(',')? "yes": "no");   // no
printf("%s\n", isspace('!')? "yes": "no");   // no
printf("%s\n", isspace('c')? "yes": "no");   // no

45.11 `isupper()`

Tests if a character is uppercase

Synopsis

    #include <ctype.h>
    
    int isupper(int c);

Description

Tests if a character is uppercase, in the range A-Z.

In other locales, there could be other uppercase characters. In all cases, to be uppercase, the following must be true:

    !iscntrl(c) && !isdigit(c) && !ispunct(c) && !isspace(c)

Return Value

Returns true if the character is uppercase.

Example

//             testing this char
//                      v
printf("%s\n", isupper('B')? "yes": "no");   // yes
printf("%s\n", isupper('c')? "yes": "no");   // no
printf("%s\n", isupper('0')? "yes": "no");   // no
printf("%s\n", isupper('?')? "yes": "no");   // no
printf("%s\n", isupper(' ')? "yes": "no");   // no

45.12 `isxdigit()`

Tests if a character is a hexadecimal digit

Synopsis

    #include <ctype.h>
    
    int isxdigit(int c);

Description

Returns true if the character is a hexadecimal digit. Namely if it’s 0-9, a-f, or A-F.

Return Value

True if the character is a hexadecimal digit.

Example

//             testing this char
//                      v
printf("%s\n", isxdigit('B')? "yes": "no");   // yes
printf("%s\n", isxdigit('c')? "yes": "no");   // yes
printf("%s\n", isxdigit('2')? "yes": "no");   // yes
printf("%s\n", isxdigit('G')? "yes": "no");   // no
printf("%s\n", isxdigit('?')? "yes": "no");   // no

45.13 `tolower()`

Convert a letter to lowercase

Synopsis

    #include <ctype.h>
    
    int tolower(int c);

Description

If the character is uppercase (i.e. isupper(c) is true), this function returns the corresponding lowercase letter.

Different locales might have different upper- and lowercase letters.

Return Value

Returns the lowercase value for an uppercase letter. If the letter isn’t uppercase, returns it unchanged.

Example

//             changing this char
//                      v
printf("%c\n", tolower('B'));  // b (made lowercase!)
printf("%c\n", tolower('e'));  // e (unchanged)
printf("%c\n", tolower('!'));  // ! (unchanged)

45.14 `toupper()`

Convert a letter to uppercase

Synopsis

    #include <ctype.h>
    
    int toupper(int c);

Description

If the character is lower (i.e. islower(c) is true), this function returns the corresponding uppercase letter.

Different locales might have different upper- and lowercase letters.

Return Value

Returns the uppercase value for a lowercase letter. If the letter isn’t lowercase, returns it unchanged.

Example

//             changing this char
//                      v
printf("%c\n", toupper('B'));  // B (unchanged)
printf("%c\n", toupper('e'));  // E (made uppercase!)
printf("%c\n", toupper('!'));  // ! (unchanged)

46 `<threads.h>` Multithreading Functions

Function	Description
`call_once()`	Call a function one time no matter how many threads try
`cnd_broadcast()`	Wake up all threads waiting on a condition variable
`cnd_destroy()`	Free up resources from a condition variable
`cnd_init()`	Initialize a condition variable to make it ready for use
`cnd_signal()`	Wake up a thread waiting on a condition variable
`cnd_timedwait()`	Wait on a condition variable with a timeout
`cnd_wait()`	Wait for a signal on a condition variable
`mtx_destroy()`	Cleanup a mutex when done with it
`mtx_init()`	Initialize a mutex for use
`mtx_lock()`	Acquire a lock on a mutex
`mtx_timedlock()`	Lock a mutex allowing for timeout
`mtx_trylock()`	Try to lock a mutex, returning if not possible
`mtx_unlock()`	Free a mutex when you’re done with the critical section
`thrd_create()`	Create a new thread of execution
`thrd_current()`	Get the ID of the calling thread
`thrd_detach()`	Automatically clean up threads when they exit
`thrd_equal()`	Compare two thread descriptors for equality
`thrd_exit()`	Stop and exit this thread
`thrd_join()`	Wait for a thread to exit
`thrd_yield()`	Stop running that other threads might run
`tss_create()`	Create new thread-specific storage
`tss_delete()`	Clean up a thread-specific storage variable
`tss_get()`	Get thread-specific data
`tss_set()`	Set thread-specific data

We have a bunch of good things at our disposal with this one:

Threads
Mutexes
Condition Variables
Thread-Specific Storage
And, last but not least, the always-fun call_once() function!

Enjoy!

46.1 `call_once()`

Call a function one time no matter how many threads try

Synopsis

    #include <threads.h>
    
    void call_once(once_flag *flag, void (*func)(void));

Description

If you have a bunch of threads running over the same piece of code that calls a function, but you only want that function to run one time, call_once() can help you out.

The catch is the function that is called doesn’t return anything and takes no arguments.

If you need more than that, you’ll have to set a threadsafe flag such as atomic_flag, or one that you protect with a mutex.

To use this, you need to pass it a pointer to a function to execute, func, and also a pointer to a flag of type once_flag.

once_flag is an opaque type, so all you need to know is that you initialize it to the value ONCE_FLAG_INIT.

Return Value

Returns nothing.

Example

#include <stdio.h>
#include <threads.h>

once_flag of = ONCE_FLAG_INIT;  // Initialize it like this

void run_once_function(void)
{
    printf("I'll only run once!\n");
}

int run(void *arg)
{
    (void)arg;

    printf("Thread running!\n");

    call_once(&of, run_once_function);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);
}

Output (might vary per run):

    Thread running!
    Thread running!
    I'll only run once!
    Thread running!
    Thread running!
    Thread running!

46.2 `cnd_broadcast()`

Wake up all threads waiting on a condition variable

Synopsis

    #include <threads.h>
    
    int cnd_broadcast(cnd_t *cond);

Description

This is just like cnd_signal() in that it wakes up threads that are waiting on a condition variable…. except instead of just rousing one thread, it wakes them all.

Of course, only one will get the mutex, and the rest will have to wait their turn. But instead of being asleep waiting for a signal, they’ll be asleep waiting to reacquire the mutex. They’re rearin’ to go, in other words.

This can make a difference in a specific set of circumstances where cnd_signal() might leave you hanging.

If you’re relying on subsequent threads to issue the next cnd_signal(), but you have the cnd_wait() in a while loop²¹⁴ that doesn’t allow any threads to escape, you’ll be stuck. No more threads will be woken up from the wait.

But if you cnd_broadcast(), all the threads will be woken, and presumably at least one of them will be allowed to escape the while loop, freeing it up to broadcast the next wakeup when its work is done.

Return Value

Returns thrd_success or thrd_error depending on how well things went.

Example

In the example below, we launch a bunch of threads, but they’re only allowed to run if their ID matches the current ID. If it doesn’t, they go back to waiting.

If you cnd_signal() to wake the next thread, it might not be the one with the proper ID to run. If it’s not, it goes back to sleep and we hang (because no thread is awake to hit cnd_signal() again).

But if you cnd_broadcast() to wake them all, then they’ll all try (one after another) to get out of the while loop. And one of them will make it.

Try switching the cnd_broadcast() to cnd_signal() to see likely deadlocks. It doesn’t happen every time, but usually does.

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    int id = *(int*)arg;

    static int current_id = 0;

    mtx_lock(&mutex);

    while (id != current_id) {
        printf("THREAD %d: waiting\n", id);
        cnd_wait(&condvar, &mutex);

        if (id != current_id)
            printf("THREAD %d: woke up, but it's not my turn!\n", id);
        else
            printf("THREAD %d: woke up, my turn! Let's go!\n", id);
    }

    current_id++;

    printf("THREAD %d: signaling thread %d to run\n", id, current_id);

    //cnd_signal(&condvar);
    cnd_broadcast(&condvar);
    mtx_unlock(&mutex);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];
    int id[] = {4, 3, 2, 1, 0};

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, id + i);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);
}

Example run with cnd_broadcast():

    THREAD 4: waiting
    THREAD 1: waiting
    THREAD 3: waiting
    THREAD 2: waiting
    THREAD 0: signaling thread 1 to run
    THREAD 2: woke up, but it's not my turn!
    THREAD 2: waiting
    THREAD 4: woke up, but it's not my turn!
    THREAD 4: waiting
    THREAD 3: woke up, but it's not my turn!
    THREAD 3: waiting
    THREAD 1: woke up, my turn! Let's go!
    THREAD 1: signaling thread 2 to run
    THREAD 4: woke up, but it's not my turn!
    THREAD 4: waiting
    THREAD 3: woke up, but it's not my turn!
    THREAD 3: waiting
    THREAD 2: woke up, my turn! Let's go!
    THREAD 2: signaling thread 3 to run
    THREAD 4: woke up, but it's not my turn!
    THREAD 4: waiting
    THREAD 3: woke up, my turn! Let's go!
    THREAD 3: signaling thread 4 to run
    THREAD 4: woke up, my turn! Let's go!
    THREAD 4: signaling thread 5 to run

Example run with cnd_signal():

    THREAD 4: waiting
    THREAD 1: waiting
    THREAD 3: waiting
    THREAD 2: waiting
    THREAD 0: signaling thread 1 to run
    THREAD 4: woke up, but it's not my turn!
    THREAD 4: waiting
    
    [deadlock at this point]

See how THREAD 0 signaled that it was THREAD 1’s turn? But—bad news—it was THREAD 4 that got woken up. So no one continued the process. cnd_broadcast() would have woken them all, so eventually THREAD 1 would have run, gotten out of the while, and broadcast for the next thread to run.

46.3 `cnd_destroy()`

Free up resources from a condition variable

Synopsis

    #include <threads.h>
    
    void cnd_destroy(cnd_t *cond);

Description

This is the opposite of cnd_init() and should be called when all threads are done using a condition variable.

Return Value

Returns nothing!

Example

General-purpose condition variable example here, but you can see the cnd_destroy() down at the end.

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    mtx_lock(&mutex);

    printf("Thread: waiting...\n");
    cnd_wait(&condvar, &mutex);
    printf("Thread: running again!\n");

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 0.1s to allow the other thread to wait
    thrd_sleep(&(struct timespec){.tv_nsec=100000000L}, NULL);

    mtx_lock(&mutex);
    printf("Main: signaling thread\n");
    cnd_signal(&condvar);
    mtx_unlock(&mutex);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);  // <-- DESTROY CONDITION VARIABLE
}

Output:

    Main creating thread
    Thread: waiting...
    Main: signaling thread
    Thread: running again!

46.4 `cnd_init()`

Initialize a condition variable to make it ready for use

Synopsis

    #include <threads.h>
    
    int cnd_init(cnd_t *cond);

Description

This is the opposite of cnd_destroy(). This prepares a condition variable for use, doing behind-the-scenes work on it.

Don’t use a condition variable without calling this first!

Return Value

If all goes well, returns thrd_success. It all doesn’t go well, it could return thrd_nomem if the system is out of memory, or thread_error in the case of any other error.

Example

General-purpose condition variable example here, but you can see the cnd_init() down at the start of main().

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    mtx_lock(&mutex);

    printf("Thread: waiting...\n");
    cnd_wait(&condvar, &mutex);
    printf("Thread: running again!\n");

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);      // <-- INITIALIZE CONDITION VARIABLE

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 0.1s to allow the other thread to wait
    thrd_sleep(&(struct timespec){.tv_nsec=100000000L}, NULL);

    mtx_lock(&mutex);
    printf("Main: signaling thread\n");
    cnd_signal(&condvar);
    mtx_unlock(&mutex);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);
}

Output:

    Main creating thread
    Thread: waiting...
    Main: signaling thread
    Thread: running again!

46.5 `cnd_signal()`

Wake up a thread waiting on a condition variable

Synopsis

    #include <threads.h>
    
    int cnd_signal(cnd_t *cond);

Description

If you have a thread (or a bunch of threads) waiting on a condition variable, this function will wake one of them up to run.

Compare to cnd_broadcast() that wakes up all the threads. See the cnd_broadcast() page for more information on when you’re want to use that versus this.

Return Value

Returns thrd_success or thrd_error depending on how happy your program is.

Example

General-purpose condition variable example here, but you can see the cnd_signal() in the middle of main().

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    mtx_lock(&mutex);

    printf("Thread: waiting...\n");
    cnd_wait(&condvar, &mutex);
    printf("Thread: running again!\n");

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 0.1s to allow the other thread to wait
    thrd_sleep(&(struct timespec){.tv_nsec=100000000L}, NULL);

    mtx_lock(&mutex);
    printf("Main: signaling thread\n");
    cnd_signal(&condvar);    // <-- SIGNAL CHILD THREAD HERE!
    mtx_unlock(&mutex);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);
}

Output:

    Main creating thread
    Thread: waiting...
    Main: signaling thread
    Thread: running again!

46.6 `cnd_timedwait()`

Wait on a condition variable with a timeout

Synopsis

    #include <threads.h>
    
    int cnd_timedwait(cnd_t *restrict cond, mtx_t *restrict mtx,
                      const struct timespec *restrict ts);

Description

This is like cnd_wait() except we get to specify a timeout, as well.

Note that the thread still must reacquire the mutex to get more work done even after the timeout. The the main difference is that regular cnd_wait() will only try to get the mutex after a cnd_signal() or cnd_broadcast(), whereas cnd_timedwait() will do that, too, and try to get the mutex after the timeout.

The timeout is specified as an absolute UTC time since Epoch. You can get this with the timespec_get() function and then add values on to the result to timeout later than now, as shown in the example.

Beware that you can’t have more than 999999999 nanoseconds in the tv_nsec field of the struct timespec. Mod those so they stay in range.

Return Value

If the thread wakes up for a non-timeout reason (e.g. signal or broadcast), returns thrd_success. If woken up due to timeout, returns thrd_timedout. Otherwise returns thrd_error.

Example

This example has a thread wait on a condition variable for a maximum of 1.75 seconds. And it always times out because no one ever sends a signal. Tragic.

#include <stdio.h>
#include <time.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    mtx_lock(&mutex);

    struct timespec ts;

    // Get the time now
    timespec_get(&ts, TIME_UTC);

    // Add on 1.75 seconds from now
    ts.tv_sec += 1;
    ts.tv_nsec += 750000000L;

    // Handle nsec overflow
    ts.tv_sec += ts.tv_nsec / 1000000000L;
    ts.tv_nsec = ts.tv_nsec % 1000000000L;

    printf("Thread: waiting...\n");
    int r = cnd_timedwait(&condvar, &mutex, &ts);

    switch (r) {
        case thrd_success:
            printf("Thread: signaled!\n");
            break;

        case thrd_timedout:
            printf("Thread: timed out!\n");
            return 1;

        case thrd_error:
            printf("Thread: Some kind of error\n");
            return 2;
    }

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 3s to allow the other thread to timeout
    thrd_sleep(&(struct timespec){.tv_sec=3}, NULL);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);
}

Output:

    Main creating thread
    Thread: waiting...
    Thread: timed out!

46.7 `cnd_wait()`

Wait for a signal on a condition variable

Synopsis

    #include <threads.h>
    
    int cnd_wait(cnd_t *cond, mtx_t *mtx);

Description

This puts the calling thread to sleep until it is awakened by a call to cnd_signal() or cnd_broadcast().

Return Value

If everything’s fantastic, returns thrd_success. Otherwise it returns thrd_error to report that something has gone fantastically, horribly awry.

Example

General-purpose condition variable example here, but you can see the cnd_wait() in the run() function.

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    mtx_lock(&mutex);

    printf("Thread: waiting...\n");
    cnd_wait(&condvar, &mutex);       // <-- WAIT HERE!
    printf("Thread: running again!\n");

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);
    cnd_init(&condvar);

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 0.1s to allow the other thread to wait
    thrd_sleep(&(struct timespec){.tv_nsec=100000000L}, NULL);

    mtx_lock(&mutex);
    printf("Main: signaling thread\n");
    cnd_signal(&condvar);    // <-- SIGNAL CHILD THREAD HERE!
    mtx_unlock(&mutex);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
    cnd_destroy(&condvar);
}

Output:

    Main creating thread
    Thread: waiting...
    Main: signaling thread
    Thread: running again!

46.8 `mtx_destroy()`

Cleanup a mutex when done with it

Synopsis

    #include <threads.h>
    
    void mtx_destroy(mtx_t *mtx);

Description

The opposite of mtx_init(), this function frees up any resources associated with the given mutex.

You should call this when all threads are done using the mutex.

Return Value

Returns nothing, the selfish ingrate!

Example

General-purpose mutex example here, but you can see the mtx_destroy() down at the end.

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    static int count = 0;

    mtx_lock(&mutex);

    printf("Thread: I got %d!\n", count);
    count++;

    mtx_unlock(&mutex);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    mtx_init(&mutex, mtx_plain);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);   // <-- DESTROY THE MUTEX HERE
}

Output:

    Thread: I got 0!
    Thread: I got 1!
    Thread: I got 2!
    Thread: I got 3!
    Thread: I got 4!

46.9 `mtx_init()`

Initialize a mutex for use

Synopsis

    #include <threads.h>
    
    int mtx_init(mtx_t *mtx, int type);

Description

Before you can use a mutex variable, you have to initialize it with this call to get it all prepped and ready to go.

But wait! It’s not quite that simple. You have to tell it what type of mutex you want to create.

Type	Description
`mtx_plain`	Regular ol’ mutex
`mtx_timed`	Mutex that supports timeouts
`mtx_plain\|mtx_recursive`	Recursive mutex
`mtx_timed\|mtx_recursive`	Recursive mutex that supports timeouts

As you can see, you can make a plain or timed mutex recursive by bitwise-ORing the value with mtx_recursive.

And the timeout gives a thread a chance to try to get the lock for a while, but then bail out if it can’t get it in that timeframe. You use the mtx_timedlock() function with mtx_timed mutexes.

Return Value

Returns thrd_success in a perfect world, and potentially thrd_error in an imperfect one.

Example

General-purpose mutex example here, but you can see the mtx_init() down at the top of main():

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    static int count = 0;

    mtx_lock(&mutex);

    printf("Thread: I got %d!\n", count);
    count++;

    mtx_unlock(&mutex);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    mtx_init(&mutex, mtx_plain);  // <-- CREATE THE MUTEX HERE

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);   // <-- DESTROY THE MUTEX HERE
}

Output:

    Thread: I got 0!
    Thread: I got 1!
    Thread: I got 2!
    Thread: I got 3!
    Thread: I got 4!

46.10 `mtx_lock()`

Acquire a lock on a mutex

Synopsis

    #include <threads.h>
    
    int mtx_lock(mtx_t *mtx);

Description

If you’re a thread and want to enter a critical section, do I have the function for you!

A thread that calls this function will wait until it can acquire the mutex, then it will grab it, wake up, and run!

If the mutex is recursive and is already locked by this thread, it will be locked again and the lock count will increase. If the mutex is not recursive and the thread already holds it, this call will error out.

Return Value

Returns thrd_success on goodness and thrd_error on badness.

Example

General-purpose mutex example here, but you can see the mtx_lock() in the run() function:

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    static int count = 0;

    mtx_lock(&mutex);  // <-- LOCK HERE

    printf("Thread: I got %d!\n", count);
    count++;

    mtx_unlock(&mutex);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    mtx_init(&mutex, mtx_plain);  // <-- CREATE THE MUTEX HERE

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);   // <-- DESTROY THE MUTEX HERE
}

Output:

    Thread: I got 0!
    Thread: I got 1!
    Thread: I got 2!
    Thread: I got 3!
    Thread: I got 4!

46.11 `mtx_timedlock()`

Lock a mutex allowing for timeout

Synopsis

    #include <threads.h>
    
    int mtx_timedlock(mtx_t *restrict mtx, const struct timespec *restrict ts);

Description

This is just like mtx_lock() except you can add a timeout if you don’t want to wait forever.

Beware that you can’t have more than 999999999 nanoseconds in the tv_nsec field of the struct timespec. Mod those so they stay in range.

Return Value

If everything works and the mutex is obtained, returns thrd_success. If a timeout happens first, returns thrd_timedout.

Otherwise, returns thrd_error. Because if nothing is right, everything is wrong.

Example

This example has a thread wait on a mutex for a maximum of 1.75 seconds. And it always times out because no one ever sends a signal.

#include <stdio.h>
#include <time.h>
#include <threads.h>

mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    struct timespec ts;

    // Get the time now
    timespec_get(&ts, TIME_UTC);

    // Add on 1.75 seconds from now
    ts.tv_sec += 1;
    ts.tv_nsec += 750000000L;

    // Handle nsec overflow
    ts.tv_sec += ts.tv_nsec / 1000000000L;
    ts.tv_nsec = ts.tv_nsec % 1000000000L;

    printf("Thread: waiting for lock...\n");
    int r = mtx_timedlock(&mutex, &ts);

    switch (r) {
        case thrd_success:
            printf("Thread: grabbed lock!\n");
            break;

        case thrd_timedout:
            printf("Thread: timed out!\n");
            break;

        case thrd_error:
            printf("Thread: Some kind of error\n");
            break;
    }

    mtx_unlock(&mutex);

    return 0;
}

int main(void)
{
    thrd_t t;

    mtx_init(&mutex, mtx_plain);

    mtx_lock(&mutex);

    printf("Main creating thread\n");
    thrd_create(&t, run, NULL);

    // Sleep 3s to allow the other thread to timeout
    thrd_sleep(&(struct timespec){.tv_sec=3}, NULL);

    mtx_unlock(&mutex);

    thrd_join(t, NULL);

    mtx_destroy(&mutex);
}

Output:

    Main creating thread
    Thread: waiting for lock...
    Thread: timed out!

46.12 `mtx_trylock()`

Try to lock a mutex, returning if not possible

Synopsis

    #include <threads.h>
    
    int mtx_trylock(mtx_t *mtx);

Description

This works just like mtx_lock except that it returns instantly if a lock can’t be obtained.

The spec notes that there’s a chance that mtx_trylock() might spuriously fail with thrd_busy even if there are no other threads holding the lock. I’m not sure why this is, but you should defensively code against it.

Return Value

Returns thrd_success if all’s well. Or thrd_busy if some other thread holds the lock. Or thrd_error, which means something went right. I mean “wrong”.

Example

#include <stdio.h>
#include <time.h>
#include <threads.h>

mtx_t mutex;

int run(void *arg)
{
    int id = *(int*)arg;

    int r = mtx_trylock(&mutex);   // <-- TRY TO GRAB THE LOCK

    switch (r) {
        case thrd_success:
            printf("Thread %d: grabbed lock!\n", id);
            break;

        case thrd_busy:
            printf("Thread %d: lock already taken :(\n", id);
            return 1;

        case thrd_error:
            printf("Thread %d: Some kind of error\n", id);
            return 2;
    }

    mtx_unlock(&mutex);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];
    int id[THREAD_COUNT];

    mtx_init(&mutex, mtx_plain);

    for (int i = 0; i < THREAD_COUNT; i++) {
        id[i] = i;
        thrd_create(t + i, run, id + i);
    }

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);
}

Output (varies by run):

    Thread 0: grabbed lock!
    Thread 1: lock already taken :(
    Thread 4: lock already taken :(
    Thread 3: grabbed lock!
    Thread 2: lock already taken :(

46.13 `mtx_unlock()`

Free a mutex when you’re done with the critical section

Synopsis

    #include <threads.h>
    
    int mtx_unlock(mtx_t *mtx);

Description

After you’ve done all the dangerous stuff you have to do, wherein the involved threads should not be stepping on each other’s toes… you can free up your stranglehold on the mutex by calling mtx_unlock().

Return Value

Returns thrd_success on success. Or thrd_error on error. It’s not very original in this regard.

Example

General-purpose mutex example here, but you can see the mtx_unlock() in the run() function:

#include <stdio.h>
#include <threads.h>

cnd_t condvar;
mtx_t mutex;

int run(void *arg)
{
    (void)arg;

    static int count = 0;

    mtx_lock(&mutex);

    printf("Thread: I got %d!\n", count);
    count++;

    mtx_unlock(&mutex);  // <-- UNLOCK HERE

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    mtx_init(&mutex, mtx_plain);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    mtx_destroy(&mutex);
}

Output:

    Thread: I got 0!
    Thread: I got 1!
    Thread: I got 2!
    Thread: I got 3!
    Thread: I got 4!

46.14 `thrd_create()`

Create a new thread of execution

Synopsis

    #include <threads.h>
    
    int thrd_create(thrd_t *thr, thrd_start_t func, void *arg);

Description

Now you have the POWER!

Right?

This is how you launch new threads to make your program do multiple things at once²¹⁵!

In order to make this happen, you need to pass a pointer to a thrd_t that will be used to represent the thread you’re spawning.

That thread will start running the function you pass a pointer to in func. This is a value of type thrd_start_t, which is a pointer to a function that returns an int and takes a single void* as a parameter, i.e.:

    int thread_run_func(void *arg)

And, as you might have guessed, the pointer you pass to thrd_create() for the arg parameter is passed on to the func function. This is how you can give additional information to the thread when it starts up.

Of course, for arg, you have to be sure to pass a pointer to an object that is thread-safe or per-thread.

If the thread returns from the function, it exits just as if it had called thrd_exit().

Finally, the value that the func function returns can be picked up by the parent thread with thrd_join().

Return Value

In the case of goodness, returns thrd_success. If you’re out of memory, will return thrd_nomem. Otherwise, thrd_error.

Example

#include <stdio.h>
#include <threads.h>

int run(void *arg)
{
    int id = *(int*)arg;

    printf("Thread %d: I'm alive!!\n", id);

    return id;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];
    int id[THREAD_COUNT];  // One of these per thread

    for (int i = 0; i < THREAD_COUNT; i++) {
        id[i] = i; // Let's pass in the thread number as the ID
        thrd_create(t + i, run, id + i);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        int res;

        thrd_join(t[i], &res);

        printf("Main: thread %d exited with code %d\n", i, res);
    }
}

Output (might vary from run to run):

    Thread 1: I'm alive!!
    Thread 0: I'm alive!!
    Thread 3: I'm alive!!
    Thread 2: I'm alive!!
    Main: thread 0 exited with code 0
    Main: thread 1 exited with code 1
    Main: thread 2 exited with code 2
    Main: thread 3 exited with code 3
    Thread 4: I'm alive!!
    Main: thread 4 exited with code 4

46.15 `thrd_current()`

Get the ID of the calling thread

Synopsis

    #include <threads.h>
    
    thrd_t thrd_current(void);

Description

Each thread has an opaque ID of type thrd_t. This is the value we see get initialized when we call thrd_create().

But what if you want to get the ID of the currently running thread?

No problem! Just call this function and it will be returned to you.

Why? Who knows!

Well, to be honest, I could see it being used a couple places.

You could use it to have a thread detach itself with thrd_detach(). I’m not sure why you’d want to do this, however.
You could use it to compare this thread’s ID with another you have stored in a variable somewhere by using the thrd_equal() function. Seems like the most legit use.
…
Profit!

If anyone has another use, please let me know.

Return Value

Returns the calling thread’s ID.

Example

Here’s a general example that shows getting the current thread ID and comparing it to a previously-recorded thread ID and taking exciting action based on the result! Starring Arnold Schwarzenegger!

#include <stdio.h>
#include <threads.h>

thrd_t first_thread_id;

int run(void *arg)
{
    (void)arg;

    thrd_t my_id = thrd_current();   // <-- GET MY THREAD ID

    if (thrd_equal(my_id, first_thread_id))
        printf("I'm the first thread!\n");
    else
        printf("I'm not the first!\n");

    return 0;
}

int main(void)
{
    thrd_t t;

    thrd_create(&first_thread_id, run, NULL);
    thrd_create(&t, run, NULL);

    thrd_join(first_thread_id, NULL);
    thrd_join(t, NULL);
}

Output:

    Come on, you got what you want, Cohaagen! Give deez people ay-ah!

No, wait, that’s an Arnold Schwarzenegger quote from Total Recall, one of the best science fiction films of all time. Watch it now and then come back to finish this reference page.

Man–what an ending! And Johnny Cab? So excellent. Anyway!

Output:

    I'm the first thread!
    I'm not the first!

46.16 `thrd_detach()`

Automatically clean up threads when they exit

Synopsis

    #include <threads.h>
    
    int thrd_detach(thrd_t thr);

Description

Normally you have to thrd_join() to get resources associated with a deceased thread cleaned up. (Most notably, its exit status is still floating around waiting to get picked up.)

But if you call thrd_detach() on the thread first, manual cleanup isn’t necessary. They just exit and are cleaned up by the OS.

(Note that when the main thread dies, all the threads die in any case.)

Return Value

thrd_success if the thread successfully detaches, thrd_error otherwise.

Example

#include <stdio.h>
#include <threads.h>

thrd_t first_thread_id;

int run(void *arg)
{
    (void)arg;

    printf("Thread running!\n");

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t;

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_create(&t, run, NULL);
        thrd_detach(t);
    }

    // No need to thrd_join()!

    // Sleep a quarter second to let them all finish
    thrd_sleep(&(struct timespec){.tv_nsec=250000000}, NULL);
}

46.17 `thrd_equal()`

Compare two thread descriptors for equality

Synopsis

    #include <threads.h>
    
    int thrd_equal(thrd_t thr0, thrd_t thr1);

Description

If you have two thread descriptors in thrd_t variables, you can test them for equality with this function.

For example, maybe one of the threads has special powers the others don’t, and the run function needs to be able to tell them apart, as in the example.

Return Value

Returns non-zero if the threads are equal. Returns 0 if they’re not.

Example

Here’s a general example that shows getting the current thread ID and comparing it to a previously-recorded thread ID and taking boring action based on the result.

#include <stdio.h>
#include <threads.h>

thrd_t first_thread_id;

int run(void *arg)
{
    (void)arg;

    thrd_t my_id = thrd_current();

    if (thrd_equal(my_id, first_thread_id))  // <-- COMPARE!
        printf("I'm the first thread!\n");
    else
        printf("I'm not the first!\n");

    return 0;
}

int main(void)
{
    thrd_t t;

    thrd_create(&first_thread_id, run, NULL);
    thrd_create(&t, run, NULL);

    thrd_join(first_thread_id, NULL);
    thrd_join(t, NULL);
}

Output:

    I'm the first thread!
    I'm not the first!

46.18 `thrd_exit()`

Stop and exit this thread

Synopsis

    #include <threads.h>
    
    _Noreturn void thrd_exit(int res);

Description

A thread commonly exits by returning from its run function. But if it wants to exit early (perhaps from deeper in the call stack), this function will get that done.

The res code can be picked up by a thread calling thrd_join(), and is equivalent to returning a value from the run function.

Like with returning from the run function, this will also properly clean up all the thread-specific storage associated with this thread—all the destructors for the threads TSS variables will be called. If there are any remaining TSS variables with destructors after the first round of destruction²¹⁶, the remaining destructors will be called. This happens repeatedly until there are no more, or the number of rounds of carnage reaches TSS_DTOR_ITERATIONS.

If the main thread calls this, it’s as if you called exit(EXIT_SUCCESS).

Return Value

This function never returns because the thread calling it is killed in the process. Trippy!

Example

Threads in this example exit early with result 22 if they get a NULL value for arg.

#include <stdio.h>
#include <threads.h>

thrd_t first_thread_id;

int run(void *arg)
{
    (void)arg;

    if (arg == NULL)
        thrd_exit(22);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, i == 2? NULL: "spatula");


    for (int i = 0; i < THREAD_COUNT; i++) {
        int res;
        thrd_join(t[i], &res);

        printf("Thread %d exited with code %d\n", i, res);
    }
}

Output:

    Thread 0 exited with code 0
    Thread 1 exited with code 0
    Thread 2 exited with code 22
    Thread 3 exited with code 0
    Thread 4 exited with code 0

46.19 `thrd_join()`

Wait for a thread to exit

Synopsis

    #include <threads.h>
    
    int thrd_join(thrd_t thr, int *res);

Description

When a parent thread fires off some child threads, it can wait for them to complete with this call

Return Value

Example

Threads in this example exit early with result 22 if they get a NULL value for arg. The parent thread picks up this result code with thrd_join().

#include <stdio.h>
#include <threads.h>

thrd_t first_thread_id;

int run(void *arg)
{
    (void)arg;

    if (arg == NULL)
        thrd_exit(22);

    return 0;
}

#define THREAD_COUNT 5

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, i == 2? NULL: "spatula");


    for (int i = 0; i < THREAD_COUNT; i++) {
        int res;
        thrd_join(t[i], &res);

        printf("Thread %d exited with code %d\n", i, res);
    }
}

Output:

    Thread 0 exited with code 0
    Thread 1 exited with code 0
    Thread 2 exited with code 22
    Thread 3 exited with code 0
    Thread 4 exited with code 0

46.20 `thrd_sleep()`

Sleep for a specific number of seconds and nanoseconds

Synopsis

    #include <threads.h>
    
    int thrd_sleep(const struct timespec *duration, struct timespec *remaining);

Description

This function puts the current thread to sleep for a while²¹⁷ allowing other threads to run.

The calling thread will wake up after the time has elapsed, or if it gets interrupted by a signal or something.

If it doesn’t get interrupted, it’ll sleep at least as long as you asked. Maybe a tad longer. You know how hard it can be to get out of bed.

The structure looks like this:

    struct timespec {
        time_t tv_sec;   // Seconds
        long   tv_nsec;  // Nanoseconds (billionths of a second)
    };

Don’t set tv_nsec greater than 999,999,999. I can’t see what officially happens if you do, but on my system thrd_sleep() returns -2 and fails.

Return Value

Returns 0 on timeout, or -1 if interrupted by a signal. Or any negative value on some other error. Weirdly, the spec allows this “other error negative value” to also be -1, so good luck with that.

Example

#include <stdio.h>
#include <threads.h>

int main(void)
{
    // Sleep for 3.25 seconds
    thrd_sleep(&(struct timespec){.tv_sec=3, .tv_nsec=250000000}, NULL);

    return 0;
}

46.21 `thrd_yield()`

Stop running that other threads might run

Synopsis

    #include <threads.h>
    
    void thrd_yield(void);

Description

If you have a thread that’s hogging the CPU and you want to give your other threads time to run, you can call thrd_yield(). If the system sees fit, it will put the calling thread to sleep and one of the other threads will run instead.

It’s a good way to be “polite” to the other threads in your program if you want the encourage them to run instead.

Return Value

Returns nothing!

Example

This example’s kinda poor because the OS is probably going to reschedule threads on the output anyway, but it gets the point across.

The main thread is giving other threads a chance to run after every block of dumb work it does.

#include <stdio.h>
#include <threads.h>

int run(void *arg)
{
    int main_thread = arg != NULL;

    if (main_thread) {
        long int total = 0;

        for (int i = 0; i < 10; i++) {
            for (long int j = 0; j < 1000L; j++)
                total++;

            printf("Main thread yielding\n");
            thrd_yield();                       // <-- YIELD HERE
        }
    } else
        printf("Other thread running!\n");

    return 0;
}

#define THREAD_COUNT 10

int main(void)
{
    thrd_t t[THREAD_COUNT];

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_create(t + i, run, i == 0? "main": NULL);

    for (int i = 0; i < THREAD_COUNT; i++)
        thrd_join(t[i], NULL);

    return 0;
}

The output will vary from run to run. Notice that even after thrd_yield() other threads might not yet be ready to run and the main thread will continue.

    Main thread yielding
    Main thread yielding
    Main thread yielding
    Other thread running!
    Other thread running!
    Other thread running!
    Other thread running!
    Main thread yielding
    Other thread running!
    Other thread running!
    Main thread yielding
    Main thread yielding
    Main thread yielding
    Other thread running!
    Main thread yielding
    Main thread yielding
    Main thread yielding
    Other thread running!
    Other thread running!

46.22 `tss_create()`

Create new thread-specific storage

Synopsis

    #include <threads.h>
    
    int tss_create(tss_t *key, tss_dtor_t dtor);

Description

This helps when you need per-thread storage of different values.

A common place this comes up is if you have a file scope variable that is shared between a bunch of functions and often returned. That’s not threadsafe. One way to refactor is to replace it with thread-specific storage so that each thread gets their own code and doesn’t step on other thread’s toes.

To make this work, you pass in a pointer to a tss_t key—this is the variable you will use in subsequent tss_set() and tss_get() calls to set and get the value associated with the key.

The interesting part of this is the dtor destructor pointer of type tss_dtor_t. This is actually a pointer to a function that takes a void* argument and returns void, i.e.

    void dtor(void *p) { ... }

This function will be called per thread when the thread exits with thrd_exit() (or returns from the run function).

It’s unspecified behavior to call this function while other threads’ destructors are running.

Return Value

Returns nothing!

Example

This is a general-purpose TSS example. Note the TSS variable is created near the top of main().

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

tss_t str;

void some_function(void)
{
    // Retrieve the per-thread value of this string
    char *tss_string = tss_get(str);

    // And print it
    printf("TSS string: %s\n", tss_string);
}

int run(void *arg)
{
    int serial = *(int*)arg;  // Get this thread's serial number
    free(arg);

    // malloc() space to hold the data for this thread
    char *s = malloc(64);
    sprintf(s, "thread %d! :)", serial);  // Happy little string

    // Set this TSS variable to point at the string
    tss_set(str, s);

    // Call a function that will get the variable
    some_function();

    return 0; // Equivalent to thrd_exit(0); fires destructors
}

#define THREAD_COUNT 15

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Make a new TSS variable, the free() function is the destructor
    tss_create(&str, free);                  // <-- CREATE TSS VAR!

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // And all threads are done, so let's free this
    tss_delete(str);
}

Output:

    TSS string: thread 0! :)
    TSS string: thread 2! :)
    TSS string: thread 1! :)
    TSS string: thread 5! :)
    TSS string: thread 3! :)
    TSS string: thread 6! :)
    TSS string: thread 4! :)
    TSS string: thread 7! :)
    TSS string: thread 8! :)
    TSS string: thread 9! :)
    TSS string: thread 10! :)
    TSS string: thread 13! :)
    TSS string: thread 12! :)
    TSS string: thread 11! :)
    TSS string: thread 14! :)

46.23 `tss_delete()`

Clean up a thread-specific storage variable

Synopsis

    #include <threads.h>
    
    void tss_delete(tss_t key);

Description

This is the opposite of tss_create(). You create (initialize) the TSS variable before using it, then, when all the threads are done that need it, you delete (deinitialize/free) it with this.

This doesn’t call any destructors! Those are all called by thrd_exit()!

Return Value

Returns nothing!

Example

This is a general-purpose TSS example. Note the TSS variable is deleted near the bottom of main().

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

tss_t str;

void some_function(void)
{
    // Retrieve the per-thread value of this string
    char *tss_string = tss_get(str);

    // And print it
    printf("TSS string: %s\n", tss_string);
}

int run(void *arg)
{
    int serial = *(int*)arg;  // Get this thread's serial number
    free(arg);

    // malloc() space to hold the data for this thread
    char *s = malloc(64);
    sprintf(s, "thread %d! :)", serial);  // Happy little string

    // Set this TSS variable to point at the string
    tss_set(str, s);

    // Call a function that will get the variable
    some_function();

    return 0; // Equivalent to thrd_exit(0); fires destructors
}

#define THREAD_COUNT 15

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Make a new TSS variable, the free() function is the destructor
    tss_create(&str, free);

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // And all threads are done, so let's free this
    tss_delete(str);    // <-- DELETE TSS VARIABLE!
}

Output:

    TSS string: thread 0! :)
    TSS string: thread 2! :)
    TSS string: thread 1! :)
    TSS string: thread 5! :)
    TSS string: thread 3! :)
    TSS string: thread 6! :)
    TSS string: thread 4! :)
    TSS string: thread 7! :)
    TSS string: thread 8! :)
    TSS string: thread 9! :)
    TSS string: thread 10! :)
    TSS string: thread 13! :)
    TSS string: thread 12! :)
    TSS string: thread 11! :)
    TSS string: thread 14! :)

46.24 `tss_get()`

Get thread-specific data

Synopsis

    #include <threads.h>
    
    void *tss_get(tss_t key);

Description

Once you’ve set a variable with tss_set(), you can retrieve the value with tss_get()—just pass in the key and you’ll get a pointer to the value back.

Don’t call this from a destructor.

Return Value

Returns the value stored for the given key, or NULL if there’s trouble.

Example

This is a general-purpose TSS example. Note the TSS variable is retrieved in some_function(), below.

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

tss_t str;

void some_function(void)
{
    // Retrieve the per-thread value of this string
    char *tss_string = tss_get(str);    // <-- GET THE VALUE

    // And print it
    printf("TSS string: %s\n", tss_string);
}

int run(void *arg)
{
    int serial = *(int*)arg;  // Get this thread's serial number
    free(arg);

    // malloc() space to hold the data for this thread
    char *s = malloc(64);
    sprintf(s, "thread %d! :)", serial);  // Happy little string

    // Set this TSS variable to point at the string
    tss_set(str, s);

    // Call a function that will get the variable
    some_function();

    return 0; // Equivalent to thrd_exit(0); fires destructors
}

#define THREAD_COUNT 15

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Make a new TSS variable, the free() function is the destructor
    tss_create(&str, free);

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // And all threads are done, so let's free this
    tss_delete(str);
}

Output:

    TSS string: thread 0! :)
    TSS string: thread 2! :)
    TSS string: thread 1! :)
    TSS string: thread 5! :)
    TSS string: thread 3! :)
    TSS string: thread 6! :)
    TSS string: thread 4! :)
    TSS string: thread 7! :)
    TSS string: thread 8! :)
    TSS string: thread 9! :)
    TSS string: thread 10! :)
    TSS string: thread 13! :)
    TSS string: thread 12! :)
    TSS string: thread 11! :)
    TSS string: thread 14! :)

46.25 `tss_set()`

Set thread-specific data

Synopsis

    #include <threads.h>
    
    int tss_set(tss_t key, void *val);

Description

Once you’ve set up your TSS variable with tss_create(), you can set it on a per thread basis with tss_set().

key is the identifier for this data, and val is a pointer to it.

The destructor specified in tss_create() will be called for the value set when the thread exits.

Also, if there’s a destructor and there is already at value for this key in place, the destructor will not be called for the already-existing value. In fact, this function will never cause a destructor to be called. So you’re on your own, there—best clean up the old value before overwriting it with the new one.

Return Value

Returns thrd_success when happy, and thrd_error when not.

Example

This is a general-purpose TSS example. Note the TSS variable is set in run(), below.

#include <stdio.h>
#include <stdlib.h>
#include <threads.h>

tss_t str;

void some_function(void)
{
    // Retrieve the per-thread value of this string
    char *tss_string = tss_get(str);

    // And print it
    printf("TSS string: %s\n", tss_string);
}

int run(void *arg)
{
    int serial = *(int*)arg;  // Get this thread's serial number
    free(arg);

    // malloc() space to hold the data for this thread
    char *s = malloc(64);
    sprintf(s, "thread %d! :)", serial);  // Happy little string

    // Set this TSS variable to point at the string
    tss_set(str, s);    // <-- SET THE TSS VARIABLE

    // Call a function that will get the variable
    some_function();

    return 0; // Equivalent to thrd_exit(0); fires destructors
}

#define THREAD_COUNT 15

int main(void)
{
    thrd_t t[THREAD_COUNT];

    // Make a new TSS variable, the free() function is the destructor
    tss_create(&str, free);

    for (int i = 0; i < THREAD_COUNT; i++) {
        int *n = malloc(sizeof *n);  // Holds a thread serial number
        *n = i;
        thrd_create(t + i, run, n);
    }

    for (int i = 0; i < THREAD_COUNT; i++) {
        thrd_join(t[i], NULL);
    }

    // And all threads are done, so let's free this
    tss_delete(str);
}

Output:

    TSS string: thread 0! :)
    TSS string: thread 2! :)
    TSS string: thread 1! :)
    TSS string: thread 5! :)
    TSS string: thread 3! :)
    TSS string: thread 6! :)
    TSS string: thread 4! :)
    TSS string: thread 7! :)
    TSS string: thread 8! :)
    TSS string: thread 9! :)
    TSS string: thread 10! :)
    TSS string: thread 13! :)
    TSS string: thread 12! :)
    TSS string: thread 11! :)
    TSS string: thread 14! :)

47 `<errno.h>` Error Information

Variable	Description
`errno`	Holds the error status of the last call

This header defines a single variable²¹⁸, errno, that can be checked to see if an error has occurred.

errno is set to 0 on startup, but no library function sets it to 0. If you’re going to use solely it to check for errors, set it to 0 before the call and then check it after. Not only that, but if there’s no error, all library functions will leave the value of errno unchanged.

Often, though, you’ll get some error indication from the function you’re calling then check errno to see what went wrong.

This is commonly used in conjunction with perror() to get a human-readable error message that corresponds to the specific error.

Important Safety Tip: You should never make your own variable called errno—that’s undefined behavior.

Note that the C Spec defines less than a handful of values errno can take on. Unix defines a bunch more ²¹⁹, as does Windows ²²⁰.

47.1 `errno`

Holds the error status of the last call

Synopsis

    errno   // Type is undefined, but it's assignable

Description

Indicates the error status of the last call (note that not all calls will set this value).

Value	Description
`0`	No error
`EDOM`	Domain error (from math)
`EILSEQ`	Encoding error (from character conversion)
`ERANGE`	Range error (from math)

If you’re doing a number of math functions, you might come across EDOM or ERANGE.

With multibyte/wide character conversion functions, you might see EILSEQ.

And your system might define any other number of values that errno could be set to, all of which will begin with the letter E.

Fun Fact: you can use EDOM, EILSEQ, and ERANGE with preprocessor directives such as #ifdef. But, frankly, I’m not sure why you’d do that other than to test their existence.

Example

The following prints an error message, since passing 2.0 to acos() is outside the function’s domain.

#include <stdio.h>
#include <math.h>
#include <errno.h>

int main(void)
{
    double x;

    errno = 0;       // Make sure this is clear before the call

    x = acos(2.0);   // Invalid argument to acos()

    if (errno == EDOM)
        perror("acos");
    else
        printf("Answer is %f\n", x);

    return 0;
}

Output:

    acos: Numerical argument out of domain

The following prints an error message (on my system), since passing 1e+30 to exp() produces a result that’s outside the range of a double.

#include <stdio.h>
#include <math.h>
#include <errno.h>

int main(void)
{
    double x;

    errno = 0;       // Make sure this is clear before the call

    x = exp(1e+30);  // Pass in some too-huge number

    if (errno == ERANGE)
        perror("exp");
    else
        printf("Answer is %f\n", x);

    return 0;
}

Output:

    exp: Numerical result out of range

This example tries to convert an invalid character into a wide character, failing. This sets errno to EILSEQ. We then use perror() to print an error message.

#include <stdio.h>
#include <string.h>
#include <wchar.h>
#include <errno.h>

int main(void)
{
    char *bad_str = "\xff";  // Probably invalid char in C locale
    wchar_t wc;
    size_t result;
    mbstate_t ps;

    memset(&ps, 0, sizeof ps);

    result = mbrtowc(&wc, bad_str, 1, &ps);

    if (result == (size_t)(-1))
        perror("mbrtowc");
    else
        printf("Converted to L'%lc'\n", wc);

    return 0;
}

Output:

    mbrtowc: Invalid or incomplete multibyte or wide character

Scientific Notation	E notation
\(1.2345\times10^{-3}=12.345\)	`1.2345e-3`
\(1.2345\times10^4=123450000\)	`1.2345e+4`

Function	Description
`cexp()`	Base-\(e\) exponential
`clog()`	Natural (base-\(e\)) logarithm

`numer`	`denom`	`quot`	`rem`
\(+\)	\(+\)	\(+\)	\(+\)
\(-\)	\(+\)	\(-\)	\(-\)
\(+\)	\(-\)	\(-\)	\(+\)
\(-\)	\(-\)	\(+\)	\(-\)

1 Foreword

1.1 Audience

1.2 Platform and Compiler

1.3 Official Homepage

1.4 Email Policy

1.5 Mirroring

1.6 Note for Translators

1.7 Copyright and Distribution

2 Hello, World!

2.1 What to Expect from C

2.2 Hello, World!

2.3 Compilation

2.4 Building with gcc

2.5 C Versions

3 Variables and Statements

3.1 Variables

3.1.1 Variable Names

3.1.2 Variable Types

3.1.3 Boolean Types

3.2 Operators and Expressions

3.2.1 The sizeof Operator

3.2.2 Arithmetic

3.2.3 Ternary Operator

3.2.4 Pre-and-Post Increment-and-Decrement

3.2.5 The Comma Operator

3.2.6 Conditional Operators

3.2.7 Boolean Operators

3.3 Flow Control

3.3.1 The if statement

3.3.2 The while statement

3.3.3 The do-while statement

3.3.4 The for statement

3.3.5 The switch Statement

4 Functions

4.1 Passing by Value

4.2 Function Prototypes

4.3 Empty Parameter Lists

5 Pointers—Cower In Fear!

5.1 Memory and Variables

5.2 Pointer Types

5.3 Dereferencing

5.4 Passing Pointers as Parameters

5.5 The NULL Pointer

5.6 A Note on Declaring Pointers

5.7 sizeof and Pointers

6 Arrays

6.1 Easy Example

6.2 Getting the Length of an Array

6.3 Array Initializers

6.4 Out of Bounds!

6.5 Multidimensional Arrays

6.6 Arrays and Pointers

6.6.1 Getting a Pointer to an Array

6.6.2 Passing Single Dimensional Arrays to Functions

6.6.3 Changing Arrays in Functions

6.6.4 Passing Multidimensional Arrays to Functions

7 Strings

7.1 Constant Strings

7.2 String Variables

7.3 String Variables as Arrays

7.4 String Initializers

7.5 Getting String Length

7.6 String Termination

7.7 Copying a String

8 Structs

8.1 Declaring a Struct

8.2 Struct Initializers

8.3 Passing Structs to Functions

8.4 The Arrow Operator

8.5 Copying and Returning structs

9 File Input/Output

9.1 The FILE* Data Type

9.2 Reading Text Files

9.3 End of File: EOF

9.3.1 Reading a Line at a Time

9.4 Formatted Input

9.5 Writing Text Files

9.6 Binary File I/O

9.6.1 struct and Number Caveats

10 typedef: Making New Types

2.4 Building with `gcc`

3.2.1 The `sizeof` Operator

3.3.1 The `if` statement

3.3.2 The `while` statement

3.3.3 The `do-while` statement

3.3.4 The `for` statement

3.3.5 The `switch` Statement

5.5 The `NULL` Pointer

5.7 `sizeof` and Pointers

8.5 Copying and Returning `struct`s

9.1 The `FILE*` Data Type

9.3 End of File: `EOF`

9.6.1 `struct` and Number Caveats

10 `typedef`: Making New Types

10.1 `typedef` in Theory

10.2 `typedef` in Practice

10.2.1 `typedef` and `struct`s

10.2.2 `typedef` and Other Types

10.2.3 `typedef` and Pointers

10.2.4 `typedef` and Capitalization

10.3 Arrays and `typedef`

11.3 `void` Pointers

12.1 Allocating and Deallocating, `malloc()` and `free()`

12.4 An Alternative: `calloc()`

12.5 Changing Allocated Size with `realloc()`

12.5.2 `realloc()` with `NULL`

13.3 `for`-loop Scope

14.3 More Integer Types: `short`, `long`, `long long`

14.4 More Float: `double` and `long double`

15.3.3 `void*`

16.1.1 `const`

16.1.1.1 `const` and Pointers

16.1.1.2 `const` Correctness

16.1.2 `restrict`

16.1.3 `volatile`

16.1.4 `_Atomic`