Introduction

Zig is an open-source programming language designed for robustness, optimality, and clarity.

Robust - behavior is correct even for edge cases such as out of memory.
Optimal - write programs the best way they can behave and perform.
Clear - precisely communicate your intent to the compiler and other programmers. The language imposes a low overhead to reading code.

Often the most efficient way to learn something new is to see examples, so this documentation shows how to use each of Zig's features. It is all on one page so you can search with your browser's search tool.

The code samples in this document are compiled and tested as part of the main test suite of Zig. This HTML document depends on no external files, so you can use it offline.

Hello World

hello.zig

const std = @import("std");

pub fn main() !void {
    // If this program is run without stdout attached, exit with an error.
    var stdout_file = try std.io.getStdOut();
    // If this program encounters pipe failure when printing to stdout, exit
    // with an error.
    try stdout_file.write("Hello, world!\n");
}

$ zig build-exe hello.zig
$ ./hello
Hello, world!

Usually you don't want to write to stdout. You want to write to stderr. And you don't care if it fails. It's more like a warning message that you want to emit. For that you can use a simpler API:

hello.zig

const warn = @import("std").debug.warn;

pub fn main() void {
    warn("Hello, world!\n");
}

$ zig build-exe hello.zig
$ ./hello
Hello, world!

Note that we also left off the ! from the return type. In Zig, if your main function cannot fail, you must use the void return type.

See also:

Values
@import
Errors
Root Source File

Comments

comments.zig

const assert = @import("std").debug.assert;

test "comments" {
    // Comments in Zig start with "//" and end at the next LF byte (end of line).
    // The below line is a comment, and won't be executed.

    //assert(false);

    const x = true;  // another comment
    assert(x);
}

$ zig test comments.zig
Test 1/1 comments...OK
All tests passed.

There are no multiline comments in Zig (e.g. like /* */ comments in C). This helps allow Zig to have the property that each line of code can be tokenized out of context.

Doc comments

A doc comment is one that begins with exactly three slashes (i.e. /// but not ////); multiple doc comments in a row are merged together to form a multiline doc comment. The doc comment documents whatever immediately follows it.

/// A structure for storing a timestamp, with nanosecond precision (this is a
/// multiline doc comment).
const Timestamp = struct {
    /// The number of seconds since the epoch (this is also a doc comment).
    seconds: i64,  // signed so we can represent pre-1970 (not a doc comment)
    /// The number of nanoseconds past the second (doc comment again).
    nanos: u32,

    /// Returns a `Timestamp` struct representing the Unix epoch; that is, the
    /// moment of 1970 Jan 1 00:00:00 UTC (this is a doc comment too).
    pub fn unixEpoch() Timestamp {
        return Timestamp{
            .seconds = 0,
            .nanos = 0,
        };
    }
};

Doc comments are only allowed in certain places; eventually, it will become a compile error have a doc comment in an unexpected place, such as in the middle of an expression, or just before a non-doc comment.

Values

values.zig

const std = @import("std");
const warn = std.debug.warn;
const os = std.os;
const assert = std.debug.assert;

pub fn main() void {
    // integers
    const one_plus_one: i32 = 1 + 1;
    warn("1 + 1 = {}\n", one_plus_one);

    // floats
    const seven_div_three: f32 = 7.0 / 3.0;
    warn("7.0 / 3.0 = {}\n", seven_div_three);

    // boolean
    warn("{}\n{}\n{}\n",
        true and false,
        true or false,
        !true);
    
    // optional
    var optional_value: ?[]const u8 = null;
    assert(optional_value == null);

    warn("\noptional 1\ntype: {}\nvalue: {}\n",
        @typeName(@typeOf(optional_value)), optional_value);

    optional_value = "hi";
    assert(optional_value != null);

    warn("\noptional 2\ntype: {}\nvalue: {}\n",
        @typeName(@typeOf(optional_value)), optional_value);

    // error union
    var number_or_error: error!i32 = error.ArgNotFound;

    warn("\nerror union 1\ntype: {}\nvalue: {}\n",
        @typeName(@typeOf(number_or_error)), number_or_error);
 
    number_or_error = 1234;

    warn("\nerror union 2\ntype: {}\nvalue: {}\n",
        @typeName(@typeOf(number_or_error)), number_or_error);
}

$ zig build-exe values.zig
$ ./values
1 + 1 = 2
7.0 / 3.0 = 2.33333325e+00
false
true
false

optional 1
type: ?[]const u8
value: null

optional 2
type: ?[]const u8
value: hi

error union 1
type: error!i32
value: error.ArgNotFound

error union 2
type: error!i32
value: 1234

Primitive Types

Name	C Equivalent	Description
`i8`	`int8_t`	signed 8-bit integer
`u8`	`uint8_t`	unsigned 8-bit integer
`i16`	`int16_t`	signed 16-bit integer
`u16`	`uint16_t`	unsigned 16-bit integer
`i32`	`int32_t`	signed 32-bit integer
`u32`	`uint32_t`	unsigned 32-bit integer
`i64`	`int64_t`	signed 64-bit integer
`u64`	`uint64_t`	unsigned 64-bit integer
`i128`	`__int128`	signed 128-bit integer
`u128`	`unsigned __int128`	unsigned 128-bit integer
`isize`	`intptr_t`	signed pointer sized integer
`usize`	`uintptr_t`	unsigned pointer sized integer
`c_short`	`short`	for ABI compatibility with C
`c_ushort`	`unsigned short`	for ABI compatibility with C
`c_int`	`int`	for ABI compatibility with C
`c_uint`	`unsigned int`	for ABI compatibility with C
`c_long`	`long`	for ABI compatibility with C
`c_ulong`	`unsigned long`	for ABI compatibility with C
`c_longlong`	`long long`	for ABI compatibility with C
`c_ulonglong`	`unsigned long long`	for ABI compatibility with C
`c_longdouble`	`long double`	for ABI compatibility with C
`c_void`	`void`	for ABI compatibility with C
`f16`	`_Float16`	16-bit floating point (10-bit mantissa) IEEE-754-2008 binary16
`f32`	`float`	32-bit floating point (23-bit mantissa) IEEE-754-2008 binary32
`f64`	`double`	64-bit floating point (52-bit mantissa) IEEE-754-2008 binary64
`f128`	`_Float128`	128-bit floating point (112-bit mantissa) IEEE-754-2008 binary128
`bool`	`bool`	`true` or `false`
`void`	(none)	0 bit type
`noreturn`	(none)	the type of `break`, `continue`, `return`, `unreachable`, and `while (true) {}`
`type`	(none)	the type of types
`error`	(none)	an error code
`comptime_int`	(none)	Only allowed for comptime-known values. The type of integer literals.
`comptime_float`	(none)	Only allowed for comptime-known values. The type of float literals.

In addition to the integer types above, arbitrary bit-width integers can be referenced by using an identifier of i or u followed by digits. For example, the identifier i7 refers to a signed 7-bit integer.

See also:

Integers
Floats
void
Errors

Primitive Values

Name	Description
`true` and `false`	`bool` values
`null`	used to set an optional type to `null`
`undefined`	used to leave a value unspecified

See also:

Optionals
undefined

String Literals

test.zig

const assert = @import("std").debug.assert;
const mem = @import("std").mem;

test "string literals" {
    // In Zig a string literal is an array of bytes.
    const normal_bytes = "hello";
    assert(@typeOf(normal_bytes) == [5]u8);
    assert(normal_bytes.len == 5);
    assert(normal_bytes[1] == 'e');
    assert('e' == '\x65');
    assert(mem.eql(u8, "hello", "h\x65llo"));

    // A C string literal is a null terminated pointer.
    const null_terminated_bytes = c"hello";
    assert(@typeOf(null_terminated_bytes) == [*]const u8);
    assert(null_terminated_bytes[5] == 0);
}

$ zig test test.zig
Test 1/1 string literals...OK
All tests passed.

See also:

Arrays
Zig Test

Escape Sequences

Escape Sequence	Name
`\n`	Newline
`\r`	Carriage Return
`\t`	Tab
`\\`	Backslash
`\'`	Single Quote
`\"`	Double Quote
`\xNN`	hexadecimal 8-bit character code (2 digits)
`\uNNNN`	hexadecimal 16-bit Unicode character code UTF-8 encoded (4 digits)
`\UNNNNNN`	hexadecimal 24-bit Unicode character code UTF-8 encoded (6 digits)

Note that the maximum valid Unicode point is 0x10ffff.

Multiline String Literals

Multiline string literals have no escapes and can span across multiple lines. To start a multiline string literal, use the \\ token. Just like a comment, the string literal goes until the end of the line. The end of the line is not included in the string literal. However, if the next line begins with \\ then a newline is appended and the string literal continues.

const hello_world_in_c =
    \\#include <stdio.h>
    \\
    \\int main(int argc, char **argv) {
    \\    printf("hello world\n");
    \\    return 0;
    \\}
;

For a multiline C string literal, prepend c to each \\:

const c_string_literal =
    c\\#include <stdio.h>
    c\\
    c\\int main(int argc, char **argv) {
    c\\    printf("hello world\n");
    c\\    return 0;
    c\\}
;

In this example the variable c_string_literal has type [*]const u8 and has a terminating null byte.

See also:

@embedFile

Assignment

Use the const keyword to assign a value to an identifier:

test.zig

const x = 1234;

fn foo() void {
    // It works at global scope as well as inside functions.
    const y = 5678;

    // Once assigned, an identifier cannot be changed.
    y += 1;
}

test "assignment" {
    foo();
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:8:7: error: cannot assign to constant
    y += 1;
      ^

const applies to all of the bytes that the identifier immediately addresses. Pointers have their own const-ness.

If you need a variable that you can modify, use the var keyword:

test.zig

const assert = @import("std").debug.assert;

test "var" {
    var y: i32 = 5678;

    y += 1;

    assert(y == 5679);
}

$ zig test test.zig
Test 1/1 var...OK
All tests passed.

Variables must be initialized:

test.zig

test "initialization" {
    var x: i32;

    x = 1;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:5: error: variables must be initialized
    var x: i32;
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:4:5: error: use of undeclared identifier 'x'
    x = 1;
    ^

undefined

Use undefined to leave variables uninitialized:

test.zig

const assert = @import("std").debug.assert;

test "init with undefined" {
    var x: i32 = undefined;
    x = 1;
    assert(x == 1);
}

$ zig test test.zig
Test 1/1 init with undefined...OK
All tests passed.

undefined can be implicitly cast to any type. Once this happens, it is no longer possible to detect that the value is undefined. undefined means the value could be anything, even something that is nonsense according to the type. Translated into English, undefined means "Not a meaningful value. Using this value would be a bug. The value will be unused, or overwritten before being used."

In Debug mode, Zig writes 0xaa bytes to undefined memory. This is to catch bugs early, and to help detect use of undefined memory in a debugger.

Integers

Integer Literals

const decimal_int = 98222;
const hex_int = 0xff;
const another_hex_int = 0xFF;
const octal_int = 0o755;
const binary_int = 0b11110000;

Runtime Integer Values

Integer literals have no size limitation, and if any undefined behavior occurs, the compiler catches it.

However, once an integer value is no longer known at compile-time, it must have a known size, and is vulnerable to undefined behavior.

fn divide(a: i32, b: i32) i32 {
    return a / b;
}

In this function, values a and b are known only at runtime, and thus this division operation is vulnerable to both integer overflow and division by zero.

Operators such as + and - cause undefined behavior on integer overflow. Also available are operations such as +% and -% which are defined to have wrapping arithmetic on all targets.

See also:

Integer Overflow
Division by Zero
Wrapping Operations

Floats

Zig has the following floating point types:

f16 - IEEE-754-2008 binary16
f32 - IEEE-754-2008 binary32
f64 - IEEE-754-2008 binary64
f128 - IEEE-754-2008 binary128
c_longdouble - matches long double for the target C ABI

Float Literals

Float literals have type comptime_float which is guaranteed to hold at least all possible values that the largest other floating point type can hold. Float literals implicitly cast to any other type.

const floating_point = 123.0E+77;
const another_float = 123.0;
const yet_another = 123.0e+77;

const hex_floating_point = 0x103.70p-5;
const another_hex_float = 0x103.70;
const yet_another_hex_float = 0x103.70P-5;

Floating Point Operations

By default floating point operations use Strict mode, but you can switch to Optimized mode on a per-block basis:

foo.zig

const builtin = @import("builtin");
const big = f64(1 << 40);

export fn foo_strict(x: f64) f64 {
    return x + big - big;
}

export fn foo_optimized(x: f64) f64 {
    @setFloatMode(builtin.FloatMode.Optimized);
    return x + big - big;
}

$ zig build-obj foo.zig --release-fast

For this test we have to separate code into two object files - otherwise the optimizer figures out all the values at compile-time, which operates in strict mode.

float_mode.zig

const warn = @import("std").debug.warn;

extern fn foo_strict(x: f64) f64;
extern fn foo_optimized(x: f64) f64;

pub fn main() void {
    const x = 0.001;
    warn("optimized = {}\n", foo_optimized(x));
    warn("strict = {}\n", foo_strict(x));
}

$ zig build-exe float_mode.zig --object foo.o
$ ./float_mode
optimized = 1.0e-03
strict = 9.765625e-04

See also:

@setFloatMode
Division by Zero

Operators

Table of Operators

Syntax	Relevant Types	Description	Example
`a + b a += b`	Integers Floats	Addition. Can cause overflow for integers. Invokes Peer Type Resolution for the operands. See also @addWithOverflow.	`2 + 5 == 7`
`a +% b a +%= b`	Integers	Wrapping Addition. Guaranteed to have twos-complement wrapping behavior. Invokes Peer Type Resolution for the operands. See also @addWithOverflow.	`u32(@maxValue(u32)) +% 1 == 0`
`a - b a -= b`	Integers Floats	Subtraction. Can cause overflow for integers. Invokes Peer Type Resolution for the operands. See also @subWithOverflow.	`2 - 5 == -3`
`a -% b a -%= b`	Integers	Wrapping Subtraction. Guaranteed to have twos-complement wrapping behavior. Invokes Peer Type Resolution for the operands. See also @subWithOverflow.	`u32(0) -% 1 == @maxValue(u32)`
`-a`	Integers Floats	Negation. Can cause overflow for integers.	`-1 == 0 - 1`
`-%a`	Integers	Wrapping Negation. Guaranteed to have twos-complement wrapping behavior.	`-%i32(@minValue(i32)) == @minValue(i32)`
`a * b a *= b`	Integers Floats	Multiplication. Can cause overflow for integers. Invokes Peer Type Resolution for the operands. See also @mulWithOverflow.	`2 * 5 == 10`
`a % b a %= b`	Integers	Wrapping Multiplication. Guaranteed to have twos-complement wrapping behavior. Invokes Peer Type Resolution for the operands. See also @mulWithOverflow.	`u8(200) *% 2 == 144`
`a / b a /= b`	Integers Floats	Divison. Can cause overflow for integers. Can cause Division by Zero for integers. Can cause Division by Zero for floats in FloatMode.Optimized Mode. For non-compile-time-known signed integers, must use @divTrunc, @divFloor, or @divExact instead of `/`. Invokes Peer Type Resolution for the operands.	`10 / 5 == 2`
`a % b a %= b`	Integers Floats	Remainder Division. Can cause Division by Zero for integers. Can cause Division by Zero for floats in FloatMode.Optimized Mode. For non-compile-time-known signed integers, must use @rem or @mod instead of `%`. Invokes Peer Type Resolution for the operands.	`10 % 3 == 1`
`a << b a <<= b`	Integers	Bit Shift Left. `b` must be comptime-known or have a type with log2 number of bits as `a`. See also @shlExact. See also @shlWithOverflow.	`1 << 8 == 256`
`a >> b a >>= b`	Integers	Bit Shift Right. `b` must be comptime-known or have a type with log2 number of bits as `a`. See also @shrExact.	`10 >> 1 == 5`
`a & b a &= b`	Integers	Bitwise AND. Invokes Peer Type Resolution for the operands.	`0b011 & 0b101 == 0b001`
`a \| b a \|= b`	Integers	Bitwise OR. Invokes Peer Type Resolution for the operands.	`0b010 \| 0b100 == 0b110`
`a ^ b a ^= b`	Integers	Bitwise XOR. Invokes Peer Type Resolution for the operands.	`0b011 ^ 0b101 == 0b110`
`~a`	Integers	Bitwise NOT.	`~u8(0b0101111) == 0b1010000`
`a orelse b`	Optionals	If `a` is `null`, returns `b` ("default value"), otherwise returns the unwrapped value of `a`. Note that `b` may be a value of type noreturn.	`const value: ?u32 = null; const unwrapped = value orelse 1234; unwrapped == 1234`
`a.?`	Optionals	Equivalent to: `a orelse unreachable`	`const value: ?u32 = 5678; value.? == 5678`
`a catch b a catch \|err\| b`	Error Unions	If `a` is an `error`, returns `b` ("default value"), otherwise returns the unwrapped value of `a`. Note that `b` may be a value of type noreturn. `err` is the `error` and is in scope of the expression `b`.	`const value: error!u32 = error.Broken; const unwrapped = value catch 1234; unwrapped == 1234`
`a and b`	bool	If `a` is `false`, returns `false` without evaluating `b`. Otherwise, returns `b`.	`false and true == false`
`a or b`	bool	If `a` is `true`, returns `true` without evaluating `b`. Otherwise, returns `b`.	`false or true == true`
`!a`	bool	Boolean NOT.	`!false == true`
`a == b`	Integers Floats bool type	Returns `true` if a and b are equal, otherwise returns `false`. Invokes Peer Type Resolution for the operands.	`(1 == 1) == true`
`a == null`	Optionals	Returns `true` if a is `null`, otherwise returns `false`.	`const value: ?u32 = null; value == null`
`a != b`	Integers Floats bool type	Returns `false` if a and b are equal, otherwise returns `true`. Invokes Peer Type Resolution for the operands.	`(1 != 1) == false`
`a > b`	Integers Floats	Returns `true` if a is greater than b, otherwise returns `false`. Invokes Peer Type Resolution for the operands.	`(2 > 1) == true`
`a >= b`	Integers Floats	Returns `true` if a is greater than or equal to b, otherwise returns `false`. Invokes Peer Type Resolution for the operands.	`(2 >= 1) == true`
`a < b`	Integers Floats	Returns `true` if a is less than b, otherwise returns `false`. Invokes Peer Type Resolution for the operands.	`(1 < 2) == true`>
`a <= b`	Integers Floats	Returns `true` if a is less than or equal to b, otherwise returns `false`. Invokes Peer Type Resolution for the operands.	`(1 <= 2) == true`
`a ++ b`	Arrays	Array concatenation. Only available when `a` and `b` are compile-time known.	`const mem = @import("std").mem; const array1 = []u32{1,2}; const array2 = []u32{3,4}; const together = array1 ++ array2; mem.eql(u32, together, []u32{1,2,3,4})`
`a ** b`	Arrays	Array multiplication. Only available when `a` and `b` are compile-time known.	`const mem = @import("std").mem; const pattern = "ab" ** 3; mem.eql(u8, pattern, "ababab")`
`a.*`	Pointers	Pointer dereference.	`const x: u32 = 1234; const ptr = &x; x.* == 1234`
`&a`	All types	Address of.	`const x: u32 = 1234; const ptr = &x; x.* == 1234`
`a \|\| b`	Error Set Type	Merging Error Sets	`const A = error{One}; const B = error{Two}; (A \|\| B) == error{One, Two}`

Precedence

x() x[] x.y
a!b
!x -x -%x ~x &x ?x
x{} x.* x.?
! * / % ** *% ||
+ - ++ +% -%
<< >>
&
^
|
== != < > <= >=
and
or
orelse catch
= *= /= %= += -= <<= >>= &= ^= |=

Arrays

arrays.zig

const assert = @import("std").debug.assert;
const mem = @import("std").mem;

// array literal
const message = []u8{ 'h', 'e', 'l', 'l', 'o' };

// get the size of an array
comptime {
    assert(message.len == 5);
}

// a string literal is an array literal
const same_message = "hello";

comptime {
    assert(mem.eql(u8, message, same_message));
    assert(@typeOf(message) == @typeOf(same_message));
}

test "iterate over an array" {
    var sum: usize = 0;
    for (message) |byte| {
        sum += byte;
    }
    assert(sum == usize('h') + usize('e') + usize('l') * 2 + usize('o'));
}

// modifiable array
var some_integers: [100]i32 = undefined;

test "modify an array" {
    for (some_integers) |*item, i| {
        item.* = @intCast(i32, i);
    }
    assert(some_integers[10] == 10);
    assert(some_integers[99] == 99);
}

// array concatenation works if the values are known
// at compile time
const part_one = []i32{ 1, 2, 3, 4 };
const part_two = []i32{ 5, 6, 7, 8 };
const all_of_it = part_one ++ part_two;
comptime {
    assert(mem.eql(i32, all_of_it, []i32{ 1, 2, 3, 4, 5, 6, 7, 8 }));
}

// remember that string literals are arrays
const hello = "hello";
const world = "world";
const hello_world = hello ++ " " ++ world;
comptime {
    assert(mem.eql(u8, hello_world, "hello world"));
}

// ** does repeating patterns
const pattern = "ab" ** 3;
comptime {
    assert(mem.eql(u8, pattern, "ababab"));
}

// initialize an array to zero
const all_zero = []u16{0} ** 10;

comptime {
    assert(all_zero.len == 10);
    assert(all_zero[5] == 0);
}

// use compile-time code to initialize an array
var fancy_array = init: {
    var initial_value: [10]Point = undefined;
    for (initial_value) |*pt, i| {
        pt.* = Point{
            .x = @intCast(i32, i),
            .y = @intCast(i32, i) * 2,
        };
    }
    break :init initial_value;
};
const Point = struct {
    x: i32,
    y: i32,
};

test "compile-time array initalization" {
    assert(fancy_array[4].x == 4);
    assert(fancy_array[4].y == 8);
}

// call a function to initialize an array
var more_points = []Point{makePoint(3)} ** 10;
fn makePoint(x: i32) Point {
    return Point{
        .x = x,
        .y = x * 2,
    };
}
test "array initialization with function calls" {
    assert(more_points[4].x == 3);
    assert(more_points[4].y == 6);
    assert(more_points.len == 10);
}

$ zig test arrays.zig
Test 1/4 iterate over an array...OK
Test 2/4 modify an array...OK
Test 3/4 compile-time array initalization...OK
Test 4/4 array initialization with function calls...OK
All tests passed.

See also:

for
Slices

Pointers

test.zig

const assert = @import("std").debug.assert;

test "address of syntax" {
    // Get the address of a variable:
    const x: i32 = 1234;
    const x_ptr = &x;

    // Deference a pointer:
    assert(x_ptr.* == 1234);

    // When you get the address of a const variable, you get a const pointer.
    assert(@typeOf(x_ptr) == *const i32);

    // If you want to mutate the value, you'd need an address of a mutable variable:
    var y: i32 = 5678;
    const y_ptr = &y;
    assert(@typeOf(y_ptr) == *i32);
    y_ptr.* += 1;
    assert(y_ptr.* == 5679);
}

test "pointer array access" {
    // Taking an address of an individual element gives a
    // pointer to a single item. This kind of pointer
    // does not support pointer arithmetic.
    var array = []u8{ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
    const ptr = &array[2];
    assert(@typeOf(ptr) == *u8);

    assert(array[2] == 3);
    ptr.* += 1;
    assert(array[2] == 4);
}

test "pointer slicing" {
    // In Zig, we prefer slices over pointers to null-terminated arrays.
    // You can turn an array into a slice using slice syntax:
    var array = []u8{ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 };
    const slice = array[2..4];
    assert(slice.len == 2);

    // Slices have bounds checking and are therefore protected
    // against this kind of undefined behavior. This is one reason
    // we prefer slices to pointers.
    assert(array[3] == 4);
    slice[1] += 1;
    assert(array[3] == 5);
}

comptime {
    // Pointers work at compile-time too, as long as you don't use
    // @ptrCast.
    var x: i32 = 1;
    const ptr = &x;
    ptr.* += 1;
    x += 1;
    assert(ptr.* == 3);
}

test "@ptrToInt and @intToPtr" {
    // To convert an integer address into a pointer, use @intToPtr:
    const ptr = @intToPtr(*i32, 0xdeadbeef);

    // To convert a pointer to an integer, use @ptrToInt:
    const addr = @ptrToInt(ptr);

    assert(@typeOf(addr) == usize);
    assert(addr == 0xdeadbeef);
}

comptime {
    // Zig is able to do this at compile-time, as long as
    // ptr is never dereferenced.
    const ptr = @intToPtr(*i32, 0xdeadbeef);
    const addr = @ptrToInt(ptr);
    assert(@typeOf(addr) == usize);
    assert(addr == 0xdeadbeef);
}

test "volatile" {
    // In Zig, loads and stores are assumed to not have side effects.
    // If a given load or store should have side effects, such as
    // Memory Mapped Input/Output (MMIO), use `volatile`:
    const mmio_ptr = @intToPtr(*volatile u8, 0x12345678);

    // Now loads and stores with mmio_ptr are guaranteed to all happen
    // and in the same order as in source code.
    assert(@typeOf(mmio_ptr) == *volatile u8);
}

test "optional pointers" {
    // Pointers cannot be null. If you want a null pointer, use the optional
    // prefix `?` to make the pointer type optional.
    var ptr: ?*i32 = null;

    var x: i32 = 1;
    ptr = &x;

    assert(ptr.?.* == 1);

    // Optional pointers are the same size as normal pointers, because pointer
    // value 0 is used as the null value.
    assert(@sizeOf(?*i32) == @sizeOf(*i32));
}

test "pointer casting" {
    // To convert one pointer type to another, use @ptrCast. This is an unsafe
    // operation that Zig cannot protect you against. Use @ptrCast only when other
    // conversions are not possible.
    const bytes align(@alignOf(u32)) = []u8{ 0x12, 0x12, 0x12, 0x12 };
    const u32_ptr = @ptrCast(*const u32, &bytes);
    assert(u32_ptr.* == 0x12121212);

    // Even this example is contrived - there are better ways to do the above than
    // pointer casting. For example, using a slice narrowing cast:
    const u32_value = @bytesToSlice(u32, bytes[0..])[0];
    assert(u32_value == 0x12121212);

    // And even another way, the most straightforward way to do it:
    assert(@bitCast(u32, bytes) == 0x12121212);
}

test "pointer child type" {
    // pointer types have a `child` field which tells you the type they point to.
    assert((*u32).Child == u32);
}

$ zig test test.zig
Test 1/8 address of syntax...OK
Test 2/8 pointer array access...OK
Test 3/8 pointer slicing...OK
Test 4/8 @ptrToInt and @intToPtr...OK
Test 5/8 volatile...OK
Test 6/8 optional pointers...OK
Test 7/8 pointer casting...OK
Test 8/8 pointer child type...OK
All tests passed.

Alignment

Each type has an alignment - a number of bytes such that, when a value of the type is loaded from or stored to memory, the memory address must be evenly divisible by this number. You can use @alignOf to find out this value for any type.

Alignment depends on the CPU architecture, but is always a power of two, and less than 1 << 29.

In Zig, a pointer type has an alignment value. If the value is equal to the alignment of the underlying type, it can be omitted from the type:

test.zig

const assert = @import("std").debug.assert;
const builtin = @import("builtin");

test "variable alignment" {
    var x: i32 = 1234;
    const align_of_i32 = @alignOf(@typeOf(x));
    assert(@typeOf(&x) == *i32);
    assert(*i32 == *align(align_of_i32) i32);
    if (builtin.arch == builtin.Arch.x86_64) {
        assert((*i32).alignment == 4);
    }
}

$ zig test test.zig
Test 1/1 variable alignment...OK
All tests passed.

In the same way that a *i32 can be implicitly cast to a *const i32, a pointer with a larger alignment can be implicitly cast to a pointer with a smaller alignment, but not vice versa.

You can specify alignment on variables and functions. If you do this, then pointers to them get the specified alignment:

test.zig

const assert = @import("std").debug.assert;

var foo: u8 align(4) = 100;

test "global variable alignment" {
    assert(@typeOf(&foo).alignment == 4);
    assert(@typeOf(&foo) == *align(4) u8);
    const slice = (*[1]u8)(&foo)[0..];
    assert(@typeOf(slice) == []align(4) u8);
}

fn derp() align(@sizeOf(usize) * 2) i32 { return 1234; }
fn noop1() align(1) void {}
fn noop4() align(4) void {}

test "function alignment" {
    assert(derp() == 1234);
    assert(@typeOf(noop1) == fn() align(1) void);
    assert(@typeOf(noop4) == fn() align(4) void);
    noop1();
    noop4();
}

$ zig test test.zig
Test 1/2 global variable alignment...OK
Test 2/2 function alignment...OK
All tests passed.

If you have a pointer or a slice that has a small alignment, but you know that it actually has a bigger alignment, use @alignCast to change the pointer into a more aligned pointer. This is a no-op at runtime, but inserts a safety check:

test.zig

const assert = @import("std").debug.assert;

test "pointer alignment safety" {
    var array align(4) = []u32{ 0x11111111, 0x11111111 };
    const bytes = @sliceToBytes(array[0..]);
    assert(foo(bytes) == 0x11111111);
}
fn foo(bytes: []u8) u32 {
    const slice4 = bytes[1..5];
    const int_slice = @bytesToSlice(u32, @alignCast(4, slice4));
    return int_slice[0];
}

$ zig test test.zig
Test 1/1 pointer alignment safety...incorrect alignment
/home/andy/dev/zig/docgen_tmp/test.zig:10:56: 0x2052bf in ??? (test)
    const int_slice = @bytesToSlice(u32, @alignCast(4, slice4));
                                                       ^
/home/andy/dev/zig/docgen_tmp/test.zig:6:15: 0x2050a7 in ??? (test)
    assert(foo(bytes) == 0x11111111);
              ^
/home/andy/dev/zig/build/lib/zig/std/special/test_runner.zig:13:25: 0x222aca in ??? (test)
        if (test_fn.func()) |_| {
                        ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:96:22: 0x22287b in ??? (test)
            root.main() catch |err| {
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x2227f5 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x222658 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222510 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Tests failed. Use the following command to reproduce the failure:
/home/andy/dev/zig/docgen_tmp/test

Type Based Alias Analysis

Zig uses Type Based Alias Analysis (also known as Strict Aliasing) to perform some optimizations. This means that pointers of different types must not alias the same memory, with the exception of u8. Pointers to u8 can alias any memory.

As an example, this code produces undefined behavior:

@ptrCast(*u32, f32(12.34)).*

Instead, use @bitCast:

@bitCast(u32, f32(12.34))

As an added benefit, the @bitCast version works at compile-time.

See also:

Slices
Memory

Slices

test.zig

const assert = @import("std").debug.assert;

test "basic slices" {
    var array = []i32{ 1, 2, 3, 4 };
    // A slice is a pointer and a length. The difference between an array and
    // a slice is that the array's length is part of the type and known at
    // compile-time, whereas the slice's length is known at runtime.
    // Both can be accessed with the `len` field.
    const slice = array[0..array.len];
    assert(&slice[0] == &array[0]);
    assert(slice.len == array.len);

    // Using the address-of operator on a slice gives a pointer to a single
    // item, while using the `ptr` field gives an unknown length pointer.
    assert(@typeOf(slice.ptr) == [*]i32);
    assert(@typeOf(&slice[0]) == *i32);
    assert(@ptrToInt(slice.ptr) == @ptrToInt(&slice[0]));

    // Slices have array bounds checking. If you try to access something out
    // of bounds, you'll get a safety check failure:
    slice[10] += 1;

    // Note that `slice.ptr` does not invoke safety checking, while `&slice[0]`
    // asserts that the slice has len >= 1.
}

$ zig test test.zig
Test 1/1 basic slices...index out of bounds
/home/andy/dev/zig/docgen_tmp/test.zig:21:10: 0x205156 in ??? (test)
    slice[10] += 1;
         ^
/home/andy/dev/zig/build/lib/zig/std/special/test_runner.zig:13:25: 0x222a8a in ??? (test)
        if (test_fn.func()) |_| {
                        ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:96:22: 0x22283b in ??? (test)
            root.main() catch |err| {
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x2227b5 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x222618 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x2224d0 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Tests failed. Use the following command to reproduce the failure:
/home/andy/dev/zig/docgen_tmp/test

This is one reason we prefer slices to pointers.

slices.zig

const assert = @import("std").debug.assert;
const mem = @import("std").mem;
const fmt = @import("std").fmt;

test "using slices for strings" {
    // Zig has no concept of strings. String literals are arrays of u8, and
    // in general the string type is []u8 (slice of u8).
    // Here we implicitly cast [5]u8 to []const u8
    const hello: []const u8 = "hello";
    const world: []const u8 = "世界";

    var all_together: [100]u8 = undefined;
    // You can use slice syntax on an array to convert an array into a slice.
    const all_together_slice = all_together[0..];
    // String concatenation example.
    const hello_world = try fmt.bufPrint(all_together_slice, "{} {}", hello, world);

    // Generally, you can use UTF-8 and not worry about whether something is a
    // string. If you don't need to deal with individual characters, no need
    // to decode.
    assert(mem.eql(u8, hello_world, "hello 世界"));
}

test "slice pointer" {
    var array: [10]u8 = undefined;
    const ptr = &array;

    // You can use slicing syntax to convert a pointer into a slice:
    const slice = ptr[0..5];
    slice[2] = 3;
    assert(slice[2] == 3);
    // The slice is mutable because we sliced a mutable pointer.
    assert(@typeOf(slice) == []u8);

    // You can also slice a slice:
    const slice2 = slice[2..3];
    assert(slice2.len == 1);
    assert(slice2[0] == 3);
}

test "slice widening" {
    // Zig supports slice widening and slice narrowing. Cast a slice of u8
    // to a slice of anything else, and Zig will perform the length conversion.
    const array align(@alignOf(u32)) = []u8{ 0x12, 0x12, 0x12, 0x12, 0x13, 0x13, 0x13, 0x13 };
    const slice = @bytesToSlice(u32, array[0..]);
    assert(slice.len == 2);
    assert(slice[0] == 0x12121212);
    assert(slice[1] == 0x13131313);
}

$ zig test slices.zig
Test 1/3 using slices for strings...OK
Test 2/3 slice pointer...OK
Test 3/3 slice widening...OK
All tests passed.

See also:

Pointers
for
Arrays

struct

structs.zig

// Declare a struct.
// Zig gives no guarantees about the order of fields and whether or
// not there will be padding.
const Point = struct {
    x: f32,
    y: f32,
};

// Maybe we want to pass it to OpenGL so we want to be particular about
// how the bytes are arranged.
const Point2 = packed struct {
    x: f32,
    y: f32,
};


// Declare an instance of a struct.
const p = Point {
    .x = 0.12,
    .y = 0.34,
};

// Maybe we're not ready to fill out some of the fields.
var p2 = Point {
    .x = 0.12,
    .y = undefined,
};

// Structs can have methods
// Struct methods are not special, they are only namespaced
// functions that you can call with dot syntax.
const Vec3 = struct {
    x: f32,
    y: f32,
    z: f32,

    pub fn init(x: f32, y: f32, z: f32) Vec3 {
        return Vec3 {
            .x = x,
            .y = y,
            .z = z,
        };
    }

    pub fn dot(self: *const Vec3, other: *const Vec3) f32 {
        return self.x * other.x + self.y * other.y + self.z * other.z;
    }
};

const assert = @import("std").debug.assert;
test "dot product" {
    const v1 = Vec3.init(1.0, 0.0, 0.0);
    const v2 = Vec3.init(0.0, 1.0, 0.0);
    assert(v1.dot(v2) == 0.0);

    // Other than being available to call with dot syntax, struct methods are
    // not special. You can reference them as any other declaration inside
    // the struct:
    assert(Vec3.dot(v1, v2) == 0.0);
}

// Structs can have global declarations.
// Structs can have 0 fields.
const Empty = struct {
    pub const PI = 3.14;
};
test "struct namespaced variable" {
    assert(Empty.PI == 3.14);
    assert(@sizeOf(Empty) == 0);

    // you can still instantiate an empty struct
    const does_nothing = Empty {};
}

// struct field order is determined by the compiler for optimal performance.
// however, you can still calculate a struct base pointer given a field pointer:
fn setYBasedOnX(x: *f32, y: f32) void {
    const point = @fieldParentPtr(Point, "x", x);
    point.y = y;
}
test "field parent pointer" {
    var point = Point {
        .x = 0.1234,
        .y = 0.5678,
    };
    setYBasedOnX(&point.x, 0.9);
    assert(point.y == 0.9);
}

// You can return a struct from a function. This is how we do generics
// in Zig:
fn LinkedList(comptime T: type) type {
    return struct {
        pub const Node = struct {
            prev: ?*Node,
            next: ?*Node,
            data: T,
        };

        first: ?*Node,
        last:  ?*Node,
        len:   usize,
    };
}

test "linked list" {
    // Functions called at compile-time are memoized. This means you can
    // do this:
    assert(LinkedList(i32) == LinkedList(i32));

    var list = LinkedList(i32) {
        .first = null,
        .last = null,
        .len = 0,
    };
    assert(list.len == 0);

    // Since types are first class values you can instantiate the type
    // by assigning it to a variable:
    const ListOfInts = LinkedList(i32);
    assert(ListOfInts == LinkedList(i32));

    var node = ListOfInts.Node {
        .prev = null,
        .next = null,
        .data = 1234,
    };
    var list2 = LinkedList(i32) {
        .first = &node,
        .last = &node,
        .len = 1,
    };
    assert(list2.first.?.data == 1234);
}

$ zig test structs.zig
Test 1/4 dot product...OK
Test 2/4 struct namespaced variable...OK
Test 3/4 field parent pointer...OK
Test 4/4 linked list...OK
All tests passed.

packed struct

packed structs have guaranteed in-memory layout.

TODO bit fields

TODO alignment

TODO endianness

TODO @bitOffsetOf and @byteOffsetOf

TODO mention how volatile loads and stores of bit packed fields could be more efficient when done by hand instead of with packed struct

struct Naming

Since all structs are anonymous, Zig infers the type name based on a few rules.

If the struct is in the initialization expression of a variable, it gets named after that variable.
If the struct is in the return expression, it gets named after the function it is returning from, with the parameter values serialized.
Otherwise, the struct gets a name such as (anonymous struct at file.zig:7:38).

struct_name.zig

const std = @import("std");

pub fn main() void {
    const Foo = struct {};
    std.debug.warn("variable: {}\n", @typeName(Foo));
    std.debug.warn("anonymous: {}\n", @typeName(struct {}));
    std.debug.warn("function: {}\n", @typeName(List(i32)));
}

fn List(comptime T: type) type {
    return struct {
        x: T,
    };
}

$ zig build-exe struct_name.zig
$ ./struct_name
variable: Foo
anonymous: (anonymous struct at /home/andy/dev/zig/docgen_tmp/struct_name.zig:6:49)
function: List(i32)

See also:

comptime
@fieldParentPtr

enum

enums.zig

const assert = @import("std").debug.assert;
const mem = @import("std").mem;

// Declare an enum.
const Type = enum {
    Ok,
    NotOk,
};

// Declare a specific instance of the enum variant.
const c = Type.Ok;

// If you want access to the ordinal value of an enum, you
// can specify the tag type.
const Value = enum(u2) {
    Zero,
    One,
    Two,
};

// Now you can cast between u2 and Value.
// The ordinal value starts from 0, counting up for each member.
test "enum ordinal value" {
    assert(@enumToInt(Value.Zero) == 0);
    assert(@enumToInt(Value.One) == 1);
    assert(@enumToInt(Value.Two) == 2);
}

// You can override the ordinal value for an enum.
const Value2 = enum(u32) {
    Hundred = 100,
    Thousand = 1000,
    Million = 1000000,
};
test "set enum ordinal value" {
    assert(@enumToInt(Value2.Hundred) == 100);
    assert(@enumToInt(Value2.Thousand) == 1000);
    assert(@enumToInt(Value2.Million) == 1000000);
}

// Enums can have methods, the same as structs and unions.
// Enum methods are not special, they are only namespaced
// functions that you can call with dot syntax.
const Suit = enum {
    Clubs,
    Spades,
    Diamonds,
    Hearts,

    pub fn isClubs(self: Suit) bool {
        return self == Suit.Clubs;
    }
};
test "enum method" {
    const p = Suit.Spades;
    assert(!p.isClubs());
}

// An enum variant of different types can be switched upon.
const Foo = enum {
    String,
    Number,
    None,
};
test "enum variant switch" {
    const p = Foo.Number;
    const what_is_it = switch (p) {
        Foo.String => "this is a string",
        Foo.Number => "this is a number",
        Foo.None => "this is a none",
    };
    assert(mem.eql(u8, what_is_it, "this is a number"));
}

// @TagType can be used to access the integer tag type of an enum.
const Small = enum {
    One,
    Two,
    Three,
    Four,
};
test "@TagType" {
    assert(@TagType(Small) == u2);
}

// @memberCount tells how many fields an enum has:
test "@memberCount" {
    assert(@memberCount(Small) == 4);
}

// @memberName tells the name of a field in an enum:
test "@memberName" {
    assert(mem.eql(u8, @memberName(Small, 1), "Two"));
}

// @tagName gives a []const u8 representation of an enum value:
test "@tagName" {
    assert(mem.eql(u8, @tagName(Small.Three), "Three"));
}

$ zig test enums.zig
Test 1/8 enum ordinal value...OK
Test 2/8 set enum ordinal value...OK
Test 3/8 enum method...OK
Test 4/8 enum variant switch...OK
Test 5/8 @TagType...OK
Test 6/8 @memberCount...OK
Test 7/8 @memberName...OK
Test 8/8 @tagName...OK
All tests passed.

extern enum

By default, enums are not guaranteed to be compatible with the C ABI:

test.zig

const Foo = enum { A, B, C };
export fn entry(foo: Foo) void { }

$ zig build-obj test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:22: error: parameter of type 'Foo' not allowed in function with calling convention 'ccc'
export fn entry(foo: Foo) void { 
                     ^

For a C-ABI-compatible enum, use extern enum:

test.zig

const Foo = extern enum { A, B, C };
export fn entry(foo: Foo) void { }

$ zig build-obj test.zig

packed enum

By default, the size of enums is not guaranteed.

packed enum causes the size of the enum to be the same as the size of the integer tag type of the enum:

test.zig

const std = @import("std");

test "packed enum" {
    const Number = packed enum(u8) {
        One,
        Two,
        Three,
    };
    std.debug.assert(@sizeOf(Number) == @sizeOf(u8));
}

$ zig test test.zig
Test 1/1 packed enum...OK
All tests passed.

See also:

@memberName
@memberCount
@tagName
@sizeOf

union

union.zig

const assert = @import("std").debug.assert;
const mem = @import("std").mem;

// A union has only 1 active field at a time.
const Payload = union {
    Int: i64,
    Float: f64,
    Bool: bool,
};
test "simple union" {
    var payload = Payload {.Int = 1234};
    // payload.Float = 12.34; // ERROR! field not active
    assert(payload.Int == 1234);
    // You can activate another field by assigning the entire union.
    payload = Payload {.Float = 12.34};
    assert(payload.Float == 12.34);
}

// Unions can be given an enum tag type:
const ComplexTypeTag = enum { Ok, NotOk }; 
const ComplexType = union(ComplexTypeTag) {
    Ok: u8,
    NotOk: void,
};

// Declare a specific instance of the union variant.
test "declare union value" {
    const c = ComplexType { .Ok = 0 };
    assert(ComplexTypeTag(c) == ComplexTypeTag.Ok);
}

// @TagType can be used to access the enum tag type of a union.
test "@TagType" {
    assert(@TagType(ComplexType) == ComplexTypeTag);
}

// Unions can be made to infer the enum tag type.
const Foo = union(enum) {
    String: []const u8,
    Number: u64,

    // void can be omitted when inferring enum tag type.
    None,
};
test "union variant switch" {
    const p = Foo { .Number = 54 };
    const what_is_it = switch (p) {
        // Capture by reference
        Foo.String => |*x| blk: {
            break :blk "this is a string";
        },

        // Capture by value
        Foo.Number => |x| blk: {
            assert(x == 54);
            break :blk "this is a number";
        },

        Foo.None => blk: {
            break :blk "this is a none";
        },
    };
    assert(mem.eql(u8, what_is_it, "this is a number"));
}

// Unions can have methods just like structs and enums:

const Variant = union(enum) {
    Int: i32,
    Bool: bool,

    fn truthy(self: *const Variant) bool {
        return switch (self.*) {
            Variant.Int => |x_int| x_int != 0,
            Variant.Bool => |x_bool| x_bool,
        };
    }
};

test "union method" {
    var v1 = Variant { .Int = 1 };
    var v2 = Variant { .Bool = false };

    assert(v1.truthy());
    assert(!v2.truthy());
}


const Small = union {
    A: i32,
    B: bool,
    C: u8,
};

// @memberCount tells how many fields a union has:
test "@memberCount" {
    assert(@memberCount(Small) == 3);
}

// @memberName tells the name of a field in an enum:
test "@memberName" {
    assert(mem.eql(u8, @memberName(Small, 1), "B"));
}

// @tagName gives a []const u8 representation of an enum value,
// but only if the union has an enum tag type.
const Small2 = union(enum) {
    A: i32,
    B: bool,
    C: u8,
};
test "@tagName" {
    assert(mem.eql(u8, @tagName(Small2.C), "C"));
}

$ zig test union.zig
Test 1/8 simple union...OK
Test 2/8 declare union value...OK
Test 3/8 @TagType...OK
Test 4/8 union variant switch...OK
Test 5/8 union method...OK
Test 6/8 @memberCount...OK
Test 7/8 @memberName...OK
Test 8/8 @tagName...OK
All tests passed.

Unions with an enum tag are generated as a struct with a tag field and union field. Zig sorts the order of the tag and union field by the largest alignment.

blocks

Blocks are used to limit the scope of variable declarations:

test.zig

test "access variable after block scope" {
    {
        var x: i32 = 1;
    }
    x += 1;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:5:5: error: use of undeclared identifier 'x'
    x += 1;
    ^

Blocks are expressions. When labeled, break can be used to return a value from the block:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "labeled break from labeled block expression" {
    var y: i32 = 123;

    const x = blk: {
        y += 1;
        break :blk y;
    };
    assert(x == 124);
    assert(y == 124);
}

$ zig test test.zig
Test 1/1 labeled break from labeled block expression...OK
All tests passed.

Here, blk can be any name.

See also:

Labeled while
Labeled for

switch

switch.zig

const assert = @import("std").debug.assert;
const builtin = @import("builtin");

test "switch simple" {
    const a: u64 = 10;
    const zz: u64 = 103;

    // All branches of a switch expression must be able to be coerced to a
    // common type.
    //
    // Branches cannot fallthrough. If fallthrough behavior is desired, combine
    // the cases and use an if.
    const b = switch (a) {
        // Multiple cases can be combined via a ','
        1, 2, 3 => 0,

        // Ranges can be specified using the ... syntax. These are inclusive
        // both ends.
        5 ... 100 => 1,

        // Branches can be arbitrarily complex.
        101 => blk: {
            const c: u64 = 5;
            break :blk c * 2 + 1;
        },

        // Switching on arbitrary expressions is allowed as long as the
        // expression is known at compile-time.
        zz => zz,
        comptime blk: {
            const d: u32 = 5;
            const e: u32 = 100;
            break :blk d + e;
        } => 107,

        // The else branch catches everything not already captured.
        // Else branches are mandatory unless the entire range of values
        // is handled.
        else => 9,
    };

    assert(b == 1);
}

test "switch enum" {
    const Item = union(enum) {
        A: u32,
        C: struct { x: u8, y: u8 },
        D,
    };

    var a = Item { .A = 3 };

    // Switching on more complex enums is allowed.
    const b = switch (a) {
        // A capture group is allowed on a match, and will return the enum
        // value matched.
        Item.A => |item| item,

        // A reference to the matched value can be obtained using `*` syntax.
        Item.C => |*item| blk: {
            item.*.x += 1;
            break :blk 6;
        },

        // No else is required if the types cases was exhaustively handled
        Item.D => 8,
    };

    assert(b == 3);
}

// Switch expressions can be used outside a function:
const os_msg = switch (builtin.os) {
    builtin.Os.linux => "we found a linux user",
    else => "not a linux user",
};

// Inside a function, switch statements implicitly are compile-time
// evaluated if the target expression is compile-time known.
test "switch inside function" {
    switch (builtin.os) {
        builtin.Os.fuchsia => {
            // On an OS other than fuchsia, block is not even analyzed,
            // so this compile error is not triggered.
            // On fuchsia this compile error would be triggered.
            @compileError("fuchsia not supported");
        },
        else => {},
    }
}

$ zig test switch.zig
Test 1/3 switch simple...OK
Test 2/3 switch enum...OK
Test 3/3 switch inside function...OK
All tests passed.

See also:

comptime
enum
@compileError
Compile Variables

while

A while loop is used to repeatedly execute an expression until some condition is no longer true.

while.zig

const assert = @import("std").debug.assert;

test "while basic" {
    var i: usize = 0;
    while (i < 10) {
        i += 1;
    }
    assert(i == 10);
}

$ zig test while.zig
Test 1/1 while basic...OK
All tests passed.

Use break to exit a while loop early.

while.zig

const assert = @import("std").debug.assert;

test "while break" {
    var i: usize = 0;
    while (true) {
        if (i == 10)
            break;
        i += 1;
    }
    assert(i == 10);
}

$ zig test while.zig
Test 1/1 while break...OK
All tests passed.

Use continue to jump back to the beginning of the loop.

while.zig

const assert = @import("std").debug.assert;

test "while continue" {
    var i: usize = 0;
    while (true) {
        i += 1;
        if (i < 10)
            continue;
        break;
    }
    assert(i == 10);
}

$ zig test while.zig
Test 1/1 while continue...OK
All tests passed.

While loops support a continue expression which is executed when the loop is continued. The continue keyword respects this expression.

while.zig

const assert = @import("std").debug.assert;

test "while loop continue expression" {
    var i: usize = 0;
    while (i < 10) : (i += 1) {}
    assert(i == 10);
}

test "while loop continue expression, more complicated" {
    var i: usize = 1;
    var j: usize = 1;
    while (i * j < 2000) : ({ i *= 2; j *= 3; }) {
        const my_ij = i * j;
        assert(my_ij < 2000);
    }
}

$ zig test while.zig
Test 1/2 while loop continue expression...OK
Test 2/2 while loop continue expression, more complicated...OK
All tests passed.

While loops are expressions. The result of the expression is the result of the else clause of a while loop, which is executed when the condition of the while loop is tested as false.

break, like return, accepts a value parameter. This is the result of the while expression. When you break from a while loop, the else branch is not evaluated.

while.zig

const assert = @import("std").debug.assert;

test "while else" {
    assert(rangeHasNumber(0, 10, 5));
    assert(!rangeHasNumber(0, 10, 15));
}

fn rangeHasNumber(begin: usize, end: usize, number: usize) bool {
    var i = begin;
    return while (i < end) : (i += 1) {
        if (i == number) {
            break true;
        }
    } else false;
}

$ zig test while.zig
Test 1/1 while else...OK
All tests passed.

Labeled while

When a while loop is labeled, it can be referenced from a break or continue from within a nested loop:

test.zig

test "nested break" {
    outer: while (true) {
        while (true) {
            break :outer;
        }
    }
}

test "nested continue" {
    var i: usize = 0;
    outer: while (i < 10) : (i += 1) {
        while (true) {
            continue :outer;
        }
    }
}

$ zig test test.zig
Test 1/2 nested break...OK
Test 2/2 nested continue...OK
All tests passed.

while with Optionals

Just like if expressions, while loops can take an optional as the condition and capture the payload. When null is encountered the loop exits.

When the |x| syntax is present on a while expression, the while condition must have an Optional Type.

The else branch is allowed on optional iteration. In this case, it will be executed on the first null value encountered.

while.zig

const assert = @import("std").debug.assert;

test "while null capture" {
    var sum1: u32 = 0;
    numbers_left = 3;
    while (eventuallyNullSequence()) |value| {
        sum1 += value;
    }
    assert(sum1 == 3);

    var sum2: u32 = 0;
    numbers_left = 3;
    while (eventuallyNullSequence()) |value| {
        sum2 += value;
    } else {
        assert(sum1 == 3);
    }
}

var numbers_left: u32 = undefined;
fn eventuallyNullSequence() ?u32 {
    return if (numbers_left == 0) null else blk: {
        numbers_left -= 1;
        break :blk numbers_left;
    };
}

$ zig test while.zig
Test 1/1 while null capture...OK
All tests passed.

while with Error Unions

Just like if expressions, while loops can take an error union as the condition and capture the payload or the error code. When the condition results in an error code the else branch is evaluated and the loop is finished.

When the else |x| syntax is present on a while expression, the while condition must have an Error Union Type.

while.zig

const assert = @import("std").debug.assert;

test "while error union capture" {
    var sum1: u32 = 0;
    numbers_left = 3;
    while (eventuallyErrorSequence()) |value| {
        sum1 += value;
    } else |err| {
        assert(err == error.ReachedZero);
    }
}

var numbers_left: u32 = undefined;

fn eventuallyErrorSequence() error!u32 {
    return if (numbers_left == 0) error.ReachedZero else blk: {
        numbers_left -= 1;
        break :blk numbers_left;
    };
}

$ zig test while.zig
Test 1/1 while error union capture...OK
All tests passed.

inline while

While loops can be inlined. This causes the loop to be unrolled, which allows the code to do some things which only work at compile time, such as use types as first class values.

test.zig

const assert = @import("std").debug.assert;

test "inline while loop" {
    comptime var i = 0;
    var sum: usize = 0;
    inline while (i < 3) : (i += 1) {
        const T = switch (i) {
            0 => f32,
            1 => i8,
            2 => bool,
            else => unreachable,
        };
        sum += typeNameLength(T);
    }
    assert(sum == 9);
}

fn typeNameLength(comptime T: type) usize {
    return @typeName(T).len;
}

$ zig test test.zig
Test 1/1 inline while loop...OK
All tests passed.

It is recommended to use inline loops only for one of these reasons:

You need the loop to execute at comptime for the semantics to work.
You have a benchmark to prove that forcibly unrolling the loop in this way is measurably faster.

See also:

if
Optionals
Errors
comptime
unreachable

for

for.zig

const assert = @import("std").debug.assert;

test "for basics" {
    const items = []i32 { 4, 5, 3, 4, 0 };
    var sum: i32 = 0;

    // For loops iterate over slices and arrays.
    for (items) |value| {
        // Break and continue are supported.
        if (value == 0) {
            continue;
        }
        sum += value;
    }
    assert(sum == 16);

    // To iterate over a portion of a slice, reslice.
    for (items[0..1]) |value| {
        sum += value;
    }
    assert(sum == 20);

    // To access the index of iteration, specify a second capture value.
    // This is zero-indexed.
    var sum2: i32 = 0;
    for (items) |value, i| {
        assert(@typeOf(i) == usize);
        sum2 += @intCast(i32, i);
    }
    assert(sum2 == 10);
}

test "for reference" {
    var items = []i32 { 3, 4, 2 };

    // Iterate over the slice by reference by
    // specifying that the capture value is a pointer.
    for (items) |*value| {
        value.* += 1;
    }

    assert(items[0] == 4);
    assert(items[1] == 5);
    assert(items[2] == 3);
}

test "for else" {
    // For allows an else attached to it, the same as a while loop.
    var items = []?i32 { 3, 4, null, 5 };

    // For loops can also be used as expressions.
    var sum: i32 = 0;
    const result = for (items) |value| {
        if (value == null) {
            break 9;
        } else {
            sum += value.?;
        }
    } else blk: {
        assert(sum == 7);
        break :blk sum;
    };
}

$ zig test for.zig
Test 1/3 for basics...OK
Test 2/3 for reference...OK
Test 3/3 for else...OK
All tests passed.

Labeled for

When a for loop is labeled, it can be referenced from a break or continue from within a nested loop:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "nested break" {
    var count: usize = 0;
    outer: for ([]i32{ 1, 2, 3, 4, 5 }) |_| {
        for ([]i32{ 1, 2, 3, 4, 5 }) |_| {
            count += 1;
            break :outer;
        }
    }
    assert(count == 1);
}

test "nested continue" {
    var count: usize = 0;
    outer: for ([]i32{ 1, 2, 3, 4, 5, 6, 7, 8 }) |_| {
        for ([]i32{ 1, 2, 3, 4, 5 }) |_| {
            count += 1;
            continue :outer;
        }
    }

    assert(count == 8);
}

$ zig test test.zig
Test 1/2 nested break...OK
Test 2/2 nested continue...OK
All tests passed.

inline for

For loops can be inlined. This causes the loop to be unrolled, which allows the code to do some things which only work at compile time, such as use types as first class values. The capture value and iterator value of inlined for loops are compile-time known.

test.zig

const assert = @import("std").debug.assert;

test "inline for loop" {
    const nums = []i32{2, 4, 6};
    var sum: usize = 0;
    inline for (nums) |i| {
        const T = switch (i) {
            2 => f32,
            4 => i8,
            6 => bool,
            else => unreachable,
        };
        sum += typeNameLength(T);
    }
    assert(sum == 9);
}

fn typeNameLength(comptime T: type) usize {
    return @typeName(T).len;
}

$ zig test test.zig
Test 1/1 inline for loop...OK
All tests passed.

It is recommended to use inline loops only for one of these reasons:

You need the loop to execute at comptime for the semantics to work.
You have a benchmark to prove that forcibly unrolling the loop in this way is measurably faster.

See also:

while
comptime
Arrays
Slices

if

if.zig

// If expressions have three uses, corresponding to the three types:
// * bool
// * ?T
// * error!T

const assert = @import("std").debug.assert;

test "if boolean" {
    // If expressions test boolean conditions.
    const a: u32 = 5;
    const b: u32 = 4;
    if (a != b) {
        assert(true);
    } else if (a == 9) {
        unreachable;
    } else {
        unreachable;
    }

    // If expressions are used instead of a ternary expression.
    const result = if (a != b) 47 else 3089;
    assert(result == 47);
}

test "if optional" {
    // If expressions test for null.

    const a: ?u32 = 0;
    if (a) |value| {
        assert(value == 0);
    } else {
        unreachable;
    }

    const b: ?u32 = null;
    if (b) |value| {
        unreachable;
    } else {
        assert(true);
    }

    // The else is not required.
    if (a) |value| {
        assert(value == 0);
    }

    // To test against null only, use the binary equality operator.
    if (b == null) {
        assert(true);
    }

    // Access the value by reference using a pointer capture.
    var c: ?u32 = 3;
    if (c) |*value| {
        value.* = 2;
    }

    if (c) |value| {
        assert(value == 2);
    } else {
        unreachable;
    }
}

test "if error union" {
    // If expressions test for errors.
    // Note the |err| capture on the else.

    const a: error!u32 = 0;
    if (a) |value| {
        assert(value == 0);
    } else |err| {
        unreachable;
    }

    const b: error!u32 = error.BadValue;
    if (b) |value| {
        unreachable;
    } else |err| {
        assert(err == error.BadValue);
    }

    // The else and |err| capture is strictly required.
    if (a) |value| {
        assert(value == 0);
    } else |_| {}

    // To check only the error value, use an empty block expression.
    if (b) |_| {} else |err| {
        assert(err == error.BadValue);
    }

    // Access the value by reference using a pointer capture.
    var c: error!u32 = 3;
    if (c) |*value| {
        value.* = 9;
    } else |err| {
        unreachable;
    }

    if (c) |value| {
        assert(value == 9);
    } else |err| {
        unreachable;
    }
}

$ zig test if.zig
Test 1/3 if boolean...OK
Test 2/3 if optional...OK
Test 3/3 if error union...OK
All tests passed.

See also:

Optionals
Errors

defer

defer.zig

const std = @import("std");
const assert = std.debug.assert;
const warn = std.debug.warn;

// defer will execute an expression at the end of the current scope.
fn deferExample() usize {
    var a: usize = 1;

    {
        defer a = 2;
        a = 1;
    }
    assert(a == 2);

    a = 5;
    return a;
}

test "defer basics" {
    assert(deferExample() == 5);
}

// If multiple defer statements are specified, they will be executed in
// the reverse order they were run.
fn deferUnwindExample() void {
    warn("\n");

    defer {
        warn("1 ");
    }
    defer {
        warn("2 ");
    }
    if (false) {
        // defers are not run if they are never executed.
        defer {
            warn("3 ");
        }
    }
}

test "defer unwinding" {
    deferUnwindExample();
}

// The errdefer keyword is similar to defer, but will only execute if the
// scope returns with an error.
//
// This is especially useful in allowing a function to clean up properly
// on error, and replaces goto error handling tactics as seen in c.
fn deferErrorExample(is_error: bool) !void {
    warn("\nstart of function\n");

    // This will always be executed on exit
    defer {
        warn("end of function\n");
    }

    errdefer {
        warn("encountered an error!\n");
    }

    if (is_error) {
        return error.DeferError;
    }
}

test "errdefer unwinding" {
    _ = deferErrorExample(false);
    _ = deferErrorExample(true);
}

$ zig test defer.zig
Test 1/3 defer basics...OK
Test 2/3 defer unwinding...
2 1 OK
Test 3/3 errdefer unwinding...
start of function
end of function

start of function
encountered an error!
end of function
OK
All tests passed.

See also:

Errors

unreachable

In Debug and ReleaseSafe mode, and when using zig test, unreachable emits a call to panic with the message reached unreachable code.

In ReleaseFast mode, the optimizer uses the assumption that unreachable code will never be hit to perform optimizations. However, zig test even in ReleaseFast mode still emits unreachable as calls to panic.

Basics

test.zig

// unreachable is used to assert that control flow will never happen upon a
// particular location:
test "basic math" {
    const x = 1;
    const y = 2;
    if (x + y != 3) {
        unreachable;
    }
}

$ zig test test.zig
Test 1/1 basic math...OK
All tests passed.

In fact, this is how assert is implemented:

test.zig

fn assert(ok: bool) void {
    if (!ok) unreachable; // assertion failure
}

// This test will fail because we hit unreachable.
test "this will fail" {
    assert(false);
}

$ zig test test.zig
Test 1/1 this will fail...reached unreachable code
/home/andy/dev/zig/docgen_tmp/test.zig:2:14: 0x2051c9 in ??? (test)
    if (!ok) unreachable; // assertion failure
             ^
/home/andy/dev/zig/docgen_tmp/test.zig:7:11: 0x20504b in ??? (test)
    assert(false);
          ^
/home/andy/dev/zig/build/lib/zig/std/special/test_runner.zig:13:25: 0x22297a in ??? (test)
        if (test_fn.func()) |_| {
                        ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:96:22: 0x22272b in ??? (test)
            root.main() catch |err| {
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x2226a5 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x222508 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x2223c0 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Tests failed. Use the following command to reproduce the failure:
/home/andy/dev/zig/docgen_tmp/test

At Compile-Time

test.zig

const assert = @import("std").debug.assert;

test "type of unreachable" {
    comptime {
        // The type of unreachable is noreturn.

        // However this assertion will still fail because
        // evaluating unreachable at compile-time is a compile error.

        assert(@typeOf(unreachable) == noreturn);
    }
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:10:16: error: unreachable code
        assert(@typeOf(unreachable) == noreturn);
               ^

See also:

Zig Test
Build Mode
comptime

noreturn

noreturn is the type of:

break
continue
return
unreachable
while (true) {}

When resolving types together, such as if clauses or switch prongs, the noreturn type is compatible with every other type. Consider:

test.zig

fn foo(condition: bool, b: u32) void {
    const a = if (condition) b else return;
    @panic("do something with a");
}
test "noreturn" {
    foo(false, 1);
}

$ zig test test.zig
Test 1/1 noreturn...OK
All tests passed.

Another use case for noreturn is the exit function:

test.zig

pub extern "kernel32" stdcallcc fn ExitProcess(exit_code: c_uint) noreturn;

test "foo" {
    const value = bar() catch ExitProcess(1);
    assert(value == 1234);
}

fn bar() error!u32 {
    return 1234;
}

const assert = @import("std").debug.assert;

$ zig test test.zig
Created /home/andy/dev/zig/docgen_tmp/test but skipping execution because it is non-native.

Functions

functions.zig

const assert = @import("std").debug.assert;

// Functions are declared like this
fn add(a: i8, b: i8) i8 {
    if (a == 0) {
        // You can still return manually if needed.
        return b;
    }

    return a + b;
}

// The export specifier makes a function externally visible in the generated
// object file, and makes it use the C ABI.
export fn sub(a: i8, b: i8) i8 { return a - b; }

// The extern specifier is used to declare a function that will be resolved
// at link time, when linking statically, or at runtime, when linking
// dynamically.
// The stdcallcc specifier changes the calling convention of the function.
extern "kernel32" stdcallcc fn ExitProcess(exit_code: u32) noreturn;
extern "c" fn atan2(a: f64, b: f64) f64;

// The @setCold builtin tells the optimizer that a function is rarely called.
fn abort() noreturn {
    @setCold(true);
    while (true) {}
}

// nakedcc makes a function not have any function prologue or epilogue.
// This can be useful when integrating with assembly.
nakedcc fn _start() noreturn {
    abort();
}

// The pub specifier allows the function to be visible when importing.
// Another file can use @import and call sub2
pub fn sub2(a: i8, b: i8) i8 { return a - b; }

// Functions can be used as values and are equivalent to pointers.
const call2_op = fn (a: i8, b: i8) i8;
fn do_op(fn_call: call2_op, op1: i8, op2: i8) i8 {
    return fn_call(op1, op2);
}

test "function" {
    assert(do_op(add, 5, 6) == 11);
    assert(do_op(sub2, 5, 6) == -1);
}

$ zig test functions.zig
Test 1/1 function...OK
All tests passed.

Function values are like pointers:

test.zig

const assert = @import("std").debug.assert;

comptime {
    assert(@typeOf(foo) == fn()void);
    assert(@sizeOf(fn()void) == @sizeOf(?fn()void));
}

fn foo() void { }

$ zig build-obj test.zig

Pass-by-value Parameters

In Zig, structs, unions, and enums with payloads can be passed directly to a function:

test.zig

const Point = struct {
    x: i32,
    y: i32,
};

fn foo(point: Point) i32 {
    return point.x + point.y;
}

const assert = @import("std").debug.assert;

test "pass aggregate type by non-copy value to function" {
    assert(foo(Point{ .x = 1, .y = 2 }) == 3);
}

$ zig test test.zig
Test 1/1 pass aggregate type by non-copy value to function...OK
All tests passed.

In this case, the value may be passed by reference, or by value, whichever way Zig decides will be faster.

For extern functions, Zig follows the C ABI for passing structs and unions by value.

Function Reflection

test.zig

const assert = @import("std").debug.assert;

test "fn reflection" {
    assert(@typeOf(assert).ReturnType == void);
    assert(@typeOf(assert).is_var_args == false);
}

$ zig test test.zig
Test 1/1 fn reflection...OK
All tests passed.

Errors

Error Set Type

An error set is like an enum. However, each error name across the entire compilation gets assigned an unsigned integer greater than 0. You are allowed to declare the same error name more than once, and if you do, it gets assigned the same integer value.

The number of unique error values across the entire compilation should determine the size of the error set type. However right now it is hard coded to be a u16. See #768.

You can implicitly cast an error from a subset to its superset:

test.zig

const std = @import("std");

const FileOpenError = error {
    AccessDenied,
    OutOfMemory,
    FileNotFound,
};

const AllocationError = error {
    OutOfMemory,
};

test "implicit cast subset to superset" {
    const err = foo(AllocationError.OutOfMemory);
    std.debug.assert(err == FileOpenError.OutOfMemory);
}

fn foo(err: AllocationError) FileOpenError {
    return err;
}

$ zig test test.zig
Test 1/1 implicit cast subset to superset...OK
All tests passed.

But you cannot implicitly cast an error from a superset to a subset:

test.zig

const FileOpenError = error {
    AccessDenied,
    OutOfMemory,
    FileNotFound,
};

const AllocationError = error {
    OutOfMemory,
};

test "implicit cast superset to subset" {
    foo(FileOpenError.OutOfMemory) catch {};
}

fn foo(err: FileOpenError) AllocationError {
    return err;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:16:12: error: expected type 'AllocationError', found 'FileOpenError'
    return err;
           ^
/home/andy/dev/zig/docgen_tmp/test.zig:2:5: note: 'error.AccessDenied' not a member of destination error set
    AccessDenied,
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:4:5: note: 'error.FileNotFound' not a member of destination error set
    FileNotFound,
    ^

There is a shortcut for declaring an error set with only 1 value, and then getting that value:

const err = error.FileNotFound;

This is equivalent to:

const err = (error {FileNotFound}).FileNotFound;

This becomes useful when using Inferred Error Sets.

The Global Error Set

error refers to the global error set. This is the error set that contains all errors in the entire compilation unit. It is a superset of all other error sets and a subset of none of them.

You can implicitly cast any error set to the global one, and you can explicitly cast an error of global error set to a non-global one. This inserts a language-level assert to make sure the error value is in fact in the destination error set.

The global error set should generally be avoided because it prevents the compiler from knowing what errors are possible at compile-time. Knowing the error set at compile-time is better for generated documentation and helpful error messages, such as forgetting a possible error value in a switch.

Error Union Type

An error set type and normal type can be combined with the ! binary operator to form an error union type. You are likely to use an error union type more often than an error set type by itself.

Here is a function to parse a string into a 64-bit integer:

test.zig

pub fn parseU64(buf: []const u8, radix: u8) !u64 {
    var x: u64 = 0;

    for (buf) |c| {
        const digit = charToDigit(c);

        if (digit >= radix) {
            return error.InvalidChar;
        }

        // x *= radix
        if (@mulWithOverflow(u64, x, radix, &x)) {
            return error.Overflow;
        }

        // x += digit
        if (@addWithOverflow(u64, x, digit, &x)) {
            return error.Overflow;
        }
    }

    return x;
}

fn charToDigit(c: u8) u8 {
    return switch (c) {
        '0' ... '9' => c - '0',
        'A' ... 'Z' => c - 'A' + 10,
        'a' ... 'z' => c - 'a' + 10,
        else => @maxValue(u8),
    };
}

test "parse u64" {
    const result = try parseU64("1234", 10);
    @import("std").debug.assert(result == 1234);
}

$ zig test test.zig
Test 1/1 parse u64...OK
All tests passed.

Notice the return type is !u64. This means that the function either returns an unsigned 64 bit integer, or an error. We left off the error set to the left of the !, so the error set is inferred.

Within the function definition, you can see some return statements that return an error, and at the bottom a return statement that returns a u64. Both types implicitly cast to error!u64.

What it looks like to use this function varies depending on what you're trying to do. One of the following:

You want to provide a default value if it returned an error.
If it returned an error then you want to return the same error.
You know with complete certainty it will not return an error, so want to unconditionally unwrap it.
You want to take a different action for each possible error.

catch

If you want to provide a default value, you can use the catch binary operator:

fn doAThing(str: []u8) void {
    const number = parseU64(str, 10) catch 13;
    // ...
}

In this code, number will be equal to the successfully parsed string, or a default value of 13. The type of the right hand side of the binary catch operator must match the unwrapped error union type, or be of type noreturn.

try

Let's say you wanted to return the error if you got one, otherwise continue with the function logic:

fn doAThing(str: []u8) !void {
    const number = parseU64(str, 10) catch |err| return err;
    // ...
}

There is a shortcut for this. The try expression:

fn doAThing(str: []u8) !void {
    const number = try parseU64(str, 10);
    // ...
}

try evaluates an error union expression. If it is an error, it returns from the current function with the same error. Otherwise, the expression results in the unwrapped value.

Maybe you know with complete certainty that an expression will never be an error. In this case you can do this:

const number = parseU64("1234", 10) catch unreachable;

Here we know for sure that "1234" will parse successfully. So we put the unreachable value on the right hand side. unreachable generates a panic in Debug and ReleaseSafe modes and undefined behavior in ReleaseFast mode. So, while we're debugging the application, if there was a surprise error here, the application would crash appropriately.

Finally, you may want to take a different action for every situation. For that, we combine the if and switch expression:

fn doAThing(str: []u8) void {
    if (parseU64(str, 10)) |number| {
        doSomethingWithNumber(number);
    } else |err| switch (err) {
        error.Overflow => {
            // handle overflow...
        },
        // we promise that InvalidChar won't happen (or crash in debug mode if it does)
        error.InvalidChar => unreachable,
    }
}

errdefer

The other component to error handling is defer statements. In addition to an unconditional defer, Zig has errdefer, which evaluates the deferred expression on block exit path if and only if the function returned with an error from the block.

Example:

fn createFoo(param: i32) !Foo {
    const foo = try tryToAllocateFoo();
    // now we have allocated foo. we need to free it if the function fails.
    // but we want to return it if the function succeeds.
    errdefer deallocateFoo(foo);

    const tmp_buf = allocateTmpBuffer() orelse return error.OutOfMemory;
    // tmp_buf is truly a temporary resource, and we for sure want to clean it up
    // before this block leaves scope
    defer deallocateTmpBuffer(tmp_buf);

    if (param > 1337) return error.InvalidParam;

    // here the errdefer will not run since we're returning success from the function.
    // but the defer will run!
    return foo;
}

The neat thing about this is that you get robust error handling without the verbosity and cognitive overhead of trying to make sure every exit path is covered. The deallocation code is always directly following the allocation code.

A couple of other tidbits about error handling:

These primitives give enough expressiveness that it's completely practical to have failing to check for an error be a compile error. If you really want to ignore the error, you can add catch unreachable and get the added benefit of crashing in Debug and ReleaseSafe modes if your assumption was wrong.
Since Zig understands error types, it can pre-weight branches in favor of errors not occuring. Just a small optimization benefit that is not available in other languages.

See also:

defer
if
switch

An error union is created with the ! binary operator. You can use compile-time reflection to access the child type of an error union:

test.zig

const assert = @import("std").debug.assert;

test "error union" {
    var foo: error!i32 = undefined;

    // Implicitly cast from child type of an error union:
    foo = 1234;

    // Implicitly cast from an error set:
    foo = error.SomeError;

    // Use compile-time reflection to access the payload type of an error union:
    comptime assert(@typeOf(foo).Payload == i32);

    // Use compile-time reflection to access the error set type of an error union:
    comptime assert(@typeOf(foo).ErrorSet == error);
}

$ zig test test.zig
Test 1/1 error union...OK
All tests passed.

Merging Error Sets

Use the || operator to merge two error sets together. The resulting error set contains the errors of both error sets. Doc comments from the left-hand side override doc comments from the right-hand side. In this example, the doc comments for C.PathNotFound is A doc comment.

This is especially useful for functions which return different error sets depending on comptime branches. For example, the Zig standard library uses LinuxFileOpenError || WindowsFileOpenError for the error set of opening files.

test.zig

const A = error{
    NotDir,

    /// A doc comment
    PathNotFound,
};
const B = error{
    OutOfMemory,

    /// B doc comment
    PathNotFound,
};

const C = A || B;

fn foo() C!void {
    return error.NotDir;
}

test "merge error sets" {
    if (foo()) {
        @panic("unexpected");
    } else |err| switch (err) {
        error.OutOfMemory => @panic("unexpected"),
        error.PathNotFound => @panic("unexpected"),
        error.NotDir => {},
    }
}

$ zig test test.zig
Test 1/1 merge error sets...OK
All tests passed.

Inferred Error Sets

Because many functions in Zig return a possible error, Zig supports inferring the error set. To infer the error set for a function, use this syntax:

test.zig

// With an inferred error set
pub fn add_inferred(comptime T: type, a: T, b: T) !T {
    var answer: T = undefined;
    return if (@addWithOverflow(T, a, b, &answer)) error.Overflow else answer;
}

// With an explicit error set
pub fn add_explicit(comptime T: type, a: T, b: T) Error!T {
    var answer: T = undefined;
    return if (@addWithOverflow(T, a, b, &answer)) error.Overflow else answer;
}

const Error = error {
    Overflow,
};

const std = @import("std");

test "inferred error set" {
    if (add_inferred(u8, 255, 1)) |_| unreachable else |err| switch (err) {
        error.Overflow => {}, // ok
    }
}

$ zig test test.zig
Test 1/1 inferred error set...OK
All tests passed.

When a function has an inferred error set, that function becomes generic and thus it becomes trickier to do certain things with it, such as obtain a function pointer, or have an error set that is consistent across different build targets. Additionally, inferred error sets are incompatible with recursion.

In these situations, it is recommended to use an explicit error set. You can generally start with an empty error set and let compile errors guide you toward completing the set.

These limitations may be overcome in a future version of Zig.

Error Return Traces

Error Return Traces show all the points in the code that an error was returned to the calling function. This makes it practical to use try everywhere and then still be able to know what happened if an error ends up bubbling all the way out of your application.

test.zig

pub fn main() !void {
    try foo(12);
}

fn foo(x: i32) !void {
    if (x >= 5) {
        try bar();
    } else {
        try bang2();
    }
}

fn bar() !void {
    if (baz()) {
        try quux();
    } else |err| switch (err) {
        error.FileNotFound => try hello(),
        else => try another(),
    }
}

fn baz() !void {
    try bang1();
}

fn quux() !void {
    try bang2();
}

fn hello() !void {
    try bang2();
}

fn another() !void {
    try bang1();
}

fn bang1() !void {
    return error.FileNotFound;
}

fn bang2() !void {
    return error.PermissionDenied;
}

$ zig build-exe test.zig
$ ./test
error: PermissionDenied
/home/andy/dev/zig/docgen_tmp/test.zig:39:5: 0x222bc9 in ??? (test)
    return error.FileNotFound;
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:23:5: 0x222aad in ??? (test)
    try bang1();
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:43:5: 0x222a79 in ??? (test)
    return error.PermissionDenied;
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:31:5: 0x222b9d in ??? (test)
    try bang2();
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:17:31: 0x222a49 in ??? (test)
        error.FileNotFound => try hello(),
                              ^
/home/andy/dev/zig/docgen_tmp/test.zig:7:9: 0x22293a in ??? (test)
        try bar();
        ^
/home/andy/dev/zig/docgen_tmp/test.zig:2:5: 0x2227e2 in ??? (test)
    try foo(12);
    ^

Look closely at this example. This is no stack trace.

You can see that the final error bubbled up was PermissionDenied, but the original error that started this whole thing was FileNotFound. In the bar function, the code handles the original error code, and then returns another one, from the switch statement. Error Return Traces make this clear, whereas a stack trace would look like this:

test.zig

pub fn main() void {
    foo(12);
}

fn foo(x: i32) void {
    if (x >= 5) {
        bar();
    } else {
        bang2();
    }
}

fn bar() void {
    if (baz()) {
        quux();
    } else {
        hello();
    }
}

fn baz() bool {
    return bang1();
}

fn quux() void {
    bang2();
}

fn hello() void {
    bang2();
}

fn bang1() bool {
    return false;
}

fn bang2() void {
    @panic("PermissionDenied");
}

$ zig build-exe test.zig
$ ./test
PermissionDenied
/home/andy/dev/zig/docgen_tmp/test.zig:38:5: 0x222734 in ??? (test)
    @panic("PermissionDenied");
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:30:10: 0x222779 in ??? (test)
    bang2();
         ^
/home/andy/dev/zig/docgen_tmp/test.zig:17:14: 0x22271b in ??? (test)
        hello();
             ^
/home/andy/dev/zig/docgen_tmp/test.zig:7:12: 0x2226e6 in ??? (test)
        bar();
           ^
/home/andy/dev/zig/docgen_tmp/test.zig:2:8: 0x2226ce in ??? (test)
    foo(12);
       ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Here, the stack trace does not explain how the control flow in bar got to the hello() call. One would have to open a debugger or further instrument the application in order to find out. The error return trace, on the other hand, shows exactly how the error bubbled up.

This debugging feature makes it easier to iterate quickly on code that robustly handles all error conditions. This means that Zig developers will naturally find themselves writing correct, robust code in order to increase their development pace.

Error Return Traces are enabled by default in Debug and ReleaseSafe builds and disabled by default in ReleaseFast and ReleaseSmall builds.

There are a few ways to activate this error return tracing feature:

Return an error from main
An error makes its way to catch unreachable and you have not overridden the default panic handler
Use errorReturnTrace to access the current return trace. You can use std.debug.dumpStackTrace to print it. This function returns comptime-known null when building without error return tracing support.

Implementation Details

To analyze performance cost, there are two cases:

when no errors are returned
when returning errors

For the case when no errors are returned, the cost is a single memory write operation, only in the first non-failable function in the call graph that calls a failable function, i.e. when a function returning void calls a function returning error. This is to initialize this struct in the stack memory:

pub const StackTrace = struct {
    index: usize,
    instruction_addresses: [N]usize,
};

Here, N is the maximum function call depth as determined by call graph analysis. Recursion is ignored and counts for 2.

A pointer to StackTrace is passed as a secret parameter to every function that can return an error, but it's always the first parameter, so it can likely sit in a register and stay there.

That's it for the path when no errors occur. It's practically free in terms of performance.

When generating the code for a function that returns an error, just before the return statement (only for the return statements that return errors), Zig generates a call to this function:

// marked as "no-inline" in LLVM IR
fn __zig_return_error(stack_trace: *StackTrace) void {
    stack_trace.instruction_addresses[stack_trace.index] = @returnAddress();
    stack_trace.index = (stack_trace.index + 1) % N;
}

The cost is 2 math operations plus some memory reads and writes. The memory accessed is constrained and should remain cached for the duration of the error return bubbling.

As for code size cost, 1 function call before a return statement is no big deal. Even so, I have a plan to make the call to __zig_return_error a tail call, which brings the code size cost down to actually zero. What is a return statement in code without error return tracing can become a jump instruction in code with error return tracing.

Optionals

One area that Zig provides safety without compromising efficiency or readability is with the optional type.

The question mark symbolizes the optional type. You can convert a type to an optional type by putting a question mark in front of it, like this:

// normal integer
const normal_int: i32 = 1234;

// optional integer
const optional_int: ?i32 = 5678;

Now the variable optional_int could be an i32, or null.

Instead of integers, let's talk about pointers. Null references are the source of many runtime exceptions, and even stand accused of being the worst mistake of computer science.

Zig does not have them.

Instead, you can use an optional pointer. This secretly compiles down to a normal pointer, since we know we can use 0 as the null value for the optional type. But the compiler can check your work and make sure you don't assign null to something that can't be null.

Typically the downside of not having null is that it makes the code more verbose to write. But, let's compare some equivalent C code and Zig code.

Task: call malloc, if the result is null, return null.

C code

// malloc prototype included for reference
void *malloc(size_t size);

struct Foo *do_a_thing(void) {
    char *ptr = malloc(1234);
    if (!ptr) return NULL;
    // ...
}

Zig code

// malloc prototype included for reference
extern fn malloc(size: size_t) ?*u8;

fn doAThing() ?*Foo {
    const ptr = malloc(1234) orelse return null;
    // ...
}

Here, Zig is at least as convenient, if not more, than C. And, the type of "ptr" is *u8 not ?*u8. The orelse keyword unwrapped the optional type and therefore ptr is guaranteed to be non-null everywhere it is used in the function.

The other form of checking against NULL you might see looks like this:

void do_a_thing(struct Foo *foo) {
    // do some stuff

    if (foo) {
        do_something_with_foo(foo);
    }

    // do some stuff
}

In Zig you can accomplish the same thing:

fn doAThing(optional_foo: ?*Foo) void {
    // do some stuff

    if (optional_foo) |foo| {
      doSomethingWithFoo(foo);
    }

    // do some stuff
}

Once again, the notable thing here is that inside the if block, foo is no longer an optional pointer, it is a pointer, which cannot be null.

One benefit to this is that functions which take pointers as arguments can be annotated with the "nonnull" attribute - __attribute__((nonnull)) in GCC. The optimizer can sometimes make better decisions knowing that pointer arguments cannot be null.

Optional Type

An optional is created by putting ? in front of a type. You can use compile-time reflection to access the child type of an optional:

test.zig

const assert = @import("std").debug.assert;

test "optional type" {
    // Declare an optional and implicitly cast from null:
    var foo: ?i32 = null;

    // Implicitly cast from child type of an optional
    foo = 1234;

    // Use compile-time reflection to access the child type of the optional:
    comptime assert(@typeOf(foo).Child == i32);
}

$ zig test test.zig
Test 1/1 optional type...OK
All tests passed.

null

Just like undefined, null has its own type, and the only way to use it is to cast it to a different type:

const optional_value: ?i32 = null;

Casting

A type cast converts a value of one type to another. Zig has Implicit Casts for conversions that are known to be completely safe and unambiguous, and Explicit Casts for conversions that one would not want to happen on accident. There is also a third kind of type conversion called Peer Type Resolution for the case when a result type must be decided given multiple operand types.

Implicit Casts

An implicit cast occurs when one type is expected, but different type is provided:

test.zig

test "implicit cast - variable declaration" {
    var a: u8 = 1;
    var b: u16 = a;
}

test "implicit cast - function call" {
    var a: u8 = 1;
    foo(a);
}

fn foo(b: u16) void {}

test "implicit cast - invoke a type as a function" {
    var a: u8 = 1;
    var b = u16(a);
}

$ zig test test.zig
Test 1/3 implicit cast - variable declaration...OK
Test 2/3 implicit cast - function call...OK
Test 3/3 implicit cast - invoke a type as a function...OK
All tests passed.

Implicit casts are only allowed when it is completely unambiguous how to get from one type to another, and the transformation is guaranteed to be safe.

Implicit Cast: Stricter Qualification

Values which have the same representation at runtime can be cast to increase the strictness of the qualifiers, no matter how nested the qualifiers are:

const - non-const to const is allowed
volatile - non-volatile to volatile is allowed
align - bigger to smaller alignment is allowed
error sets to supersets is allowed

These casts are no-ops at runtime since the value representation does not change.

test.zig

test "implicit cast - const qualification" {
    var a: i32 = 1;
    var b: *i32 = &a;
    foo(b);
}

fn foo(a: *const i32) void {}

$ zig test test.zig
Test 1/1 implicit cast - const qualification...OK
All tests passed.

In addition, pointers implicitly cast to const optional pointers:

test.zig

const std = @import("std");
const assert = std.debug.assert;
const mem = std.mem;

test "cast *[1][*]const u8 to [*]const ?[*]const u8" {
    const window_name = [1][*]const u8{c"window name"};
    const x: [*]const ?[*]const u8 = &window_name;
    assert(mem.eql(u8, std.cstr.toSliceConst(x[0].?), "window name"));
}

$ zig test test.zig
Test 1/1 cast *[1][*]const u8 to [*]const ?[*]const u8...OK
All tests passed.

Implicit Cast: Integer and Float Widening

Integers implicitly cast to integer types which can represent every value of the old type, and likewise Floats implicitly cast to float types which can represent every value of the old type.

test.zig

const std = @import("std");
const assert = std.debug.assert;
const mem = std.mem;

test "integer widening" {
    var a: u8 = 250;
    var b: u16 = a;
    var c: u32 = b;
    var d: u64 = c;
    var e: u64 = d;
    var f: u128 = e;
    assert(f == a);
}

test "implicit unsigned integer to signed integer" {
    var a: u8 = 250;
    var b: i16 = a;
    assert(b == 250);
}

test "float widening" {
    var a: f16 = 12.34;
    var b: f32 = a;
    var c: f64 = b;
    var d: f128 = c;
    assert(d == a);
}

$ zig test test.zig
Test 1/3 integer widening...OK
Test 2/3 implicit unsigned integer to signed integer...OK
Test 3/3 float widening...OK
All tests passed.

Implicit Cast: Arrays

TODO: [N]T to []const T

TODO: *const [N]T to []const T

TODO: [N]T to *const []const T

TODO: [N]T to ?[]const T

TODO: *[N]T to []T

TODO: *[N]T to [*]T

TODO: *[N]T to ?[*]T

TODO: *T to *[1]T

TODO: [N]T to E![]const T

Implicit Cast: Optionals

TODO: T to ?T

TODO: T to E!?T

TODO: null to ?T

Implicit Cast: T to E!T

TODO

Implicit Cast: E to E!T

TODO

Implicit Cast: comptime_int to *const integer

TODO

Implicit Cast: comptime_float to *const float

TODO

Implicit Cast: compile-time known numbers

TODO

Implicit Cast: union to enum

TODO

Implicit Cast: enum to union

TODO

Implicit Cast: T to *T when @sizeOf(T) == 0

TODO

Implicit Cast: undefined

TODO

Implicit Cast: T to *const T

TODO

Explicit Casts

Explicit casts are performed via Builtin Functions. Some explicit casts are safe; some are not. Some explicit casts perform language-level assertions; some do not. Some explicit casts are no-ops at runtime; some are not.

@bitCast - change type but maintain bit representation
@alignCast - make a pointer have more alignment
@boolToInt - convert true to 1 and false to 0
@bytesToSlice - convert a slice of bytes to a slice of another type
@enumToInt - obtain the integer tag value of an enum or tagged union
@errSetCast - convert to a smaller error set
@errorToInt - obtain the integer value of an error code
@floatCast - convert a larger float to a smaller float
@floatToInt - obtain the integer part of a float value
@intCast - convert between integer types
@intToEnum - obtain an enum value based on its integer tag value
@intToError - obtain an error code based on its integer value
@intToFloat - convert an integer to a float value
@intToPtr - convert an address to a pointer
@ptrCast - convert between pointer types
@ptrToInt - obtain the address of a pointer
@sliceToBytes - convert a slice of anything to a slice of bytes
@truncate - convert between integer types, chopping off bits

Peer Type Resolution

Peer Type Resolution occurs in these places:

switch expressions
if expressions
while expressions
for expressions
Multiple break statements in a block
Some binary operations

This kind of type resolution chooses a type that all peer types can implicitly cast into. Here are some examples:

test.zig

const std = @import("std");
const assert = std.debug.assert;
const mem = std.mem;

test "peer resolve int widening" {
    var a: i8 = 12;
    var b: i16 = 34;
    var c = a + b;
    assert(c == 46);
    assert(@typeOf(c) == i16);
}

test "peer resolve arrays of different size to const slice" {
    assert(mem.eql(u8, boolToStr(true), "true"));
    assert(mem.eql(u8, boolToStr(false), "false"));
    comptime assert(mem.eql(u8, boolToStr(true), "true"));
    comptime assert(mem.eql(u8, boolToStr(false), "false"));
}
fn boolToStr(b: bool) []const u8 {
    return if (b) "true" else "false";
}

test "peer resolve array and const slice" {
    testPeerResolveArrayConstSlice(true);
    comptime testPeerResolveArrayConstSlice(true);
}
fn testPeerResolveArrayConstSlice(b: bool) void {
    const value1 = if (b) "aoeu" else ([]const u8)("zz");
    const value2 = if (b) ([]const u8)("zz") else "aoeu";
    assert(mem.eql(u8, value1, "aoeu"));
    assert(mem.eql(u8, value2, "zz"));
}

test "peer type resolution: ?T and T" {
    assert(peerTypeTAndOptionalT(true, false).? == 0);
    assert(peerTypeTAndOptionalT(false, false).? == 3);
    comptime {
        assert(peerTypeTAndOptionalT(true, false).? == 0);
        assert(peerTypeTAndOptionalT(false, false).? == 3);
    }
}
fn peerTypeTAndOptionalT(c: bool, b: bool) ?usize {
    if (c) {
        return if (b) null else usize(0);
    }

    return usize(3);
}

test "peer type resolution: [0]u8 and []const u8" {
    assert(peerTypeEmptyArrayAndSlice(true, "hi").len == 0);
    assert(peerTypeEmptyArrayAndSlice(false, "hi").len == 1);
    comptime {
        assert(peerTypeEmptyArrayAndSlice(true, "hi").len == 0);
        assert(peerTypeEmptyArrayAndSlice(false, "hi").len == 1);
    }
}
fn peerTypeEmptyArrayAndSlice(a: bool, slice: []const u8) []const u8 {
    if (a) {
        return []const u8{};
    }

    return slice[0..1];
}
test "peer type resolution: [0]u8, []const u8, and error![]u8" {
    {
        var data = "hi";
        const slice = data[0..];
        assert((try peerTypeEmptyArrayAndSliceAndError(true, slice)).len == 0);
        assert((try peerTypeEmptyArrayAndSliceAndError(false, slice)).len == 1);
    }
    comptime {
        var data = "hi";
        const slice = data[0..];
        assert((try peerTypeEmptyArrayAndSliceAndError(true, slice)).len == 0);
        assert((try peerTypeEmptyArrayAndSliceAndError(false, slice)).len == 1);
    }
}
fn peerTypeEmptyArrayAndSliceAndError(a: bool, slice: []u8) error![]u8 {
    if (a) {
        return []u8{};
    }

    return slice[0..1];
}

$ zig test test.zig
Test 1/6 peer resolve int widening...OK
Test 2/6 peer resolve arrays of different size to const slice...OK
Test 3/6 peer resolve array and const slice...OK
Test 4/6 peer type resolution: ?T and T...OK
Test 5/6 peer type resolution: [0]u8 and []const u8...OK
Test 6/6 peer type resolution: [0]u8, []const u8, and error![]u8...OK
All tests passed.

void

void represents a type that has no value. Code that makes use of void values is not included in the final generated code:

export fn entry() void {
    var x: void = {};
    var y: void = {};
    x = y;
}

When this turns into LLVM IR, there is no code generated in the body of entry, even in debug mode. For example, on x86_64:

0000000000000010 <entry>:
  10:	55                   	push   %rbp
  11:	48 89 e5             	mov    %rsp,%rbp
  14:	5d                   	pop    %rbp
  15:	c3                   	retq

These assembly instructions do not have any code associated with the void values - they only perform the function call prologue and epilog.

void can be useful for instantiating generic types. For example, given a Map(Key, Value), one can pass void for the Value type to make it into a Set:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "turn HashMap into a set with void" {
    var map = std.HashMap(i32, void, hash_i32, eql_i32).init(std.debug.global_allocator);
    defer map.deinit();

    _ = try map.put(1, {});
    _ = try map.put(2, {});

    assert(map.contains(2));
    assert(!map.contains(3));

    _ = map.remove(2);
    assert(!map.contains(2));
}

fn hash_i32(x: i32) u32 {
    return @bitCast(u32, x);
}

fn eql_i32(a: i32, b: i32) bool {
    return a == b;
}

$ zig test test.zig
Test 1/1 turn HashMap into a set with void...OK
All tests passed.

Note that this is different than using a dummy value for the hash map value. By using void as the type of the value, the hash map entry type has no value field, and thus the hash map takes up less space. Further, all the code that deals with storing and loading the value is deleted, as seen above.

void is distinct from c_void, which is defined like this: pub const c_void = @OpaqueType();. void has a known size of 0 bytes, and c_void has an unknown, but non-zero, size.

Expressions of type void are the only ones whose value can be ignored. For example:

test.zig

test "ignoring expression value" {
    foo();
}

fn foo() i32 {
    return 1234;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:8: error: expression value is ignored
    foo();
       ^

However, if the expression has type void:

test.zig

test "ignoring expression value" {
    foo();
}

fn foo() void {}

$ zig test test.zig
Test 1/1 ignoring expression value...OK
All tests passed.

comptime

Zig places importance on the concept of whether an expression is known at compile-time. There are a few different places this concept is used, and these building blocks are used to keep the language small, readable, and powerful.

Introducing the Compile-Time Concept

Compile-Time Parameters

Compile-time parameters is how Zig implements generics. It is compile-time duck typing.

fn max(comptime T: type, a: T, b: T) T {
    return if (a > b) a else b;
}
fn gimmeTheBiggerFloat(a: f32, b: f32) f32 {
    return max(f32, a, b);
}
fn gimmeTheBiggerInteger(a: u64, b: u64) u64 {
    return max(u64, a, b);
}

In Zig, types are first-class citizens. They can be assigned to variables, passed as parameters to functions, and returned from functions. However, they can only be used in expressions which are known at compile-time, which is why the parameter T in the above snippet must be marked with comptime.

A comptime parameter means that:

At the callsite, the value must be known at compile-time, or it is a compile error.
In the function definition, the value is known at compile-time.

For example, if we were to introduce another function to the above snippet:

test.zig

fn max(comptime T: type, a: T, b: T) T {
    return if (a > b) a else b;
}
test "try to pass a runtime type" {
    foo(false);
}
fn foo(condition: bool) void {
    const result = max(
        if (condition) f32 else u64,
        1234,
        5678);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:1:29: error: unable to evaluate constant expression
fn max(comptime T: type, a: T, b: T) T {
                            ^

This is an error because the programmer attempted to pass a value only known at run-time to a function which expects a value known at compile-time.

Another way to get an error is if we pass a type that violates the type checker when the function is analyzed. This is what it means to have compile-time duck typing.

For example:

test.zig

fn max(comptime T: type, a: T, b: T) T {
    return if (a > b) a else b;
}
test "try to compare bools" {
    _ = max(bool, true, false);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:18: error: operator not allowed for type 'bool'
    return if (a > b) a else b;
                 ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:12: note: called from here
    _ = max(bool, true, false);
           ^

On the flip side, inside the function definition with the comptime parameter, the value is known at compile-time. This means that we actually could make this work for the bool type if we wanted to:

test.zig

fn max(comptime T: type, a: T, b: T) T {
    if (T == bool) {
        return a or b;
    } else if (a > b) {
        return a;
    } else {
        return b;
    }
}
test "try to compare bools" {
    @import("std").debug.assert(max(bool, false, true) == true);
}

$ zig test test.zig
Test 1/1 try to compare bools...OK
All tests passed.

This works because Zig implicitly inlines if expressions when the condition is known at compile-time, and the compiler guarantees that it will skip analysis of the branch not taken.

This means that the actual function generated for max in this situation looks like this:

fn max(a: bool, b: bool) bool {
    return a or b;
}

All the code that dealt with compile-time known values is eliminated and we are left with only the necessary run-time code to accomplish the task.

This works the same way for switch expressions - they are implicitly inlined when the target expression is compile-time known.

Compile-Time Variables

In Zig, the programmer can label variables as comptime. This guarantees to the compiler that every load and store of the variable is performed at compile-time. Any violation of this results in a compile error.

This combined with the fact that we can inline loops allows us to write a function which is partially evaluated at compile-time and partially at run-time.

For example:

comptime_vars.zig

const assert = @import("std").debug.assert;

const CmdFn = struct {
    name: []const u8,
    func: fn(i32) i32,
};

const cmd_fns = []CmdFn{
    CmdFn {.name = "one", .func = one},
    CmdFn {.name = "two", .func = two},
    CmdFn {.name = "three", .func = three},
};
fn one(value: i32) i32 { return value + 1; }
fn two(value: i32) i32 { return value + 2; }
fn three(value: i32) i32 { return value + 3; }

fn performFn(comptime prefix_char: u8, start_value: i32) i32 {
    var result: i32 = start_value;
    comptime var i = 0;
    inline while (i < cmd_fns.len) : (i += 1) {
        if (cmd_fns[i].name[0] == prefix_char) {
            result = cmd_fns[i].func(result);
        }
    }
    return result;
}

test "perform fn" {
    assert(performFn('t', 1) == 6);
    assert(performFn('o', 0) == 1);
    assert(performFn('w', 99) == 99);
}

$ zig test comptime_vars.zig
Test 1/1 perform fn...OK
All tests passed.

This example is a bit contrived, because the compile-time evaluation component is unnecessary; this code would work fine if it was all done at run-time. But it does end up generating different code. In this example, the function performFn is generated three different times, for the different values of prefix_char provided:

// From the line:
// assert(performFn('t', 1) == 6);
fn performFn(start_value: i32) i32 {
    var result: i32 = start_value;
    result = two(result);
    result = three(result);
    return result;
}

// From the line:
// assert(performFn('o', 0) == 1);
fn performFn(start_value: i32) i32 {
    var result: i32 = start_value;
    result = one(result);
    return result;
}

// From the line:
// assert(performFn('w', 99) == 99);
fn performFn(start_value: i32) i32 {
    var result: i32 = start_value;
    return result;
}

Note that this happens even in a debug build; in a release build these generated functions still pass through rigorous LLVM optimizations. The important thing to note, however, is not that this is a way to write more optimized code, but that it is a way to make sure that what should happen at compile-time, does happen at compile-time. This catches more errors and as demonstrated later in this article, allows expressiveness that in other languages requires using macros, generated code, or a preprocessor to accomplish.

Compile-Time Expressions

In Zig, it matters whether a given expression is known at compile-time or run-time. A programmer can use a comptime expression to guarantee that the expression will be evaluated at compile-time. If this cannot be accomplished, the compiler will emit an error. For example:

test.zig

extern fn exit() noreturn;

test "foo" {
    comptime {
        exit();
    }
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:5:9: error: unable to evaluate constant expression
        exit();
        ^

It doesn't make sense that a program could call exit() (or any other external function) at compile-time, so this is a compile error. However, a comptime expression does much more than sometimes cause a compile error.

Within a comptime expression:

All variables are comptime variables.
All if, while, for, and switch expressions are evaluated at compile-time, or emit a compile error if this is not possible.
All function calls cause the compiler to interpret the function at compile-time, emitting a compile error if the function tries to do something that has global run-time side effects.

This means that a programmer can create a function which is called both at compile-time and run-time, with no modification to the function required.

Let's look at an example:

test.zig

const assert = @import("std").debug.assert;

fn fibonacci(index: u32) u32 {
    if (index < 2) return index;
    return fibonacci(index - 1) + fibonacci(index - 2);
}

test "fibonacci" {
    // test fibonacci at run-time
    assert(fibonacci(7) == 13);

    // test fibonacci at compile-time
    comptime {
        assert(fibonacci(7) == 13);
    }
}

$ zig test test.zig
Test 1/1 fibonacci...OK
All tests passed.

Imagine if we had forgotten the base case of the recursive function and tried to run the tests:

test.zig

const assert = @import("std").debug.assert;

fn fibonacci(index: u32) u32 {
    //if (index < 2) return index;
    return fibonacci(index - 1) + fibonacci(index - 2);
}

test "fibonacci" {
    comptime {
        assert(fibonacci(7) == 13);
    }
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:5:28: error: operation caused overflow
    return fibonacci(index - 1) + fibonacci(index - 2);
                           ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:10:25: note: called from here
        assert(fibonacci(7) == 13);
                        ^

The compiler produces an error which is a stack trace from trying to evaluate the function at compile-time.

Luckily, we used an unsigned integer, and so when we tried to subtract 1 from 0, it triggered undefined behavior, which is always a compile error if the compiler knows it happened. But what would have happened if we used a signed integer?

test.zig

const assert = @import("std").debug.assert;

fn fibonacci(index: i32) i32 {
    //if (index < 2) return index;
    return fibonacci(index - 1) + fibonacci(index - 2);
}

test "fibonacci" {
    comptime {
        assert(fibonacci(7) == 13);
    }
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: error: evaluation exceeded 1000 backwards branches
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:21: note: called from here
    return fibonacci(index - 1) + fibonacci(index - 2);
                    ^

The compiler noticed that evaluating this function at compile-time took a long time, and thus emitted a compile error and gave up. If the programmer wants to increase the budget for compile-time computation, they can use a built-in function called @setEvalBranchQuota to change the default number 1000 to something else.

What if we fix the base case, but put the wrong value in the assert line?

test.zig

const assert = @import("std").debug.assert;

fn fibonacci(index: i32) i32 {
    if (index < 2) return index;
    return fibonacci(index - 1) + fibonacci(index - 2);
}

test "fibonacci" {
    comptime {
        assert(fibonacci(7) == 99999);
    }
}

$ zig test test.zig
/home/andy/dev/zig/build/lib/zig/std/debug/index.zig:118:13: error: encountered @panic at compile-time
            @panic("assertion failure");
            ^
/home/andy/dev/zig/docgen_tmp/test.zig:10:15: note: called from here
        assert(fibonacci(7) == 99999);
              ^

What happened is Zig started interpreting the assert function with the parameter ok set to false. When the interpreter hit unreachable it emitted a compile error, because reaching unreachable code is undefined behavior, and undefined behavior causes a compile error if it is detected at compile-time.

In the global scope (outside of any function), all expressions are implicitly comptime expressions. This means that we can use functions to initialize complex static data. For example:

test.zig

const first_25_primes = firstNPrimes(25);
const sum_of_first_25_primes = sum(first_25_primes);

fn firstNPrimes(comptime n: usize) [n]i32 {
    var prime_list: [n]i32 = undefined;
    var next_index: usize = 0;
    var test_number: i32 = 2;
    while (next_index < prime_list.len) : (test_number += 1) {
        var test_prime_index: usize = 0;
        var is_prime = true;
        while (test_prime_index < next_index) : (test_prime_index += 1) {
            if (test_number % prime_list[test_prime_index] == 0) {
                is_prime = false;
                break;
            }
        }
        if (is_prime) {
            prime_list[next_index] = test_number;
            next_index += 1;
        }
    }
    return prime_list;
}

fn sum(numbers: []const i32) i32 {
    var result: i32 = 0;
    for (numbers) |x| {
        result += x;
    }
    return result;
}

test "variable values" {
    @import("std").debug.assert(sum_of_first_25_primes == 1060);
}

$ zig test test.zig
Test 1/1 variable values...OK
All tests passed.

When we compile this program, Zig generates the constants with the answer pre-computed. Here are the lines from the generated LLVM IR:

@0 = internal unnamed_addr constant [25 x i32] [i32 2, i32 3, i32 5, i32 7, i32 11, i32 13, i32 17, i32 19, i32 23, i32 29, i32 31, i32 37, i32 41, i32 43, i32 47, i32 53, i32 59, i32 61, i32 67, i32 71, i32 73, i32 79, i32 83, i32 89, i32 97]
@1 = internal unnamed_addr constant i32 1060

Note that we did not have to do anything special with the syntax of these functions. For example, we could call the sum function as is with a slice of numbers whose length and values were only known at run-time.

Generic Data Structures

Zig uses these capabilities to implement generic data structures without introducing any special-case syntax. If you followed along so far, you may already know how to create a generic data structure.

Here is an example of a generic List data structure, that we will instantiate with the type i32. In Zig we refer to the type as List(i32).

fn List(comptime T: type) type {
    return struct {
        items: []T,
        len: usize,
    };
}

That's it. It's a function that returns an anonymous struct. For the purposes of error messages and debugging, Zig infers the name "List(i32)" from the function name and parameters invoked when creating the anonymous struct.

To keep the language small and uniform, all aggregate types in Zig are anonymous. To give a type a name, we assign it to a constant:

const Node = struct {
    next: *Node,
    name: []u8,
};

This works because all top level declarations are order-independent, and as long as there isn't an actual infinite regression, values can refer to themselves, directly or indirectly. In this case, Node refers to itself as a pointer, which is not actually an infinite regression, so it works fine.

Case Study: printf in Zig

Putting all of this together, let's see how printf works in Zig.

printf.zig

const warn = @import("std").debug.warn;

const a_number: i32 = 1234;
const a_string = "foobar";

pub fn main() void {
    warn("here is a string: '{}' here is a number: {}\n", a_string, a_number);
}

$ zig build-exe printf.zig
$ ./printf
here is a string: 'foobar' here is a number: 1234

Let's crack open the implementation of this and see how it works:

/// Calls print and then flushes the buffer.
pub fn printf(self: *OutStream, comptime format: []const u8, args: ...) error!void {
    const State = enum {
        Start,
        OpenBrace,
        CloseBrace,
    };

    comptime var start_index: usize = 0;
    comptime var state = State.Start;
    comptime var next_arg: usize = 0;

    inline for (format) |c, i| {
        switch (state) {
            State.Start => switch (c) {
                '{' => {
                    if (start_index < i) try self.write(format[start_index..i]);
                    state = State.OpenBrace;
                },
                '}' => {
                    if (start_index < i) try self.write(format[start_index..i]);
                    state = State.CloseBrace;
                },
                else => {},
            },
            State.OpenBrace => switch (c) {
                '{' => {
                    state = State.Start;
                    start_index = i;
                },
                '}' => {
                    try self.printValue(args[next_arg]);
                    next_arg += 1;
                    state = State.Start;
                    start_index = i + 1;
                },
                else => @compileError("Unknown format character: " ++ c),
            },
            State.CloseBrace => switch (c) {
                '}' => {
                    state = State.Start;
                    start_index = i;
                },
                else => @compileError("Single '}' encountered in format string"),
            },
        }
    }
    comptime {
        if (args.len != next_arg) {
            @compileError("Unused arguments");
        }
        if (state != State.Start) {
            @compileError("Incomplete format string: " ++ format);
        }
    }
    if (start_index < format.len) {
        try self.write(format[start_index..format.len]);
    }
    try self.flush();
}

This is a proof of concept implementation; the actual function in the standard library has more formatting capabilities.

Note that this is not hard-coded into the Zig compiler; this is userland code in the standard library.

When this function is analyzed from our example code above, Zig partially evaluates the function and emits a function that actually looks like this:

pub fn printf(self: *OutStream, arg0: i32, arg1: []const u8) !void {
    try self.write("here is a string: '");
    try self.printValue(arg0);
    try self.write("' here is a number: ");
    try self.printValue(arg1);
    try self.write("\n");
    try self.flush();
}

printValue is a function that takes a parameter of any type, and does different things depending on the type:

pub fn printValue(self: *OutStream, value: var) !void {
    const T = @typeOf(value);
    if (@isInteger(T)) {
        return self.printInt(T, value);
    } else if (@isFloat(T)) {
        return self.printFloat(T, value);
    } else {
        @compileError("Unable to print type '" ++ @typeName(T) ++ "'");
    }
}

And now, what happens if we give too many arguments to printf?

test.zig

const warn = @import("std").debug.warn;

const a_number: i32 = 1234;
const a_string = "foobar";

test "printf too many arguments" {
    warn("here is a string: '{}' here is a number: {}\n",
        a_string, a_number, a_number);
}

$ zig test test.zig
/home/andy/dev/zig/build/lib/zig/std/fmt/index.zig:95:13: error: Unused arguments
            @compileError("Unused arguments");
            ^
/home/andy/dev/zig/build/lib/zig/std/io.zig:227:34: note: called from here
            return std.fmt.format(self, Error, self.writeFn, format, args);
                                 ^
/home/andy/dev/zig/build/lib/zig/std/debug/index.zig:46:17: note: called from here
    stderr.print(fmt, args) catch return;
                ^
/home/andy/dev/zig/docgen_tmp/test.zig:7:9: note: called from here
    warn("here is a string: '{}' here is a number: {}\n",
        ^

Zig gives programmers the tools needed to protect themselves against their own mistakes.

Zig doesn't care whether the format argument is a string literal, only that it is a compile-time known value that is implicitly castable to a []const u8:

printf.zig

const warn = @import("std").debug.warn;

const a_number: i32 = 1234;
const a_string = "foobar";
const fmt = "here is a string: '{}' here is a number: {}\n";

pub fn main() void {
    warn(fmt, a_string, a_number);
}

$ zig build-exe printf.zig
$ ./printf
here is a string: 'foobar' here is a number: 1234

This works fine.

Zig does not special case string formatting in the compiler and instead exposes enough power to accomplish this task in userland. It does so without introducing another language on top of Zig, such as a macro language or a preprocessor language. It's Zig all the way down.

See also:

inline while
inline for

Assembly

TODO: example of inline assembly

TODO: example of module level assembly

TODO: example of using inline assembly return value

TODO: example of using inline assembly assigning values to variables

Atomics

TODO: @fence()

TODO: @atomic rmw

TODO: builtin atomic memory ordering enum

Coroutines

A coroutine is a generalization of a function.

When you call a function, it creates a stack frame, and then the function runs until it reaches a return statement, and then the stack frame is destroyed. At the callsite, the next line of code does not run until the function returns.

A coroutine is like a function, but it can be suspended and resumed any number of times, and then it must be explicitly destroyed. When a coroutine suspends, it returns to the resumer.

Minimal Coroutine Example

Declare a coroutine with the async keyword. The expression in angle brackets must evaluate to a struct which has these fields:

allocFn: fn (self: *Allocator, byte_count: usize, alignment: u29) Error![]u8 - where Error can be any error set.
freeFn: fn (self: *Allocator, old_mem: []u8) void

You may notice that this corresponds to the std.mem.Allocator interface. This makes it convenient to integrate with existing allocators. Note, however, that the language feature does not depend on the standard library, and any struct which has these fields is allowed.

Omitting the angle bracket expression when defining an async function makes the function generic. Zig will infer the allocator type when the async function is called.

Call a coroutine with the async keyword. Here, the expression in angle brackets is a pointer to the allocator struct that the coroutine expects.

The result of an async function call is a promise->T type, where T is the return type of the async function. Once a promise has been created, it must be consumed, either with cancel or await:

Async functions start executing when created, so in the following example, the entire async function completes before it is canceled:

test.zig

const std = @import("std");
const assert = std.debug.assert;

var x: i32 = 1;

test "create a coroutine and cancel it" {
    const p = try async<std.debug.global_allocator> simpleAsyncFn();
    comptime assert(@typeOf(p) == promise->void);
    cancel p;
    assert(x == 2);
}
async<*std.mem.Allocator> fn simpleAsyncFn() void {
    x += 1;
}

$ zig test test.zig
Test 1/1 create a coroutine and cancel it...OK
All tests passed.

Suspend and Resume

At any point, an async function may suspend itself. This causes control flow to return to the caller or resumer. The following code demonstrates where control flow goes:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "coroutine suspend, resume, cancel" {
    seq('a');
    const p = try async<std.debug.global_allocator> testAsyncSeq();
    seq('c');
    resume p;
    seq('f');
    cancel p;
    seq('g');

    assert(std.mem.eql(u8, points, "abcdefg"));
}
async fn testAsyncSeq() void {
    defer seq('e');

    seq('b');
    suspend;
    seq('d');
}
var points = []u8{0} ** "abcdefg".len;
var index: usize = 0;

fn seq(c: u8) void {
    points[index] = c;
    index += 1;
}

$ zig test test.zig
Test 1/1 coroutine suspend, resume, cancel...OK
All tests passed.

When an async function suspends itself, it must be sure that it will be resumed or canceled somehow, for example by registering its promise handle in an event loop. Use a suspend capture block to gain access to the promise:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "coroutine suspend with block" {
    const p = try async<std.debug.global_allocator> testSuspendBlock();
    std.debug.assert(!result);
    resume a_promise;
    std.debug.assert(result);
    cancel p;
}

var a_promise: promise = undefined;
var result = false;
async fn testSuspendBlock() void {
    suspend {
        comptime assert(@typeOf(@handle()) == promise->void);
        a_promise = @handle();
    }
    result = true;
}

$ zig test test.zig
Test 1/1 coroutine suspend with block...OK
All tests passed.

Every suspend point in an async function represents a point at which the coroutine could be destroyed. If that happens, defer expressions that are in scope are run, as well as errdefer expressions.

Await counts as a suspend point.

Resuming from Suspend Blocks

Upon entering a suspend block, the coroutine is already considered suspended, and can be resumed. For example, if you started another kernel thread, and had that thread call resume on the promise handle provided by the suspend block, the new thread would begin executing after the suspend block, while the old thread continued executing the suspend block.

However, the coroutine can be directly resumed from the suspend block, in which case it never returns to its resumer and continues executing.

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "resume from suspend" {
    var buf: [500]u8 = undefined;
    var a = &std.heap.FixedBufferAllocator.init(buf[0..]).allocator;
    var my_result: i32 = 1;
    const p = try async<a> testResumeFromSuspend(&my_result);
    cancel p;
    std.debug.assert(my_result == 2);
}
async fn testResumeFromSuspend(my_result: *i32) void {
    suspend {
        resume @handle();
    }
    my_result.* += 1;
    suspend;
    my_result.* += 1;
}

$ zig test test.zig
Test 1/1 resume from suspend...OK
All tests passed.

This is guaranteed to be a tail call, and therefore will not cause a new stack frame.

Await

The await keyword is used to coordinate with an async function's return statement.

await is valid only in an async function, and it takes as an operand a promise handle. If the async function associated with the promise handle has already returned, then await destroys the target async function, and gives the return value. Otherwise, await suspends the current async function, registering its promise handle with the target coroutine. It becomes the target coroutine's responsibility to have ensured that it will be resumed or destroyed. When the target coroutine reaches its return statement, it gives the return value to the awaiter, destroys itself, and then resumes the awaiter.

A promise handle must be consumed exactly once after it is created, either by cancel or await.

await counts as a suspend point, and therefore at every await, a coroutine can be potentially destroyed, which would run defer and errdefer expressions.

test.zig

const std = @import("std");
const assert = std.debug.assert;

var a_promise: promise = undefined;
var final_result: i32 = 0;

test "coroutine await" {
    seq('a');
    const p = async<std.debug.global_allocator> amain() catch unreachable;
    seq('f');
    resume a_promise;
    seq('i');
    assert(final_result == 1234);
    assert(std.mem.eql(u8, seq_points, "abcdefghi"));
}
async fn amain() void {
    seq('b');
    const p = async another() catch unreachable;
    seq('e');
    final_result = await p;
    seq('h');
}
async fn another() i32 {
    seq('c');
    suspend {
        seq('d');
        a_promise = @handle();
    }
    seq('g');
    return 1234;
}

var seq_points = []u8{0} ** "abcdefghi".len;
var seq_index: usize = 0;

fn seq(c: u8) void {
    seq_points[seq_index] = c;
    seq_index += 1;
}

$ zig test test.zig
Test 1/1 coroutine await...OK
All tests passed.

In general, suspend is lower level than await. Most application code will use only async and await, but event loop implementations will make use of suspend internally.

Open Issues

There are a few issues with coroutines that are considered unresolved. Best be aware of them, as the situation is likely to change before 1.0.0:

Async functions have optimizations disabled - even in release modes - due to an LLVM bug.
There are some situations where we can know statically that there will not be memory allocation failure, but Zig still forces us to handle it. TODO file an issue for this and link it here.
Zig does not take advantage of LLVM's allocation elision optimization for coroutines. It crashed LLVM when I tried to do it the first time. This is related to the other 2 bullet points here. See #802.

Builtin Functions

Builtin functions are provided by the compiler and are prefixed with @. The comptime keyword on a parameter means that the parameter must be known at compile time.

@addWithOverflow

@addWithOverflow(comptime T: type, a: T, b: T, result: *T) bool

Performs result.* = a + b. If overflow or underflow occurs, stores the overflowed bits in result and returns true. If no overflow or underflow occurs, returns false.

@ArgType

@ArgType(comptime T: type, comptime n: usize) type

This builtin function takes a function type and returns the type of the parameter at index n.

T must be a function type.

Note: This function is deprecated. Use @typeInfo instead.

@atomicLoad

@atomicLoad(comptime T: type, ptr: *const T, comptime ordering: builtin.AtomicOrder) T

This builtin function atomically dereferences a pointer and returns the value.

T must be a pointer type, a bool, or an integer whose bit count meets these requirements:

At least 8
At most the same as usize
Power of 2

TODO right now bool is not accepted. Also I think we could make non powers of 2 work fine, maybe we can remove this restriction

@atomicRmw

@atomicRmw(comptime T: type, ptr: *T, comptime op: builtin.AtomicRmwOp, operand: T, comptime ordering: builtin.AtomicOrder) T

This builtin function atomically modifies memory and then returns the previous value.

T must be a pointer type, a bool, or an integer whose bit count meets these requirements:

At least 8
At most the same as usize
Power of 2

TODO right now bool is not accepted. Also I think we could make non powers of 2 work fine, maybe we can remove this restriction

@bitCast

@bitCast(comptime DestType: type, value: var) DestType

Converts a value of one type to another type.

Asserts that @sizeOf(@typeOf(value)) == @sizeOf(DestType).

Asserts that @typeId(DestType) != @import("builtin").TypeId.Pointer. Use @ptrCast or @intToPtr if you need this.

Can be used for these things for example:

Convert f32 to u32 bits
Convert i32 to u32 preserving twos complement

Works at compile-time if value is known at compile time. It's a compile error to bitcast a struct to a scalar type of the same size since structs have undefined layout. However if the struct is packed then it works.

@bitOffsetOf

@bitOffsetOf(comptime T: type, comptime field_name: [] const u8) comptime_int

Returns the bit offset of a field relative to its containing struct.

For non packed structs, this will always be divisible by 8. For packed structs, non-byte-aligned fields will share a byte offset, but they will have different bit offsets.

See also:

@byteOffsetOf

@breakpoint

@breakpoint()

This function inserts a platform-specific debug trap instruction which causes debuggers to break there.

This function is only valid within function scope.

@byteOffsetOf

@byteOffsetOf(comptime T: type, comptime field_name: [] const u8) comptime_int

Returns the byte offset of a field relative to its containing struct.

See also:

@bitOffsetOf

@alignCast

@alignCast(comptime alignment: u29, ptr: var) var

ptr can be *T, fn(), ?*T, ?fn(), or []T. It returns the same type as ptr except with the alignment adjusted to the new value.

A pointer alignment safety check is added to the generated code to make sure the pointer is aligned as promised.

@alignOf

@alignOf(comptime T: type) comptime_int

This function returns the number of bytes that this type should be aligned to for the current target to match the C ABI. When the child type of a pointer has this alignment, the alignment can be omitted from the type.

const assert = @import("std").debug.assert;
comptime {
    assert(*u32 == *align(@alignOf(u32)) u32);
}

The result is a target-specific compile time constant. It is guaranteed to be less than or equal to @sizeOf(T).

See also:

Alignment

@boolToInt

@boolToInt(value: bool) u1

Converts true to u1(1) and false to u1(0).

If the value is known at compile-time, the return type is comptime_int instead of u1.

@bytesToSlice

@bytesToSlice(comptime Element: type, bytes: []u8) []Element

Converts a slice of bytes or array of bytes into a slice of Element. The resulting slice has the same pointer properties as the parameter.

Attempting to convert a number of bytes with a length that does not evenly divide into a slice of elements results in safety-protected Undefined Behavior.

@cDefine

@cDefine(comptime name: []u8, value)

This function can only occur inside @cImport.

This appends #define $name $value to the @cImport temporary buffer.

To define without a value, like this:

#define _GNU_SOURCE

Use the void value, like this:

@cDefine("_GNU_SOURCE", {})

See also:

Import from C Header File
@cInclude
@cImport
@cUndef
void

@cImport

@cImport(expression) (namespace)

This function parses C code and imports the functions, types, variables, and compatible macro definitions into the result namespace.

expression is interpreted at compile time. The builtin functions @cInclude, @cDefine, and @cUndef work within this expression, appending to a temporary buffer which is then parsed as C code.

Usually you should only have one @cImport in your entire application, because it saves the compiler from invoking clang multiple times, and prevents inline functions from being duplicated.

Reasons for having multiple @cImport expressions would be:

To avoid a symbol collision, for example if foo.h and bar.h both #define CONNECTION_COUNT
To analyze the C code with different preprocessor defines

See also:

Import from C Header File
@cInclude
@cDefine
@cUndef

@cInclude

@cInclude(comptime path: []u8)

This function can only occur inside @cImport.

This appends #include <$path>\n to the c_import temporary buffer.

See also:

Import from C Header File
@cImport
@cDefine
@cUndef

@cUndef

@cUndef(comptime name: []u8)

This function can only occur inside @cImport.

This appends #undef $name to the @cImport temporary buffer.

See also:

Import from C Header File
@cImport
@cDefine
@cInclude

@clz

@clz(x: T) U

This function counts the number of leading zeroes in x which is an integer type T.

The return type U is an unsigned integer with the minimum number of bits that can represent the value T.bit_count.

If x is zero, @clz returns T.bit_count.

See also:

@ctz
@popCount

@cmpxchgStrong

@cmpxchgStrong(comptime T: type, ptr: *T, expected_value: T, new_value: T, success_order: AtomicOrder, fail_order: AtomicOrder) ?T

This function performs a strong atomic compare exchange operation. It's the equivalent of this code, except atomic:

fn cmpxchgStrongButNotAtomic(comptime T: type, ptr: *T, expected_value: T, new_value: T) ?T {
    const old_value = ptr.*;
    if (old_value == expected_value) {
        ptr.* = new_value;
        return null;
    } else {
        return old_value;
    }
}

If you are using cmpxchg in a loop, @cmpxchgWeak is the better choice, because it can be implemented more efficiently in machine instructions.

AtomicOrder can be found with @import("builtin").AtomicOrder.

@typeOf(ptr).alignment must be >= @sizeOf(T).

See also:

Compile Variables
cmpxchgWeak

@cmpxchgWeak

@cmpxchgWeak(comptime T: type, ptr: *T, expected_value: T, new_value: T, success_order: AtomicOrder, fail_order: AtomicOrder) ?T

This function performs a weak atomic compare exchange operation. It's the equivalent of this code, except atomic:

fn cmpxchgWeakButNotAtomic(comptime T: type, ptr: *T, expected_value: T, new_value: T) ?T {
    const old_value = ptr.*;
    if (old_value == expected_value and usuallyTrueButSometimesFalse()) {
        ptr.* = new_value;
        return null;
    } else {
        return old_value;
    }
}

If you are using cmpxchg in a loop, the sporadic failure will be no problem, and cmpxchgWeak is the better choice, because it can be implemented more efficiently in machine instructions. However if you need a stronger guarantee, use @cmpxchgStrong.

AtomicOrder can be found with @import("builtin").AtomicOrder.

@typeOf(ptr).alignment must be >= @sizeOf(T).

See also:

Compile Variables
cmpxchgStrong

@compileError

@compileError(comptime msg: []u8)

This function, when semantically analyzed, causes a compile error with the message msg.

There are several ways that code avoids being semantically checked, such as using if or switch with compile time constants, and comptime functions.

@compileLog

@compileLog(args: ...)

This function prints the arguments passed to it at compile-time.

To prevent accidentally leaving compile log statements in a codebase, a compilation error is added to the build, pointing to the compile log statement. This error prevents code from being generated, but does not otherwise interfere with analysis.

This function can be used to do "printf debugging" on compile-time executing code.

test.zig

const warn = @import("std").debug.warn;

const num1 = blk: {
    var val1: i32 = 99;
    @compileLog("comptime val1 = ", val1); 
    val1 = val1 + 1;
    break :blk val1;
};

test "main" {
    @compileLog("comptime in main"); 

    warn("Runtime in main, num1 = {}.\n", num1);
}

$ zig test test.zig
| "comptime in main"
| "comptime val1 = ", 99
/home/andy/dev/zig/docgen_tmp/test.zig:11:5: error: found compile log statement
    @compileLog("comptime in main"); 
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:5:5: error: found compile log statement
    @compileLog("comptime val1 = ", val1); 
    ^

will ouput:

If all @compileLog calls are removed or not encountered by analysis, the program compiles successfully and the generated executable prints:

test.zig

const warn = @import("std").debug.warn;

const num1 = blk: {
    var val1: i32 = 99;
    val1 = val1 + 1;
    break :blk val1;
};

test "main" {
    warn("Runtime in main, num1 = {}.\n", num1);
}

$ zig test test.zig
Test 1/1 main...Runtime in main, num1 = 100.
OK
All tests passed.

@ctz

@ctz(x: T) U

This function counts the number of trailing zeroes in x which is an integer type T.

The return type U is an unsigned integer with the minimum number of bits that can represent the value T.bit_count.

If x is zero, @ctz returns T.bit_count.

See also:

@clz
@popCount

@divExact

@divExact(numerator: T, denominator: T) T

Exact division. Caller guarantees denominator != 0 and @divTrunc(numerator, denominator) * denominator == numerator.

@divExact(6, 3) == 2
@divExact(a, b) * b == a

For a function that returns a possible error code, use @import("std").math.divExact.

See also:

@divTrunc
@divFloor

@divFloor

@divFloor(numerator: T, denominator: T) T

Floored division. Rounds toward negative infinity. For unsigned integers it is the same as numerator / denominator. Caller guarantees denominator != 0 and !(@typeId(T) == builtin.TypeId.Int and T.is_signed and numerator == @minValue(T) and denominator == -1).

@divFloor(-5, 3) == -2
@divFloor(a, b) + @mod(a, b) == a

For a function that returns a possible error code, use @import("std").math.divFloor.

See also:

@divTrunc
@divExact

@divTrunc

@divTrunc(numerator: T, denominator: T) T

Truncated division. Rounds toward zero. For unsigned integers it is the same as numerator / denominator. Caller guarantees denominator != 0 and !(@typeId(T) == builtin.TypeId.Int and T.is_signed and numerator == @minValue(T) and denominator == -1).

@divTrunc(-5, 3) == -1
@divTrunc(a, b) + @rem(a, b) == a

For a function that returns a possible error code, use @import("std").math.divTrunc.

See also:

@divFloor
@divExact

@embedFile

@embedFile(comptime path: []const u8) [X]u8

This function returns a compile time constant fixed-size array with length equal to the byte count of the file given by path. The contents of the array are the contents of the file.

path is absolute or relative to the current file, just like @import.

See also:

@import

@enumToInt

@enumToInt(enum_value: var) var

Converts an enumeration value into its integer tag type.

If the enum has only 1 possible value, the resut is a comptime_int known at comptime.

See also:

@intToEnum

@errSetCast

@errSetCast(comptime T: DestType, value: var) DestType

Converts an error value from one error set to another error set. Attempting to convert an error which is not in the destination error set results in safety-protected Undefined Behavior.

@errorName

@errorName(err: error) []u8

This function returns the string representation of an error. If an error declaration is:

error OutOfMem

Then the string representation is "OutOfMem".

If there are no calls to @errorName in an entire application, or all calls have a compile-time known value for err, then no error name table will be generated.

@errorReturnTrace

@errorReturnTrace() ?*builtin.StackTrace

If the binary is built with error return tracing, and this function is invoked in a function that calls a function with an error or error union return type, returns a stack trace object. Otherwise returns `null`.

@errorToInt

@errorToInt(err: var) @IntType(false, @sizeOf(error) * 8)

Supports the following types:

error unions
E!void

Converts an error to the integer representation of an error.

It is generally recommended to avoid this cast, as the integer representation of an error is not stable across source code changes.

See also:

@intToError

@export

@export(comptime name: []const u8, target: var, linkage: builtin.GlobalLinkage) []const u8

Creates a symbol in the output object file.

@fence

@fence(order: AtomicOrder)

The fence function is used to introduce happens-before edges between operations.

AtomicOrder can be found with @import("builtin").AtomicOrder.

See also:

Compile Variables

@field

@field(lhs: var, comptime field_name: []const u8) (field)

Preforms field access equivalent to lhs.field_name, except instead of the field "field_name", it accesses the field named by the string value of field_name.

@fieldParentPtr

@fieldParentPtr(comptime ParentType: type, comptime field_name: []const u8,
    field_ptr: *T) *ParentType

Given a pointer to a field, returns the base pointer of a struct.

@floatCast

@floatCast(comptime DestType: type, value: var) DestType

Convert from one float type to another. This cast is safe, but may cause the numeric value to lose precision.

@floatToInt

@floatToInt(comptime DestType: type, float: var) DestType

Converts the integer part of a floating point number to the destination type.

If the integer part of the floating point number cannot fit in the destination type, it invokes safety-checked Undefined Behavior.

See also:

@intToFloat

@frameAddress

@frameAddress()

This function returns the base pointer of the current stack frame.

The implications of this are target specific and not consistent across all platforms. The frame address may not be available in release mode due to aggressive optimizations.

This function is only valid within function scope.

@handle

@handle()

This function returns a promise->T type, where T is the return type of the async function in scope.

This function is only valid within an async function scope.

@import

@import(comptime path: []u8) (namespace)

This function finds a zig file corresponding to path and imports all the public top level declarations into the resulting namespace.

path can be a relative or absolute path, or it can be the name of a package. If it is a relative path, it is relative to the file that contains the @import function call.

The following packages are always available:

@import("std") - Zig Standard Library
@import("builtin") - Compiler-provided types and variables

See also:

Compile Variables
@embedFile

@inlineCall

@inlineCall(function: X, args: ...) Y

This calls a function, in the same way that invoking an expression with parentheses does:

test.zig

const assert = @import("std").debug.assert;

test "inline function call" {
    assert(@inlineCall(add, 3, 9) == 12);
}

fn add(a: i32, b: i32) i32 { return a + b; }

$ zig test test.zig
Test 1/1 inline function call...OK
All tests passed.

Unlike a normal function call, however, @inlineCall guarantees that the call will be inlined. If the call cannot be inlined, a compile error is emitted.

See also:

@noInlineCall

@intCast

@intCast(comptime DestType: type, int: var) DestType

Converts an integer to another integer while keeping the same numerical value. Attempting to convert a number which is out of range of the destination type results in safety-protected Undefined Behavior.

@intToEnum

@intToEnum(comptime DestType: type, int_value: @TagType(DestType)) DestType

Converts an integer into an enum value.

Attempting to convert an integer which represents no value in the chosen enum type invokes safety-checked Undefined Behavior.

See also:

@enumToInt

@intToError

@intToError(value: @IntType(false, @sizeOf(error) * 8)) error

Converts from the integer representation of an error into the global error set type.

It is generally recommended to avoid this cast, as the integer representation of an error is not stable across source code changes.

Attempting to convert an integer that does not correspond to any error results in safety-protected Undefined Behavior.

See also:

@errorToInt

@intToFloat

@intToFloat(comptime DestType: type, int: var) DestType

Converts an integer to the closest floating point representation. To convert the other way, use @floatToInt. This cast is always safe.

@intToPtr

@intToPtr(comptime DestType: type, int: usize) DestType

Converts an integer to a pointer. To convert the other way, use @ptrToInt.

@IntType

@IntType(comptime is_signed: bool, comptime bit_count: u32) type

This function returns an integer type with the given signness and bit count.

@maxValue

@maxValue(comptime T: type) comptime_int

This function returns the maximum value of the integer type T.

The result is a compile time constant.

@memberCount

@memberCount(comptime T: type) comptime_int

This function returns the number of members in a struct, enum, or union type.

The result is a compile time constant.

It does not include functions, variables, or constants.

@memberName

@memberName(comptime T: type, comptime index: usize) [N]u8

Returns the field name of a struct, union, or enum.

The result is a compile time constant.

It does not include functions, variables, or constants.

@memberType

@memberType(comptime T: type, comptime index: usize) type

Returns the field type of a struct or union.

@memcpy

@memcpy(noalias dest: [*]u8, noalias source: [*]const u8, byte_count: usize)

This function copies bytes from one region of memory to another. dest and source are both pointers and must not overlap.

This function is a low level intrinsic with no safety mechanisms. Most code should not use this function, instead using something like this:

for (source[0...byte_count]) |b, i| dest[i] = b;

The optimizer is intelligent enough to turn the above snippet into a memcpy.

There is also a standard library function for this:

const mem = @import("std").mem;
mem.copy(u8, dest[0...byte_count], source[0...byte_count]);

@memset

@memset(dest: [*]u8, c: u8, byte_count: usize)

This function sets a region of memory to c. dest is a pointer.

This function is a low level intrinsic with no safety mechanisms. Most code should not use this function, instead using something like this:

for (dest[0...byte_count]) |*b| b.* = c;

The optimizer is intelligent enough to turn the above snippet into a memset.

There is also a standard library function for this:

const mem = @import("std").mem;
mem.set(u8, dest, c);

@minValue

@minValue(comptime T: type) comptime_int

This function returns the minimum value of the integer type T.

The result is a compile time constant.

@mod

@mod(numerator: T, denominator: T) T

Modulus division. For unsigned integers this is the same as numerator % denominator. Caller guarantees denominator > 0.

@mod(-5, 3) == 1
@divFloor(a, b) + @mod(a, b) == a

For a function that returns an error code, see @import("std").math.mod.

See also:

@rem

@mulWithOverflow

@mulWithOverflow(comptime T: type, a: T, b: T, result: *T) bool

Performs result.* = a * b. If overflow or underflow occurs, stores the overflowed bits in result and returns true. If no overflow or underflow occurs, returns false.

@newStackCall

@newStackCall(new_stack: []u8, function: var, args: ...) var

This calls a function, in the same way that invoking an expression with parentheses does. However, instead of using the same stack as the caller, the function uses the stack provided in the new_stack parameter.

test.zig

const std = @import("std");
const assert = std.debug.assert;

var new_stack_bytes: [1024]u8 = undefined;

test "calling a function with a new stack" {
    const arg = 1234;

    const a = @newStackCall(new_stack_bytes[0..512], targetFunction, arg);
    const b = @newStackCall(new_stack_bytes[512..], targetFunction, arg);
    _ = targetFunction(arg);

    assert(arg == 1234);
    assert(a < b);
}

fn targetFunction(x: i32) usize {
    assert(x == 1234);

    var local_variable: i32 = 42;
    const ptr = &local_variable;
    ptr.* += 1;

    assert(local_variable == 43);
    return @ptrToInt(ptr);
}

$ zig test test.zig
Test 1/1 calling a function with a new stack...OK
All tests passed.

@noInlineCall

@noInlineCall(function: var, args: ...) var

This calls a function, in the same way that invoking an expression with parentheses does:

test.zig

const assert = @import("std").debug.assert;

test "noinline function call" {
    assert(@noInlineCall(add, 3, 9) == 12);
}

fn add(a: i32, b: i32) i32 {
    return a + b;
}

$ zig test test.zig
Test 1/1 noinline function call...OK
All tests passed.

Unlike a normal function call, however, @noInlineCall guarantees that the call will not be inlined. If the call must be inlined, a compile error is emitted.

See also:

@inlineCall

@OpaqueType

@OpaqueType() type

Creates a new type with an unknown (but non-zero) size and alignment.

This is typically used for type safety when interacting with C code that does not expose struct details. Example:

test.zig

const Derp = @OpaqueType();
const Wat = @OpaqueType();

extern fn bar(d: *Derp) void;
export fn foo(w: *Wat) void {
    bar(w);
}

test "call foo" {
    foo(undefined);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:6:9: error: expected type '*Derp', found '*Wat'
    bar(w);
        ^
/home/andy/dev/zig/docgen_tmp/test.zig:6:9: note: pointer type child 'Wat' cannot cast into pointer type child 'Derp'
    bar(w);
        ^

@panic

@panic(message: []const u8) noreturn

Invokes the panic handler function. By default the panic handler function calls the public panic function exposed in the root source file, or if there is not one specified, invokes the one provided in std/special/panic.zig.

Generally it is better to use @import("std").debug.panic. However, @panic can be useful for 2 scenarios:

From library code, calling the programmer's panic function if they exposed one in the root source file.
When mixing C and Zig code, calling the canonical panic implementation across multiple .o files.

See also:

Root Source File

@popCount

@popCount(integer: var) var

Counts the number of bits set in an integer.

If integer is known at comptime, the return type is comptime_int. Otherwise, the return type is an unsigned integer with the minimum number of bits that can represent the bit count of the integer type.

See also:

@ctz
@clz

@ptrCast

@ptrCast(comptime DestType: type, value: var) DestType

Converts a pointer of one type to a pointer of another type.

@ptrToInt

@ptrToInt(value: var) usize

Converts value to a usize which is the address of the pointer. value can be one of these types:

*T
?*T
fn()
?fn()

To convert the other way, use @intToPtr

@rem

@rem(numerator: T, denominator: T) T

Remainder division. For unsigned integers this is the same as numerator % denominator. Caller guarantees denominator > 0.

@rem(-5, 3) == -2
@divTrunc(a, b) + @rem(a, b) == a

For a function that returns an error code, see @import("std").math.rem.

See also:

@mod

@returnAddress

@returnAddress()

This function returns a pointer to the return address of the current stack frame.

The implications of this are target specific and not consistent across all platforms.

This function is only valid within function scope.

@setAlignStack

@setAlignStack(comptime alignment: u29)

Ensures that a function will have a stack alignment of at least alignment bytes.

@setCold

@setCold(is_cold: bool)

Tells the optimizer that a function is rarely called.

@setRuntimeSafety

@setRuntimeSafety(safety_on: bool)

Sets whether runtime safety checks are on for the scope that contains the function call.

@setEvalBranchQuota

@setEvalBranchQuota(new_quota: usize)

Changes the maximum number of backwards branches that compile-time code execution can use before giving up and making a compile error.

If the new_quota is smaller than the default quota (1000) or a previously explicitly set quota, it is ignored.

Example:

test.zig

test "foo" {
    comptime {
        var i = 0;
        while (i < 1001) : (i += 1) {}
    }
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:4:9: error: evaluation exceeded 1000 backwards branches
        while (i < 1001) : (i += 1) {}
        ^

Now we use @setEvalBranchQuota:

test.zig

test "foo" {
    comptime {
        @setEvalBranchQuota(1001);
        var i = 0;
        while (i < 1001) : (i += 1) {}
    }
}

$ zig test test.zig
Test 1/1 foo...OK
All tests passed.

See also:

comptime

@setFloatMode

@setFloatMode(mode: @import("builtin").FloatMode)

Sets the floating point mode of the current scope. Possible values are:

pub const FloatMode = enum {
    Strict,
    Optimized,
};

Strict (default) - Floating point operations follow strict IEEE compliance.
Optimized - Floating point operations may do all of the following:
- Assume the arguments and result are not NaN. Optimizations are required to retain defined behavior over NaNs, but the value of the result is undefined.
- Assume the arguments and result are not +/-Inf. Optimizations are required to retain defined behavior over +/-Inf, but the value of the result is undefined.
- Treat the sign of a zero argument or result as insignificant.
- Use the reciprocal of an argument rather than perform division.
- Perform floating-point contraction (e.g. fusing a multiply followed by an addition into a fused multiply-and-add).
- Perform algebraically equivalent transformations that may change results in floating point (e.g. reassociate).
This is equivalent to -ffast-math in GCC.

The floating point mode is inherited by child scopes, and can be overridden in any scope. You can set the floating point mode in a struct or module scope by using a comptime block.

See also:

Floating Point Operations

@setGlobalLinkage

@setGlobalLinkage(global_variable_name, comptime linkage: GlobalLinkage)

GlobalLinkage can be found with @import("builtin").GlobalLinkage.

See also:

Compile Variables

@shlExact

@shlExact(value: T, shift_amt: Log2T) T

Performs the left shift operation (<<). Caller guarantees that the shift will not shift any 1 bits out.

The type of shift_amt is an unsigned integer with log2(T.bit_count) bits. This is because shift_amt >= T.bit_count is undefined behavior.

See also:

@shrExact
@shlWithOverflow

@shlWithOverflow

@shlWithOverflow(comptime T: type, a: T, shift_amt: Log2T, result: *T) bool

Performs result.* = a << b. If overflow or underflow occurs, stores the overflowed bits in result and returns true. If no overflow or underflow occurs, returns false.

The type of shift_amt is an unsigned integer with log2(T.bit_count) bits. This is because shift_amt >= T.bit_count is undefined behavior.

See also:

@shlExact
@shrExact

@shrExact

@shrExact(value: T, shift_amt: Log2T) T

Performs the right shift operation (>>). Caller guarantees that the shift will not shift any 1 bits out.

The type of shift_amt is an unsigned integer with log2(T.bit_count) bits. This is because shift_amt >= T.bit_count is undefined behavior.

See also:

@shlExact
@shlWithOverflow

@sizeOf

@sizeOf(comptime T: type) comptime_int

This function returns the number of bytes it takes to store T in memory.

The result is a target-specific compile time constant.

@sliceToBytes

@sliceToBytes(value: var) []u8

Converts a slice or array to a slice of u8. The resulting slice has the same pointer properties as the parameter.

@sqrt

@sqrt(comptime T: type, value: T) T

Performs the square root of a floating point number. Uses a dedicated hardware instruction when available. Currently only supports f32 and f64 at runtime. f128 at runtime is TODO.

This is a low-level intrinsic. Most code can use std.math.sqrt instead.

@subWithOverflow

@subWithOverflow(comptime T: type, a: T, b: T, result: *T) bool

Performs result.* = a - b. If overflow or underflow occurs, stores the overflowed bits in result and returns true. If no overflow or underflow occurs, returns false.

@tagName

@tagName(value: var) []const u8

Converts an enum value or union value to a slice of bytes representing the name.

@TagType

@TagType(T: type) type

For an enum, returns the integer type that is used to store the enumeration value.

For a union, returns the enum type that is used to store the tag value.

@This

@This() type

Returns the innermost struct or union that this function call is inside. This can be useful for an anonymous struct that needs to refer to itself:

test.zig

const std = @import("std");
const assert = std.debug.assert;

test "@This()" {
    var items = []i32{ 1, 2, 3, 4 };
    const list = List(i32){ .items = items[0..] };
    assert(list.length() == 4);
}

fn List(comptime T: type) type {
    return struct {
        const Self = @This();

        items: []T,

        fn length(self: Self) usize {
            return self.items.len;
        }
    };
}

$ zig test test.zig
Test 1/1 @This()...OK
All tests passed.

When @This() is used at global scope, it returns a reference to the current import. There is a proposal to remove the import type and use an empty struct type instead. See #1047 for details.

@truncate

@truncate(comptime T: type, integer) T

This function truncates bits from an integer type, resulting in a smaller integer type.

The following produces a crash in debug mode and undefined behavior in release mode:

const a: u16 = 0xabcd;
const b: u8 = u8(a);

However this is well defined and working code:

const a: u16 = 0xabcd;
const b: u8 = @truncate(u8, a);
// b is now 0xcd

This function always truncates the significant bits of the integer, regardless of endianness on the target platform.

@typeId

@typeId(comptime T: type) @import("builtin").TypeId

Returns which kind of type something is. Possible values:

pub const TypeId = enum {
    Type,
    Void,
    Bool,
    NoReturn,
    Int,
    Float,
    Pointer,
    Array,
    Struct,
    ComptimeFloat,
    ComptimeInt,
    Undefined,
    Null,
    Optional,
    ErrorUnion,
    Error,
    Enum,
    Union,
    Fn,
    Namespace,
    Block,
    BoundFn,
    ArgTuple,
    Opaque,
};

@typeInfo

@typeInfo(comptime T: type) @import("builtin").TypeInfo

Returns information on the type. Returns a value of the following union:

pub const TypeInfo = union(TypeId) {
    Type: void,
    Void: void,
    Bool: void,
    NoReturn: void,
    Int: Int,
    Float: Float,
    Pointer: Pointer,
    Array: Array,
    Struct: Struct,
    ComptimeFloat: void,
    ComptimeInt: void,
    Undefined: void,
    Null: void,
    Optional: Optional,
    ErrorUnion: ErrorUnion,
    ErrorSet: ErrorSet,
    Enum: Enum,
    Union: Union,
    Fn: Fn,
    Namespace: void,
    BoundFn: Fn,
    ArgTuple: void,
    Opaque: void,
    Promise: Promise,


    pub const Int = struct {
        is_signed: bool,
        bits: u8,
    };

    pub const Float = struct {
        bits: u8,
    };

    pub const Pointer = struct {
        size: Size,
        is_const: bool,
        is_volatile: bool,
        alignment: u32,
        child: type,

        pub const Size = enum {
            One,
            Many,
            Slice,
        };
    };

    pub const Array = struct {
        len: usize,
        child: type,
    };

    pub const ContainerLayout = enum {
        Auto,
        Extern,
        Packed,
    };

    pub const StructField = struct {
        name: []const u8,
        offset: ?usize,
        field_type: type,
    };

    pub const Struct = struct {
        layout: ContainerLayout,
        fields: []StructField,
        defs: []Definition,
    };

    pub const Optional = struct {
        child: type,
    };

    pub const ErrorUnion = struct {
        error_set: type,
        payload: type,
    };

    pub const Error = struct {
        name: []const u8,
        value: usize,
    };

    pub const ErrorSet = struct {
        errors: []Error,
    };

    pub const EnumField = struct {
        name: []const u8,
        value: usize,
    };

    pub const Enum = struct {
        layout: ContainerLayout,
        tag_type: type,
        fields: []EnumField,
        defs: []Definition,
    };

    pub const UnionField = struct {
        name: []const u8,
        enum_field: ?EnumField,
        field_type: type,
    };

    pub const Union = struct {
        layout: ContainerLayout,
        tag_type: ?type,
        fields: []UnionField,
        defs: []Definition,
    };

    pub const CallingConvention = enum {
        Unspecified,
        C,
        Cold,
        Naked,
        Stdcall,
        Async,
    };

    pub const FnArg = struct {
        is_generic: bool,
        is_noalias: bool,
        arg_type: ?type,
    };

    pub const Fn = struct {
        calling_convention: CallingConvention,
        is_generic: bool,
        is_var_args: bool,
        return_type: ?type,
        async_allocator_type: ?type,
        args: []FnArg,
    };

    pub const Promise = struct {
        child: ?type,
    };

    pub const Definition = struct {
        name: []const u8,
        is_pub: bool,
        data: Data,

        pub const Data = union(enum) {
            Type: type,
            Var: type,
            Fn: FnDef,

            pub const FnDef = struct {
                fn_type: type,
                inline_type: Inline,
                calling_convention: CallingConvention,
                is_var_args: bool,
                is_extern: bool,
                is_export: bool,
                lib_name: ?[]const u8,
                return_type: type,
                arg_names: [][] const u8,

                pub const Inline = enum {
                    Auto,
                    Always,
                    Never,
                };
            };
        };
    };
};

@typeName

@typeName(T: type) []u8

This function returns the string representation of a type.

@typeOf

@typeOf(expression) type

This function returns a compile-time constant, which is the type of the expression passed as an argument. The expression is evaluated.

Build Mode

Zig has four build modes:

Debug (default)
ReleaseFast
ReleaseSafe
ReleaseSmall

To add standard build options to a build.zig file:

const Builder = @import("std").build.Builder;

pub fn build(b: *Builder) void {
    const exe = b.addExecutable("example", "example.zig");
    exe.setBuildMode(b.standardReleaseOptions());
    b.default_step.dependOn(&exe.step);
}

This causes these options to be available:

  -Drelease-safe=[bool] optimizations on and safety on
  -Drelease-fast=[bool] optimizations on and safety off
  -Drelease-small=[bool] size optimizations on and safety off

Debug

$ zig build-exe example.zig

Fast compilation speed
Safety checks enabled
Slow runtime performance
Large binary size
No reproducible build requirement

ReleaseFast

$ zig build-exe example.zig --release-fast

Fast runtime performance
Safety checks disabled
Slow compilation speed
Large binary size
Reproducible build

ReleaseSafe

$ zig build-exe example.zig --release-safe

Medium runtime performance
Safety checks enabled
Slow compilation speed
Large binary size
Reproducible build

ReleaseSmall

$ zig build-exe example.zig --release-small

Medium runtime performance
Safety checks disabled
Slow compilation speed
Small binary size
Reproducible build

See also:

Compile Variables
Zig Build System
Undefined Behavior

Undefined Behavior

Zig has many instances of undefined behavior. If undefined behavior is detected at compile-time, Zig emits a compile error and refuses to continue. Most undefined behavior that cannot be detected at compile-time can be detected at runtime. In these cases, Zig has safety checks. Safety checks can be disabled on a per-block basis with setRuntimeSafety. The ReleaseFast build mode disables all safety checks in order to facilitate optimizations.

When a safety check fails, Zig crashes with a stack trace, like this:

test.zig

test "safety check" {
    unreachable;
}

$ zig test test.zig
Test 1/1 safety check...reached unreachable code
/home/andy/dev/zig/docgen_tmp/test.zig:2:5: 0x205054 in ??? (test)
    unreachable;
    ^
/home/andy/dev/zig/build/lib/zig/std/special/test_runner.zig:13:25: 0x22294a in ??? (test)
        if (test_fn.func()) |_| {
                        ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:96:22: 0x2226fb in ??? (test)
            root.main() catch |err| {
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222675 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224d8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222390 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Tests failed. Use the following command to reproduce the failure:
/home/andy/dev/zig/docgen_tmp/test

Reaching Unreachable Code

At compile-time:

test.zig

comptime {
    assert(false);
}
fn assert(ok: bool) void {
    if (!ok) unreachable; // assertion failure
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:5:14: error: unable to evaluate constant expression
    if (!ok) unreachable; // assertion failure
             ^
/home/andy/dev/zig/docgen_tmp/test.zig:2:11: note: called from here
    assert(false);
          ^
/home/andy/dev/zig/docgen_tmp/test.zig:1:10: note: called from here
comptime {
         ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    std.debug.assert(false);
}

$ zig build-exe test.zig
$ ./test
reached unreachable code
/home/andy/dev/zig/build/lib/zig/std/debug/index.zig:120:13: 0x205029 in ??? (test)
            unreachable; // assertion failure
            ^
/home/andy/dev/zig/docgen_tmp/test.zig:4:21: 0x2226cb in ??? (test)
    std.debug.assert(false);
                    ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Index out of Bounds

At compile-time:

test.zig

comptime {
    const array = "hello";
    const garbage = array[5];
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:26: error: index 5 outside array of size 5
    const garbage = array[5];
                         ^

At runtime:

test.zig

pub fn main() void {
    var x = foo("hello");
}

fn foo(x: []const u8) u8 {
    return x[5];
}

$ zig build-exe test.zig
$ ./test
index out of bounds
/home/andy/dev/zig/docgen_tmp/test.zig:6:13: 0x222709 in ??? (test)
    return x[5];
            ^
/home/andy/dev/zig/docgen_tmp/test.zig:2:16: 0x2226d4 in ??? (test)
    var x = foo("hello");
               ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Cast Negative Number to Unsigned Integer

At compile-time:

test.zig

comptime {
    const value: i32 = -1;
    const unsigned = @intCast(u32, value);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:22: error: attempt to cast negative value to unsigned integer
    const unsigned = @intCast(u32, value);
                     ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var value: i32 = -1;
    var unsigned = @intCast(u32, value);
    std.debug.warn("value: {}\n", unsigned);
}

$ zig build-exe test.zig
$ ./test
attempt to cast negative value to unsigned integer
/home/andy/dev/zig/docgen_tmp/test.zig:5:20: 0x2226fe in ??? (test)
    var unsigned = @intCast(u32, value);
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

To obtain the maximum value of an unsigned integer, use @maxValue.

Cast Truncates Data

At compile-time:

test.zig

comptime {
    const spartan_count: u16 = 300;
    const byte = @intCast(u8, spartan_count);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:18: error: cast from 'u16' to 'u8' truncates bits
    const byte = @intCast(u8, spartan_count);
                 ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var spartan_count: u16 = 300;
    const byte = @intCast(u8, spartan_count);
    std.debug.warn("value: {}\n", byte);
}

$ zig build-exe test.zig
$ ./test
integer cast truncated bits
/home/andy/dev/zig/docgen_tmp/test.zig:5:18: 0x222707 in ??? (test)
    const byte = @intCast(u8, spartan_count);
                 ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

To truncate bits, use @truncate.

Integer Overflow

Default Operations

The following operators can cause integer overflow:

+ (addition)
- (subtraction)
- (negation)
* (multiplication)
/ (division)
@divTrunc (division)
@divFloor (division)
@divExact (division)

Example with addition at compile-time:

test.zig

comptime {
    var byte: u8 = 255;
    byte += 1;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:10: error: operation caused overflow
    byte += 1;
         ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var byte: u8 = 255;
    byte += 1;
    std.debug.warn("value: {}\n", byte);
}

$ zig build-exe test.zig
$ ./test
integer overflow
/home/andy/dev/zig/docgen_tmp/test.zig:5:10: 0x2226ee in ??? (test)
    byte += 1;
         ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Standard Library Math Functions

These functions provided by the standard library return possible errors.

@import("std").math.add
@import("std").math.sub
@import("std").math.mul
@import("std").math.divTrunc
@import("std").math.divFloor
@import("std").math.divExact
@import("std").math.shl

Example of catching an overflow for addition:

test.zig

const math = @import("std").math;
const warn = @import("std").debug.warn;
pub fn main() !void {
    var byte: u8 = 255;

    byte = if (math.add(u8, byte, 1)) |result| result else |err| {
        warn("unable to add one: {}\n", @errorName(err));
        return err;
    };

    warn("result: {}\n", byte);
}

$ zig build-exe test.zig
$ ./test
unable to add one: Overflow
error: Overflow
/home/andy/dev/zig/build/lib/zig/std/math/index.zig:252:5: 0x2229fa in ??? (test)
    return if (@addWithOverflow(T, a, b, &answer)) error.Overflow else answer;
    ^
/home/andy/dev/zig/docgen_tmp/test.zig:8:9: 0x222858 in ??? (test)
        return err;
        ^

Builtin Overflow Functions

These builtins return a bool of whether or not overflow occurred, as well as returning the overflowed bits:

@addWithOverflow
@subWithOverflow
@mulWithOverflow
@shlWithOverflow

Example of @addWithOverflow:

test.zig

const warn = @import("std").debug.warn;
pub fn main() void {
    var byte: u8 = 255;

    var result: u8 = undefined;
    if (@addWithOverflow(u8, byte, 10, &result)) {
        warn("overflowed result: {}\n", result);
    } else {
        warn("result: {}\n", result);
    }
}

$ zig build-exe test.zig
$ ./test
overflowed result: 9

Wrapping Operations

These operations have guaranteed wraparound semantics.

+% (wraparound addition)
-% (wraparound subtraction)
-% (wraparound negation)
*% (wraparound multiplication)

test.zig

const assert = @import("std").debug.assert;

test "wraparound addition and subtraction" {
    const x: i32 = @maxValue(i32);
    const min_val = x +% 1;
    assert(min_val == @minValue(i32));
    const max_val = min_val -% 1;
    assert(max_val == @maxValue(i32));
}

$ zig test test.zig
Test 1/1 wraparound addition and subtraction...OK
All tests passed.

Exact Left Shift Overflow

At compile-time:

test.zig

comptime {
    const x = @shlExact(u8(0b01010101), 2);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:15: error: operation caused overflow
    const x = @shlExact(u8(0b01010101), 2);
              ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var x: u8 = 0b01010101;
    var y = @shlExact(x, 2);
    std.debug.warn("value: {}\n", y);
}

$ zig build-exe test.zig
$ ./test
left shift overflowed bits
/home/andy/dev/zig/docgen_tmp/test.zig:5:13: 0x222705 in ??? (test)
    var y = @shlExact(x, 2);
            ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Exact Right Shift Overflow

At compile-time:

test.zig

comptime {
    const x = @shrExact(u8(0b10101010), 2);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:15: error: exact shift shifted out 1 bits
    const x = @shrExact(u8(0b10101010), 2);
              ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var x: u8 = 0b10101010;
    var y = @shrExact(x, 2);
    std.debug.warn("value: {}\n", y);
}

$ zig build-exe test.zig
$ ./test
right shift overflowed bits
/home/andy/dev/zig/docgen_tmp/test.zig:5:13: 0x222705 in ??? (test)
    var y = @shrExact(x, 2);
            ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Division by Zero

At compile-time:

test.zig

comptime {
    const a: i32 = 1;
    const b: i32 = 0;
    const c = a / b;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:4:17: error: division by zero
    const c = a / b;
                ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var a: u32 = 1;
    var b: u32 = 0;
    var c = a / b;
    std.debug.warn("value: {}\n", c);
}

$ zig build-exe test.zig
$ ./test
division by zero
/home/andy/dev/zig/docgen_tmp/test.zig:6:15: 0x222712 in ??? (test)
    var c = a / b;
              ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Remainder Division by Zero

At compile-time:

test.zig

comptime {
    const a: i32 = 10;
    const b: i32 = 0;
    const c = a % b;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:4:17: error: division by zero
    const c = a % b;
                ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var a: u32 = 10;
    var b: u32 = 0;
    var c = a % b;
    std.debug.warn("value: {}\n", c);
}

$ zig build-exe test.zig
$ ./test
remainder division by zero or negative value
/home/andy/dev/zig/docgen_tmp/test.zig:6:15: 0x222712 in ??? (test)
    var c = a % b;
              ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Exact Division Remainder

At compile-time:

test.zig

comptime {
    const a: u32 = 10;
    const b: u32 = 3;
    const c = @divExact(a, b);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:4:15: error: exact division had a remainder
    const c = @divExact(a, b);
              ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var a: u32 = 10;
    var b: u32 = 3;
    var c = @divExact(a, b);
    std.debug.warn("value: {}\n", c);
}

$ zig build-exe test.zig
$ ./test
exact division produced remainder
/home/andy/dev/zig/docgen_tmp/test.zig:6:13: 0x222733 in ??? (test)
    var c = @divExact(a, b);
            ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Slice Widen Remainder

At compile-time:

test.zig

comptime {
    var bytes = [5]u8{ 1, 2, 3, 4, 5 };
    var slice = @bytesToSlice(u32, bytes);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:17: error: unable to convert [5]u8 to []align(1) const u32: size mismatch
    var slice = @bytesToSlice(u32, bytes);
                ^
/home/andy/dev/zig/docgen_tmp/test.zig:3:31: note: u32 has size 4; remaining bytes: 1
    var slice = @bytesToSlice(u32, bytes);
                              ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var bytes = [5]u8{ 1, 2, 3, 4, 5 };
    var slice = @bytesToSlice(u32, bytes[0..]);
    std.debug.warn("value: {}\n", slice[0]);
}

$ zig build-exe test.zig
$ ./test
slice widening size mismatch
/home/andy/dev/zig/docgen_tmp/test.zig:5:17: 0x222756 in ??? (test)
    var slice = @bytesToSlice(u32, bytes[0..]);
                ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Attempt to Unwrap Null

At compile-time:

test.zig

comptime {
    const optional_number: ?i32 = null;
    const number = optional_number.?;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:35: error: unable to unwrap null
    const number = optional_number.?;
                                  ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var optional_number: ?i32 = null;
    var number = optional_number.?;
    std.debug.warn("value: {}\n", number);
}

$ zig build-exe test.zig
$ ./test
attempt to unwrap null
/home/andy/dev/zig/docgen_tmp/test.zig:5:33: 0x2226ff in ??? (test)
    var number = optional_number.?;
                                ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

One way to avoid this crash is to test for null instead of assuming non-null, with the if expression:

test.zig

const warn = @import("std").debug.warn;
pub fn main() void {
    const optional_number: ?i32 = null;

    if (optional_number) |number| {
        warn("got number: {}\n", number);
    } else {
        warn("it's null\n");
    }
}

$ zig build-exe test.zig
$ ./test
it's null

See also:

Optionals

Attempt to Unwrap Error

At compile-time:

test.zig

comptime {
    const number = getNumberOrFail() catch unreachable;
}

fn getNumberOrFail() !i32 {
    return error.UnableToReturnNumber;
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:2:38: error: caught unexpected error 'UnableToReturnNumber'
    const number = getNumberOrFail() catch unreachable;
                                     ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    const number = getNumberOrFail() catch unreachable;
    std.debug.warn("value: {}\n", number);
}

fn getNumberOrFail() !i32 {
    return error.UnableToReturnNumber;
}

$ zig build-exe test.zig
$ ./test
attempt to unwrap error: UnableToReturnNumber
/home/andy/dev/zig/docgen_tmp/test.zig:9:5: 0x22276b in ??? (test)
    return error.UnableToReturnNumber;
    ^
???:?:?: 0x21fa3e in ??? (???)


/home/andy/dev/zig/docgen_tmp/test.zig:4:38: 0x22272c in ??? (test)
    const number = getNumberOrFail() catch unreachable;
                                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

One way to avoid this crash is to test for an error instead of assuming a successful result, with the if expression:

test.zig

const warn = @import("std").debug.warn;

pub fn main() void {
    const result = getNumberOrFail();

    if (result) |number| {
        warn("got number: {}\n", number);
    } else |err| {
        warn("got error: {}\n", @errorName(err));
    }
}

fn getNumberOrFail() !i32 {
    return error.UnableToReturnNumber;
}

$ zig build-exe test.zig
$ ./test
got error: UnableToReturnNumber

See also:

Errors

Invalid Error Code

At compile-time:

test.zig

comptime {
    const err = error.AnError;
    const number = @errorToInt(err) + 10;
    const invalid_err = @intToError(number);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:4:25: error: integer value 11 represents no error
    const invalid_err = @intToError(number);
                        ^

At runtime:

test.zig

const std = @import("std");

pub fn main() void {
    var err = error.AnError;
    var number = @errorToInt(err) + 500;
    var invalid_err = @intToError(number);
    std.debug.warn("value: {}\n", number);
}

$ zig build-exe test.zig
$ ./test
invalid error code
/home/andy/dev/zig/docgen_tmp/test.zig:6:23: 0x22271a in ??? (test)
    var invalid_err = @intToError(number);
                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Invalid Enum Cast

At compile-time:

test.zig

const Foo = enum {
    A,
    B,
    C,
};
comptime {
    const a: u2 = 3;
    const b = @intToEnum(Foo, a);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:8:15: error: enum 'Foo' has no tag matching integer value 3
    const b = @intToEnum(Foo, a);
              ^
/home/andy/dev/zig/docgen_tmp/test.zig:1:13: note: 'Foo' declared here
const Foo = enum {
            ^

At runtime:

test.zig

const std = @import("std");

const Foo = enum {
    A,
    B,
    C,
};

pub fn main() void {
    var a: u2 = 3;
    var b = @intToEnum(Foo, a);
    std.debug.warn("value: {}\n", @tagName(b));
}

$ zig build-exe test.zig
$ ./test
invalid enum value
/home/andy/dev/zig/docgen_tmp/test.zig:11:13: 0x2226fd in ??? (test)
    var b = @intToEnum(Foo, a);
            ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Invalid Error Set Cast

At compile-time:

test.zig

const Set1 = error{
    A,
    B,
};
const Set2 = error{
    A,
    C,
};
comptime {
    _ = @errSetCast(Set2, Set1.B);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:10:9: error: error.B not a member of error set 'Set2'
    _ = @errSetCast(Set2, Set1.B);
        ^

At runtime:

test.zig

const std = @import("std");

const Set1 = error{
    A,
    B,
};
const Set2 = error{
    A,
    C,
};
pub fn main() void {
    foo(Set1.B);
}
fn foo(set1: Set1) void {
    const x = @errSetCast(Set2, set1);
    std.debug.warn("value: {}\n", x);
}

$ zig build-exe test.zig
$ ./test
invalid error code
/home/andy/dev/zig/docgen_tmp/test.zig:15:15: 0x222724 in ??? (test)
    const x = @errSetCast(Set2, set1);
              ^
/home/andy/dev/zig/docgen_tmp/test.zig:12:8: 0x2226ce in ??? (test)
    foo(Set1.B);
       ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x2226a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Incorrect Pointer Alignment

At compile-time:

test.zig

comptime {
    const ptr = @intToPtr(*i32, 0x1);
    const aligned = @alignCast(4, ptr);
}

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:35: error: pointer address 0x1 is not aligned to 4 bytes
    const aligned = @alignCast(4, ptr);
                                  ^

At runtime:

test.zig

pub fn main() !void {
    var array align(4) = []u32{ 0x11111111, 0x11111111 };
    const bytes = @sliceToBytes(array[0..]);
    if (foo(bytes) != 0x11111111) return error.Wrong;
}
fn foo(bytes: []u8) u32 {
    const slice4 = bytes[1..5];
    const int_slice = @bytesToSlice(u32, @alignCast(4, slice4));
    return int_slice[0];
}

$ zig build-exe test.zig
$ ./test
incorrect alignment
/home/andy/dev/zig/docgen_tmp/test.zig:8:56: 0x2229ff in ??? (test)
    const int_slice = @bytesToSlice(u32, @alignCast(4, slice4));
                                                       ^
/home/andy/dev/zig/docgen_tmp/test.zig:4:12: 0x22283e in ??? (test)
    if (foo(bytes) != 0x11111111) return error.Wrong;
           ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:96:22: 0x2226db in ??? (test)
            root.main() catch |err| {
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x222655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x2224b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x222370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

Wrong Union Field Access

At compile-time:

test.zig

comptime {
    var f = Foo{ .int = 42 };
    f.float = 12.34;
}

const Foo = union {
    float: f32,
    int: u32,
};

$ zig test test.zig
/home/andy/dev/zig/docgen_tmp/test.zig:3:6: error: accessing union field 'float' while field 'int' is set
    f.float = 12.34;
     ^

At runtime:

test.zig

const std = @import("std");

const Foo = union {
    float: f32,
    int: u32,
};

pub fn main() void {
    var f = Foo{ .int = 42 };
    bar(&f);
}

fn bar(f: *Foo) void {
    f.float = 12.34;
    std.debug.warn("value: {}\n", f.float);
}

$ zig build-exe test.zig
$ ./test
access of inactive union field
/home/andy/dev/zig/docgen_tmp/test.zig:14:6: 0x22a738 in ??? (test)
    f.float = 12.34;
     ^
/home/andy/dev/zig/docgen_tmp/test.zig:10:8: 0x22a6dc in ??? (test)
    bar(&f);
       ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:86:22: 0x22a6a9 in ??? (test)
            root.main();
                     ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:70:20: 0x22a655 in ??? (test)
    return callMain();
                   ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:64:39: 0x22a4b8 in ??? (test)
    std.os.posix.exit(callMainWithArgs(argc, argv, envp));
                                      ^
/home/andy/dev/zig/build/lib/zig/std/special/bootstrap.zig:37:5: 0x22a370 in ??? (test)
    @noInlineCall(posixCallMainAndExit);
    ^

This safety is not available for extern or packed unions.

To change the active field of a union, assign the entire union, like this:

test.zig

const std = @import("std");

const Foo = union {
    float: f32,
    int: u32,
};

pub fn main() void {
    var f = Foo{ .int = 42 };
    bar(&f);
}

fn bar(f: *Foo) void {
    f.* = Foo{ .float = 12.34 };
    std.debug.warn("value: {}\n", f.float);
}

$ zig build-exe test.zig
$ ./test
value: 1.23400001e+01

To change the active field of a union when a meaningful value for the field is not known, use undefined, like this:

test.zig

const std = @import("std");

const Foo = union {
    float: f32,
    int: u32,
};

pub fn main() void {
    var f = Foo{ .int = 42 };
    f = Foo{ .float = undefined };
    bar(&f);
    std.debug.warn("value: {}\n", f.float);
}

fn bar(f: *Foo) void {
    f.float = 12.34;
}

$ zig build-exe test.zig
$ ./test
value: 1.23400001e+01

Out of Bounds Float To Integer Cast

TODO

Memory

TODO: explain no default allocator in zig

TODO: show how to use the allocator interface

TODO: mention debug allocator

TODO: importance of checking for allocation failure

TODO: mention overcommit and the OOM Killer

TODO: mention recursion

See also:

Pointers

Compile Variables

Compile variables are accessible by importing the "builtin" package, which the compiler makes available to every Zig source file. It contains compile-time constants such as the current target, endianness, and release mode.

const builtin = @import("builtin");
const separator = if (builtin.os == builtin.Os.windows) '\\' else '/';

Example of what is imported with @import("builtin"):

pub const StackTrace = struct {
    index: usize,
    instruction_addresses: []usize,
};

pub const Os = enum {
    freestanding,
    ananas,
    cloudabi,
    dragonfly,
    freebsd,
    fuchsia,
    ios,
    kfreebsd,
    linux,
    lv2,
    macosx,
    netbsd,
    openbsd,
    solaris,
    windows,
    haiku,
    minix,
    rtems,
    nacl,
    cnk,
    aix,
    cuda,
    nvcl,
    amdhsa,
    ps4,
    elfiamcu,
    tvos,
    watchos,
    mesa3d,
    contiki,
    amdpal,
    zen,
};

pub const Arch = enum {
    armv8_3a,
    armv8_2a,
    armv8_1a,
    armv8,
    armv8r,
    armv8m_baseline,
    armv8m_mainline,
    armv7,
    armv7em,
    armv7m,
    armv7s,
    armv7k,
    armv7ve,
    armv6,
    armv6m,
    armv6k,
    armv6t2,
    armv5,
    armv5te,
    armv4t,
    armebv8_3a,
    armebv8_2a,
    armebv8_1a,
    armebv8,
    armebv8r,
    armebv8m_baseline,
    armebv8m_mainline,
    armebv7,
    armebv7em,
    armebv7m,
    armebv7s,
    armebv7k,
    armebv7ve,
    armebv6,
    armebv6m,
    armebv6k,
    armebv6t2,
    armebv5,
    armebv5te,
    armebv4t,
    aarch64,
    aarch64_be,
    arc,
    avr,
    bpfel,
    bpfeb,
    hexagon,
    mips,
    mipsel,
    mips64,
    mips64el,
    msp430,
    nios2,
    powerpc,
    powerpc64,
    powerpc64le,
    r600,
    amdgcn,
    riscv32,
    riscv64,
    sparc,
    sparcv9,
    sparcel,
    s390x,
    tce,
    tcele,
    thumb,
    thumbeb,
    i386,
    x86_64,
    xcore,
    nvptx,
    nvptx64,
    le32,
    le64,
    amdil,
    amdil64,
    hsail,
    hsail64,
    spir,
    spir64,
    kalimbav3,
    kalimbav4,
    kalimbav5,
    shave,
    lanai,
    wasm32,
    wasm64,
    renderscript32,
    renderscript64,
};

pub const Environ = enum {
    unknown,
    gnu,
    gnuabin32,
    gnuabi64,
    gnueabi,
    gnueabihf,
    gnux32,
    code16,
    eabi,
    eabihf,
    android,
    musl,
    musleabi,
    musleabihf,
    msvc,
    itanium,
    cygnus,
    coreclr,
    simulator,
};

pub const ObjectFormat = enum {
    unknown,
    coff,
    elf,
    macho,
    wasm,
};

pub const GlobalLinkage = enum {
    Internal,
    Strong,
    Weak,
    LinkOnce,
};

pub const AtomicOrder = enum {
    Unordered,
    Monotonic,
    Acquire,
    Release,
    AcqRel,
    SeqCst,
};

pub const AtomicRmwOp = enum {
    Xchg,
    Add,
    Sub,
    And,
    Nand,
    Or,
    Xor,
    Max,
    Min,
};

pub const Mode = enum {
    Debug,
    ReleaseSafe,
    ReleaseFast,
    ReleaseSmall,
};

pub const TypeId = enum {
    Type,
    Void,
    Bool,
    NoReturn,
    Int,
    Float,
    Pointer,
    Array,
    Struct,
    ComptimeFloat,
    ComptimeInt,
    Undefined,
    Null,
    Optional,
    ErrorUnion,
    ErrorSet,
    Enum,
    Union,
    Fn,
    Namespace,
    BoundFn,
    ArgTuple,
    Opaque,
    Promise,
};

pub const TypeInfo = union(TypeId) {
    Type: void,
    Void: void,
    Bool: void,
    NoReturn: void,
    Int: Int,
    Float: Float,
    Pointer: Pointer,
    Array: Array,
    Struct: Struct,
    ComptimeFloat: void,
    ComptimeInt: void,
    Undefined: void,
    Null: void,
    Optional: Optional,
    ErrorUnion: ErrorUnion,
    ErrorSet: ErrorSet,
    Enum: Enum,
    Union: Union,
    Fn: Fn,
    Namespace: void,
    BoundFn: Fn,
    ArgTuple: void,
    Opaque: void,
    Promise: Promise,


    pub const Int = struct {
        is_signed: bool,
        bits: u8,
    };

    pub const Float = struct {
        bits: u8,
    };

    pub const Pointer = struct {
        size: Size,
        is_const: bool,
        is_volatile: bool,
        alignment: u32,
        child: type,

        pub const Size = enum {
            One,
            Many,
            Slice,
        };
    };

    pub const Array = struct {
        len: usize,
        child: type,
    };

    pub const ContainerLayout = enum {
        Auto,
        Extern,
        Packed,
    };

    pub const StructField = struct {
        name: []const u8,
        offset: ?usize,
        field_type: type,
    };

    pub const Struct = struct {
        layout: ContainerLayout,
        fields: []StructField,
        defs: []Definition,
    };

    pub const Optional = struct {
        child: type,
    };

    pub const ErrorUnion = struct {
        error_set: type,
        payload: type,
    };

    pub const Error = struct {
        name: []const u8,
        value: usize,
    };

    pub const ErrorSet = struct {
        errors: []Error,
    };

    pub const EnumField = struct {
        name: []const u8,
        value: usize,
    };

    pub const Enum = struct {
        layout: ContainerLayout,
        tag_type: type,
        fields: []EnumField,
        defs: []Definition,
    };

    pub const UnionField = struct {
        name: []const u8,
        enum_field: ?EnumField,
        field_type: type,
    };

    pub const Union = struct {
        layout: ContainerLayout,
        tag_type: ?type,
        fields: []UnionField,
        defs: []Definition,
    };

    pub const CallingConvention = enum {
        Unspecified,
        C,
        Cold,
        Naked,
        Stdcall,
        Async,
    };

    pub const FnArg = struct {
        is_generic: bool,
        is_noalias: bool,
        arg_type: ?type,
    };

    pub const Fn = struct {
        calling_convention: CallingConvention,
        is_generic: bool,
        is_var_args: bool,
        return_type: ?type,
        async_allocator_type: ?type,
        args: []FnArg,
    };

    pub const Promise = struct {
        child: ?type,
    };

    pub const Definition = struct {
        name: []const u8,
        is_pub: bool,
        data: Data,

        pub const Data = union(enum) {
            Type: type,
            Var: type,
            Fn: FnDef,

            pub const FnDef = struct {
                fn_type: type,
                inline_type: Inline,
                calling_convention: CallingConvention,
                is_var_args: bool,
                is_extern: bool,
                is_export: bool,
                lib_name: ?[]const u8,
                return_type: type,
                arg_names: [][] const u8,

                pub const Inline = enum {
                    Auto,
                    Always,
                    Never,
                };
            };
        };
    };
};

pub const FloatMode = enum {
    Strict,
    Optimized,
};

pub const Endian = enum {
    Big,
    Little,
};

pub const endian = Endian.Little;
pub const is_test = false;
pub const os = Os.linux;
pub const arch = Arch.x86_64;
pub const environ = Environ.gnu;
pub const object_format = ObjectFormat.elf;
pub const mode = Mode.Debug;
pub const link_libc = false;
pub const have_error_return_tracing = false;
pub const __zig_test_fn_slice = {}; // overwritten later

See also:

Build Mode

Root Source File

TODO: explain how root source file finds other files

TODO: pub fn main

TODO: pub fn panic

TODO: if linking with libc you can use export fn main

TODO: order independent top level declarations

TODO: lazy analysis

TODO: using comptime { _ = @import() }

Zig Test

TODO: basic usage

TODO: lazy analysis

TODO: --test-filter

TODO: --test-name-prefix

TODO: testing in releasefast and releasesafe mode. assert still works

Zig Build System

TODO: explain purpose, it's supposed to replace make/cmake

TODO: example of building a zig executable

TODO: example of building a C library

C

Although Zig is independent of C, and, unlike most other languages, does not depend on libc, Zig acknowledges the importance of interacting with existing C code.

There are a few ways that Zig facilitates C interop.

C Type Primitives

These have guaranteed C ABI compatibility and can be used like any other type.

c_short
c_ushort
c_int
c_uint
c_long
c_ulong
c_longlong
c_ulonglong
c_longdouble
c_void

See also:

Primitive Types

C String Literals

test.zig

extern fn puts([*]const u8) void;

pub fn main() void {
    puts(c"this has a null terminator");
    puts(
        c\\and so
        c\\does this
        c\\multiline C string literal
    );
}

$ zig build-exe test.zig --library c
$ ./test
this has a null terminator
and so
does this
multiline C string literal

See also:

String Literals

Import from C Header File

The @cImport builtin function can be used to directly import symbols from .h files:

test.zig

const c = @cImport({
    // See https://github.com/ziglang/zig/issues/515
    @cDefine("_NO_CRT_STDIO_INLINE", "1");
    @cInclude("stdio.h");
});
pub fn main() void {
    _ = c.printf(c"hello\n");
}

$ zig build-exe test.zig --library c
$ ./test
hello

The @cImport function takes an expression as a parameter. This expression is evaluated at compile-time and is used to control preprocessor directives and include multiple .h files:

const builtin = @import("builtin");

const c = @cImport({
    @cDefine("NDEBUG", builtin.mode == builtin.Mode.ReleaseFast);
    if (something) {
        @cDefine("_GNU_SOURCE", {});
    }
    @cInclude("stdlib.h");
    if (something) {
        @cUndef("_GNU_SOURCE");
    }
    @cInclude("soundio.h");
});

See also:

@cImport
@cInclude
@cDefine
@cUndef
@import

Exporting a C Library

One of the primary use cases for Zig is exporting a library with the C ABI for other programming languages to call into. The export keyword in front of functions, variables, and types causes them to be part of the library API:

mathtest.zig

export fn add(a: i32, b: i32) i32 {
    return a + b;
}

To make a shared library:

$ zig build-lib mathtest.zig

To make a static library:

$ zig build-lib mathtest.zig --static

Here is an example with the Zig Build System:

test.c

// This header is generated by zig from mathtest.zig
#include "mathtest.h"
#include <assert.h>

int main(int argc, char **argv) {
    assert(add(42, 1337) == 1379);
    return 0;
}

build.zig

const Builder = @import("std").build.Builder;

pub fn build(b: *Builder) void {
    const lib = b.addSharedLibrary("mathtest", "mathtest.zig", b.version(1, 0, 0));

    const exe = b.addCExecutable("test");
    exe.addCompileFlags([][]const u8{"-std=c99"});
    exe.addSourceFile("test.c");
    exe.linkLibrary(lib);

    b.default_step.dependOn(&exe.step);

    const run_cmd = b.addCommand(".", b.env_map, [][]const u8{exe.getOutputPath()});
    run_cmd.step.dependOn(&exe.step);

    const test_step = b.step("test", "Test the program");
    test_step.dependOn(&run_cmd.step);
}

terminal

$ zig build
$ ./test
$ echo $?
0

Mixing Object Files

You can mix Zig object files with any other object files that respect the C ABI. Example:

base64.zig

const base64 = @import("std").base64;

export fn decode_base_64(
    dest_ptr: [*]u8,
    dest_len: usize,
    source_ptr: [*]const u8,
    source_len: usize,
) usize {
    const src = source_ptr[0..source_len];
    const dest = dest_ptr[0..dest_len];
    const base64_decoder = base64.standard_decoder_unsafe;
    const decoded_size = base64_decoder.calcSize(src);
    base64_decoder.decode(dest[0..decoded_size], src);
    return decoded_size;
}

test.c

// This header is generated by zig from base64.zig
#include "base64.h"

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

int main(int argc, char **argv) {
    const char *encoded = "YWxsIHlvdXIgYmFzZSBhcmUgYmVsb25nIHRvIHVz";
    char buf[200];

    size_t len = decode_base_64(buf, 200, encoded, strlen(encoded));
    buf[len] = 0;
    puts(buf);

    return 0;
}

build.zig

const Builder = @import("std").build.Builder;

pub fn build(b: *Builder) void {
    const obj = b.addObject("base64", "base64.zig");

    const exe = b.addCExecutable("test");
    exe.addCompileFlags([][]const u8 {
        "-std=c99",
    });
    exe.addSourceFile("test.c");
    exe.addObject(obj);
    exe.setOutputPath(".");

    b.default_step.dependOn(&exe.step);
}

terminal

$ zig build
$ ./test
all your base are belong to us

See also:

Targets
Zig Build System

Targets

Zig supports generating code for all targets that LLVM supports. Here is what it looks like to execute zig targets on a Linux x86_64 computer:

$ zig targets
Architectures:
  armv8_2a
  armv8_1a
  armv8
  armv8r
  armv8m_baseline
  armv8m_mainline
  armv7
  armv7em
  armv7m
  armv7s
  armv7k
  armv7ve
  armv6
  armv6m
  armv6k
  armv6t2
  armv5
  armv5te
  armv4t
  armeb
  aarch64
  aarch64_be
  avr
  bpfel
  bpfeb
  hexagon
  mips
  mipsel
  mips64
  mips64el
  msp430
  nios2
  powerpc
  powerpc64
  powerpc64le
  r600
  amdgcn
  riscv32
  riscv64
  sparc
  sparcv9
  sparcel
  s390x
  tce
  tcele
  thumb
  thumbeb
  i386
  x86_64 (native)
  xcore
  nvptx
  nvptx64
  le32
  le64
  amdil
  amdil64
  hsail
  hsail64
  spir
  spir64
  kalimbav3
  kalimbav4
  kalimbav5
  shave
  lanai
  wasm32
  wasm64
  renderscript32
  renderscript64

Operating Systems:
  freestanding
  ananas
  cloudabi
  dragonfly
  freebsd
  fuchsia
  ios
  kfreebsd
  linux (native)
  lv2
  macosx
  netbsd
  openbsd
  solaris
  windows
  haiku
  minix
  rtems
  nacl
  cnk
  bitrig
  aix
  cuda
  nvcl
  amdhsa
  ps4
  elfiamcu
  tvos
  watchos
  mesa3d
  contiki
  zen

Environments:
  unknown
  gnu (native)
  gnuabi64
  gnueabi
  gnueabihf
  gnux32
  code16
  eabi
  eabihf
  android
  musl
  musleabi
  musleabihf
  msvc
  itanium
  cygnus
  amdopencl
  coreclr
  opencl

The Zig Standard Library (@import("std")) has architecture, environment, and operating sytsem abstractions, and thus takes additional work to support more platforms. Not all standard library code requires operating system abstractions, however, so things such as generic data structures work an all above platforms.

The current list of targets supported by the Zig Standard Library is:

Linux x86_64
Windows x86_64
MacOS x86_64

Style Guide

These coding conventions are not enforced by the compiler, but they are shipped in this documentation along with the compiler in order to provide a point of reference, should anyone wish to point to an authority on agreed upon Zig coding style.

Whitespace

4 space indentation
Open braces on same line, unless you need to wrap.
If a list of things is longer than 2, put each item on its own line and exercise the abilty to put an extra comma at the end.
Line length: aim for 100; use common sense.

Names

Roughly speaking: camelCaseFunctionName, TitleCaseTypeName, snake_case_variable_name. More precisely:

If x is a struct (or an alias of a struct), then x should be TitleCase.
If x otherwise identifies a type, x should have snake_case.
If x is callable, and x's return type is type, then x should be TitleCase.
If x is otherwise callable, then x should be camelCase.
Otherwise, x should be snake_case.

Acronyms, initialisms, proper nouns, or any other word that has capitalization rules in written English are subject to naming conventions just like any other word. Even acronyms that are only 2 letters long are subject to these conventions.

These are general rules of thumb; if it makes sense to do something different, do what makes sense. For example, if there is an established convention such as ENOENT, follow the established convention.

Examples

const namespace_name = @import("dir_name/file_name.zig");
var global_var: i32 = undefined;
const const_name = 42;
const primitive_type_alias = f32;
const string_alias = []u8;

const StructName = struct {};
const StructAlias = StructName;

fn functionName(param_name: TypeName) void {
    var functionPointer = functionName;
    functionPointer();
    functionPointer = otherFunction;
    functionPointer();
}
const functionAlias = functionName;

fn ListTemplateFunction(comptime ChildType: type, comptime fixed_size: usize) type {
    return List(ChildType, fixed_size);
}

fn ShortList(comptime T: type, comptime n: usize) type {
    return struct {
        field_name: [n]T,
        fn methodName() void {}
    };
}

// The word XML loses its casing when used in Zig identifiers.
const xml_document =
    \\<?xml version="1.0" encoding="UTF-8"?>
    \\<document>
    \\</document>
;
const XmlParser = struct {};

// The initials BE (Big Endian) are just another word in Zig identifier names.
fn readU32Be() u32 {}

See the Zig Standard Library for more examples.

Source Encoding

Zig source code is encoded in UTF-8. An invalid UTF-8 byte sequence results in a compile error.

Throughout all zig source code (including in comments), some codepoints are never allowed:

Ascii control characters, except for U+000a (LF): U+0000 - U+0009, U+000b - U+0001f, U+007f. (Note that Windows line endings (CRLF) are not allowed, and hard tabs are not allowed.)
Non-Ascii Unicode line endings: U+0085 (NEL), U+2028 (LS), U+2029 (PS).

The codepoint U+000a (LF) (which is encoded as the single-byte value 0x0a) is the line terminator character. This character always terminates a line of zig source code (except possbly the last line of the file).

For some discussion on the rationale behind these design decisions, see issue #663

Grammar

Root = many(TopLevelItem) EOF

TopLevelItem = CompTimeExpression(Block) | TopLevelDecl | TestDecl

TestDecl = "test" String Block

TopLevelDecl = option("pub") (FnDef | ExternDecl | GlobalVarDecl | UseDecl)

GlobalVarDecl = option("export") VariableDeclaration ";"

LocalVarDecl = option("comptime") VariableDeclaration

VariableDeclaration = ("var" | "const") Symbol option(":" TypeExpr) option("align" "(" Expression ")") option("section" "(" Expression ")") "=" Expression

ContainerMember = (ContainerField | FnDef | GlobalVarDecl)

ContainerField = Symbol option(":" PrefixOpExpression) option("=" PrefixOpExpression) ","

UseDecl = "use" Expression ";"

ExternDecl = "extern" option(String) (FnProto | VariableDeclaration) ";"

FnProto = option("nakedcc" | "stdcallcc" | "extern" | ("async" option("<" Expression ">"))) "fn" option(Symbol) ParamDeclList option("align" "(" Expression ")") option("section" "(" Expression ")") option("!") (TypeExpr | "var")

FnDef = option("inline" | "export") FnProto Block

ParamDeclList = "(" list(ParamDecl, ",") ")"

ParamDecl = option("noalias" | "comptime") option(Symbol ":") (TypeExpr | "var" | "...")

Block = option(Symbol ":") "{" many(Statement) "}"

Statement = LocalVarDecl ";" | Defer(Block) | Defer(Expression) ";" | BlockExpression(Block) | Expression ";" | ";"

TypeExpr = (PrefixOpExpression "!" PrefixOpExpression) | PrefixOpExpression

BlockOrExpression = Block | Expression

Expression = TryExpression | ReturnExpression | BreakExpression | AssignmentExpression | CancelExpression | ResumeExpression

AsmExpression = "asm" option("volatile") "(" String option(AsmOutput) ")"

AsmOutput = ":" list(AsmOutputItem, ",") option(AsmInput)

AsmInput = ":" list(AsmInputItem, ",") option(AsmClobbers)

AsmOutputItem = "[" Symbol "]" String "(" (Symbol | "->" TypeExpr) ")"

AsmInputItem = "[" Symbol "]" String "(" Expression ")"

AsmClobbers= ":" list(String, ",")

UnwrapExpression = BoolOrExpression (UnwrapOptional | UnwrapError) | BoolOrExpression

UnwrapOptional = "orelse" Expression

UnwrapError = "catch" option("|" Symbol "|") Expression

AssignmentExpression = UnwrapExpression AssignmentOperator UnwrapExpression | UnwrapExpression

AssignmentOperator = "=" | "*=" | "/=" | "%=" | "+=" | "-=" | "<<=" | ">>=" | "&=" | "^=" | "|=" | "*%=" | "+%=" | "-%="

BlockExpression(body) = Block | IfExpression(body) | IfErrorExpression(body) | TestExpression(body) | WhileExpression(body) | ForExpression(body) | SwitchExpression | CompTimeExpression(body) | SuspendExpression(body)

CompTimeExpression(body) = "comptime" body

SwitchExpression = "switch" "(" Expression ")" "{" many(SwitchProng) "}"

SwitchProng = (list(SwitchItem, ",") | "else") "=>" option("|" option("*") Symbol "|") Expression ","

SwitchItem = Expression | (Expression "..." Expression)

ForExpression(body) = option(Symbol ":") option("inline") "for" "(" Expression ")" option("|" option("*") Symbol option("," Symbol) "|") body option("else" BlockExpression(body))

BoolOrExpression = BoolAndExpression "or" BoolOrExpression | BoolAndExpression

ReturnExpression = "return" option(Expression)

TryExpression = "try" Expression

AwaitExpression = "await" Expression

BreakExpression = "break" option(":" Symbol) option(Expression)

CancelExpression = "cancel" Expression;

ResumeExpression = "resume" Expression;

Defer(body) = ("defer" | "deferror") body

IfExpression(body) = "if" "(" Expression ")" body option("else" BlockExpression(body))

SuspendExpression(body) = "suspend" option( body )

IfErrorExpression(body) = "if" "(" Expression ")" option("|" option("*") Symbol "|") body "else" "|" Symbol "|" BlockExpression(body)

TestExpression(body) = "if" "(" Expression ")" option("|" option("*") Symbol "|") body option("else" BlockExpression(body))

WhileExpression(body) = option(Symbol ":") option("inline") "while" "(" Expression ")" option("|" option("*") Symbol "|") option(":" "(" Expression ")") body option("else" option("|" Symbol "|") BlockExpression(body))

BoolAndExpression = ComparisonExpression "and" BoolAndExpression | ComparisonExpression

ComparisonExpression = BinaryOrExpression ComparisonOperator BinaryOrExpression | BinaryOrExpression

ComparisonOperator = "==" | "!=" | "<" | ">" | "<=" | ">="

BinaryOrExpression = BinaryXorExpression "|" BinaryOrExpression | BinaryXorExpression

BinaryXorExpression = BinaryAndExpression "^" BinaryXorExpression | BinaryAndExpression

BinaryAndExpression = BitShiftExpression "&" BinaryAndExpression | BitShiftExpression

BitShiftExpression = AdditionExpression BitShiftOperator BitShiftExpression | AdditionExpression

BitShiftOperator = "<<" | ">>"

AdditionExpression = MultiplyExpression AdditionOperator AdditionExpression | MultiplyExpression

AdditionOperator = "+" | "-" | "++" | "+%" | "-%"

MultiplyExpression = CurlySuffixExpression MultiplyOperator MultiplyExpression | CurlySuffixExpression

CurlySuffixExpression = TypeExpr option(ContainerInitExpression)

MultiplyOperator = "||" | "*" | "/" | "%" | "**" | "*%"

PrefixOpExpression = PrefixOp TypeExpr | SuffixOpExpression

SuffixOpExpression = ("async" option("<" SuffixOpExpression ">") SuffixOpExpression FnCallExpression) | PrimaryExpression option(FnCallExpression | ArrayAccessExpression | FieldAccessExpression | SliceExpression | ".*" | ".?")

FieldAccessExpression = "." Symbol

FnCallExpression = "(" list(Expression, ",") ")"

ArrayAccessExpression = "[" Expression "]"

SliceExpression = "[" Expression ".." option(Expression) "]"

ContainerInitExpression = "{" ContainerInitBody "}"

ContainerInitBody = list(StructLiteralField, ",") | list(Expression, ",")

StructLiteralField = "." Symbol "=" Expression

PrefixOp = "!" | "-" | "~" | (("*" | "[*]") option("align" "(" Expression option(":" Integer ":" Integer) ")" ) option("const") option("volatile")) | "?" | "-%" | "try" | "await"

PrimaryExpression = Integer | Float | String | CharLiteral | KeywordLiteral | GroupedExpression | BlockExpression(BlockOrExpression) | Symbol | ("@" Symbol FnCallExpression) | ArrayType | FnProto | AsmExpression | ContainerDecl | ("continue" option(":" Symbol)) | ErrorSetDecl | PromiseType

PromiseType = "promise" option("->" TypeExpr)

ArrayType : "[" option(Expression) "]" option("align" "(" Expression option(":" Integer ":" Integer) ")")) option("const") option("volatile") TypeExpr

GroupedExpression = "(" Expression ")"

KeywordLiteral = "true" | "false" | "null" | "undefined" | "error" | "unreachable" | "suspend"

ErrorSetDecl = "error" "{" list(Symbol, ",") "}"

ContainerDecl = option("extern" | "packed")
  ("struct" option(GroupedExpression) | "union" option("enum" option(GroupedExpression) | GroupedExpression) | ("enum" option(GroupedExpression)))
  "{" many(ContainerMember) "}"

Zen

Communicate intent precisely.
Edge cases matter.
Favor reading code over writing code.
Only one obvious way to do things.
Runtime crashes are better than bugs.
Compile errors are better than runtime crashes.
Incremental improvements.
Avoid local maximums.
Reduce the amount one must remember.
Minimize energy spent on coding style.
Together we serve end users.