zig/lib/std/fmt.zig

1786 lines
61 KiB
Zig

const std = @import("std.zig");
const math = std.math;
const assert = std.debug.assert;
const mem = std.mem;
const builtin = @import("builtin");
const errol = @import("fmt/errol.zig");
const lossyCast = std.math.lossyCast;
pub const default_max_depth = 3;
pub const Alignment = enum {
Left,
Center,
Right,
};
pub const FormatOptions = struct {
precision: ?usize = null,
width: ?usize = null,
alignment: Alignment = .Left,
fill: u8 = ' ',
};
fn peekIsAlign(comptime fmt: []const u8) bool {
// Should only be called during a state transition to the format segment.
comptime assert(fmt[0] == ':');
inline for (([_]u8{ 1, 2 })[0..]) |i| {
if (fmt.len > i and (fmt[i] == '<' or fmt[i] == '^' or fmt[i] == '>')) {
return true;
}
}
return false;
}
/// Renders fmt string with args, calling output with slices of bytes.
/// If `output` returns an error, the error is returned from `format` and
/// `output` is not called again.
///
/// The format string must be comptime known and may contain placeholders following
/// this format:
/// `{[position][specifier]:[fill][alignment][width].[precision]}`
///
/// Each word between `[` and `]` is a parameter you have to replace with something:
///
/// - *position* is the index of the argument that should be inserted
/// - *specifier* is a type-dependent formatting option that determines how a type should formatted (see below)
/// - *fill* is a single character which is used to pad the formatted text
/// - *alignment* is one of the three characters `<`, `^` or `>`. they define if the text is *left*, *center*, or *right* aligned
/// - *width* is the total width of the field in characters
/// - *precision* specifies how many decimals a formatted number should have
///
/// Note that most of the parameters are optional and may be omitted. Also you can leave out separators like `:` and `.` when
/// all parameters after the separator are omitted.
/// Only exception is the *fill* parameter. If *fill* is required, one has to specify *alignment* as well, as otherwise
/// the digits after `:` is interpreted as *width*, not *fill*.
///
/// The *specifier* has several options for types:
/// - `x` and `X`:
/// - format the non-numeric value as a string of bytes in hexadecimal notation ("binary dump") in either lower case or upper case
/// - output numeric value in hexadecimal notation
/// - `s`: print a pointer-to-many as a c-string, use zero-termination
/// - `B` and `Bi`: output a memory size in either metric (1000) or power-of-two (1024) based notation. works for both float and integer values.
/// - `e`: output floating point value in scientific notation
/// - `d`: output numeric value in decimal notation
/// - `b`: output integer value in binary notation
/// - `c`: output integer as an ASCII character. Integer type must have 8 bits at max.
/// - `*`: output the address of the value instead of the value itself.
///
/// If a formatted user type contains a function of the type
/// ```
/// pub fn format(value: ?, comptime fmt: []const u8, options: std.fmt.FormatOptions, writer: var) !void
/// ```
/// with `?` being the type formatted, this function will be called instead of the default implementation.
/// This allows user types to be formatted in a logical manner instead of dumping all fields of the type.
///
/// A user type may be a `struct`, `vector`, `union` or `enum` type.
pub fn format(
writer: var,
comptime fmt: []const u8,
args: var,
) !void {
const ArgSetType = u32;
if (@typeInfo(@TypeOf(args)) != .Struct) {
@compileError("Expected tuple or struct argument, found " ++ @typeName(@TypeOf(args)));
}
if (args.len > ArgSetType.bit_count) {
@compileError("32 arguments max are supported per format call");
}
const State = enum {
Start,
Positional,
CloseBrace,
Specifier,
FormatFillAndAlign,
FormatWidth,
FormatPrecision,
};
comptime var start_index = 0;
comptime var state = State.Start;
comptime var maybe_pos_arg: ?comptime_int = null;
comptime var specifier_start = 0;
comptime var specifier_end = 0;
comptime var options = FormatOptions{};
comptime var arg_state: struct {
next_arg: usize = 0,
used_args: ArgSetType = 0,
args_len: usize = args.len,
fn hasUnusedArgs(comptime self: *@This()) bool {
return (@popCount(ArgSetType, self.used_args) != self.args_len);
}
fn nextArg(comptime self: *@This(), comptime pos_arg: ?comptime_int) comptime_int {
const next_idx = pos_arg orelse blk: {
const arg = self.next_arg;
self.next_arg += 1;
break :blk arg;
};
if (next_idx >= self.args_len) {
@compileError("Too few arguments");
}
// Mark this argument as used
self.used_args |= 1 << next_idx;
return next_idx;
}
} = .{};
inline for (fmt) |c, i| {
switch (state) {
.Start => switch (c) {
'{' => {
if (start_index < i) {
try writer.writeAll(fmt[start_index..i]);
}
start_index = i;
specifier_start = i + 1;
specifier_end = i + 1;
maybe_pos_arg = null;
state = .Positional;
options = FormatOptions{};
},
'}' => {
if (start_index < i) {
try writer.writeAll(fmt[start_index..i]);
}
state = .CloseBrace;
},
else => {},
},
.Positional => switch (c) {
'{' => {
state = .Start;
start_index = i;
},
':' => {
state = if (comptime peekIsAlign(fmt[i..])) State.FormatFillAndAlign else State.FormatWidth;
specifier_end = i;
},
'0'...'9' => {
if (maybe_pos_arg == null) {
maybe_pos_arg = 0;
}
maybe_pos_arg.? *= 10;
maybe_pos_arg.? += c - '0';
specifier_start = i + 1;
if (maybe_pos_arg.? >= args.len) {
@compileError("Positional value refers to non-existent argument");
}
},
'}' => {
const arg_to_print = comptime arg_state.nextArg(maybe_pos_arg);
try formatType(
args[arg_to_print],
fmt[0..0],
options,
writer,
default_max_depth,
);
state = .Start;
start_index = i + 1;
},
else => {
state = .Specifier;
specifier_start = i;
},
},
.CloseBrace => switch (c) {
'}' => {
state = .Start;
start_index = i;
},
else => @compileError("Single '}' encountered in format string"),
},
.Specifier => switch (c) {
':' => {
specifier_end = i;
state = if (comptime peekIsAlign(fmt[i..])) State.FormatFillAndAlign else State.FormatWidth;
},
'}' => {
const arg_to_print = comptime arg_state.nextArg(maybe_pos_arg);
try formatType(
args[arg_to_print],
fmt[specifier_start..i],
options,
writer,
default_max_depth,
);
state = .Start;
start_index = i + 1;
},
else => {},
},
// Only entered if the format string contains a fill/align segment.
.FormatFillAndAlign => switch (c) {
'<' => {
options.alignment = Alignment.Left;
state = .FormatWidth;
},
'^' => {
options.alignment = Alignment.Center;
state = .FormatWidth;
},
'>' => {
options.alignment = Alignment.Right;
state = .FormatWidth;
},
else => {
options.fill = c;
},
},
.FormatWidth => switch (c) {
'0'...'9' => {
if (options.width == null) {
options.width = 0;
}
options.width.? *= 10;
options.width.? += c - '0';
},
'.' => {
state = .FormatPrecision;
},
'}' => {
const arg_to_print = comptime arg_state.nextArg(maybe_pos_arg);
try formatType(
args[arg_to_print],
fmt[specifier_start..specifier_end],
options,
writer,
default_max_depth,
);
state = .Start;
start_index = i + 1;
},
else => {
@compileError("Unexpected character in width value: " ++ [_]u8{c});
},
},
.FormatPrecision => switch (c) {
'0'...'9' => {
if (options.precision == null) {
options.precision = 0;
}
options.precision.? *= 10;
options.precision.? += c - '0';
},
'}' => {
const arg_to_print = comptime arg_state.nextArg(maybe_pos_arg);
try formatType(
args[arg_to_print],
fmt[specifier_start..specifier_end],
options,
writer,
default_max_depth,
);
state = .Start;
start_index = i + 1;
},
else => {
@compileError("Unexpected character in precision value: " ++ [_]u8{c});
},
},
}
}
comptime {
if (comptime arg_state.hasUnusedArgs()) {
@compileError("Unused arguments");
}
if (state != State.Start) {
@compileError("Incomplete format string: " ++ fmt);
}
}
if (start_index < fmt.len) {
try writer.writeAll(fmt[start_index..]);
}
}
pub fn formatType(
value: var,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
max_depth: usize,
) @TypeOf(writer).Error!void {
if (comptime std.mem.eql(u8, fmt, "*")) {
try writer.writeAll(@typeName(@TypeOf(value).Child));
try writer.writeAll("@");
try formatInt(@ptrToInt(value), 16, false, FormatOptions{}, writer);
return;
}
const T = @TypeOf(value);
if (comptime std.meta.trait.hasFn("format")(T)) {
return try value.format(fmt, options, writer);
}
switch (@typeInfo(T)) {
.ComptimeInt, .Int, .ComptimeFloat, .Float => {
return formatValue(value, fmt, options, writer);
},
.Void => {
return formatBuf("void", options, writer);
},
.Bool => {
return formatBuf(if (value) "true" else "false", options, writer);
},
.Optional => {
if (value) |payload| {
return formatType(payload, fmt, options, writer, max_depth);
} else {
return formatBuf("null", options, writer);
}
},
.ErrorUnion => {
if (value) |payload| {
return formatType(payload, fmt, options, writer, max_depth);
} else |err| {
return formatType(err, fmt, options, writer, max_depth);
}
},
.ErrorSet => {
try writer.writeAll("error.");
return writer.writeAll(@errorName(value));
},
.Enum => |enumInfo| {
try writer.writeAll(@typeName(T));
if (enumInfo.is_exhaustive) {
try writer.writeAll(".");
try writer.writeAll(@tagName(value));
return;
}
// Use @tagName only if value is one of known fields
@setEvalBranchQuota(3 * enumInfo.fields.len);
inline for (enumInfo.fields) |enumField| {
if (@enumToInt(value) == enumField.value) {
try writer.writeAll(".");
try writer.writeAll(@tagName(value));
return;
}
}
try writer.writeAll("(");
try formatType(@enumToInt(value), fmt, options, writer, max_depth);
try writer.writeAll(")");
},
.Union => {
try writer.writeAll(@typeName(T));
if (max_depth == 0) {
return writer.writeAll("{ ... }");
}
const info = @typeInfo(T).Union;
if (info.tag_type) |UnionTagType| {
try writer.writeAll("{ .");
try writer.writeAll(@tagName(@as(UnionTagType, value)));
try writer.writeAll(" = ");
inline for (info.fields) |u_field| {
if (@enumToInt(@as(UnionTagType, value)) == u_field.enum_field.?.value) {
try formatType(@field(value, u_field.name), fmt, options, writer, max_depth - 1);
}
}
try writer.writeAll(" }");
} else {
try format(writer, "@{x}", .{@ptrToInt(&value)});
}
},
.Struct => |StructT| {
try writer.writeAll(@typeName(T));
if (max_depth == 0) {
return writer.writeAll("{ ... }");
}
try writer.writeAll("{");
inline for (StructT.fields) |f, i| {
if (i == 0) {
try writer.writeAll(" .");
} else {
try writer.writeAll(", .");
}
try writer.writeAll(f.name);
try writer.writeAll(" = ");
try formatType(@field(value, f.name), fmt, options, writer, max_depth - 1);
}
try writer.writeAll(" }");
},
.Pointer => |ptr_info| switch (ptr_info.size) {
.One => switch (@typeInfo(ptr_info.child)) {
.Array => |info| {
if (info.child == u8) {
return formatText(value, fmt, options, writer);
}
return format(writer, "{}@{x}", .{ @typeName(T.Child), @ptrToInt(value) });
},
.Enum, .Union, .Struct => {
return formatType(value.*, fmt, options, writer, max_depth);
},
else => return format(writer, "{}@{x}", .{ @typeName(T.Child), @ptrToInt(value) }),
},
.Many, .C => {
if (ptr_info.sentinel) |sentinel| {
return formatType(mem.span(value), fmt, options, writer, max_depth);
}
if (ptr_info.child == u8) {
if (fmt.len > 0 and fmt[0] == 's') {
return formatText(mem.span(value), fmt, options, writer);
}
}
return format(writer, "{}@{x}", .{ @typeName(T.Child), @ptrToInt(value) });
},
.Slice => {
if (fmt.len > 0 and ((fmt[0] == 'x') or (fmt[0] == 'X'))) {
return formatText(value, fmt, options, writer);
}
if (ptr_info.child == u8) {
return formatText(value, fmt, options, writer);
}
return format(writer, "{}@{x}", .{ @typeName(ptr_info.child), @ptrToInt(value.ptr) });
},
},
.Array => |info| {
const Slice = @Type(builtin.TypeInfo{
.Pointer = .{
.size = .Slice,
.is_const = true,
.is_volatile = false,
.is_allowzero = false,
.alignment = @alignOf(info.child),
.child = info.child,
.sentinel = null,
},
});
return formatType(@as(Slice, &value), fmt, options, writer, max_depth);
},
.Vector => {
const len = @typeInfo(T).Vector.len;
try writer.writeAll("{ ");
var i: usize = 0;
while (i < len) : (i += 1) {
try formatValue(value[i], fmt, options, writer);
if (i < len - 1) {
try writer.writeAll(", ");
}
}
try writer.writeAll(" }");
},
.Fn => {
return format(writer, "{}@{x}", .{ @typeName(T), @ptrToInt(value) });
},
.Type => return writer.writeAll(@typeName(T)),
.EnumLiteral => {
const buffer = [_]u8{'.'} ++ @tagName(value);
return formatType(buffer, fmt, options, writer, max_depth);
},
else => @compileError("Unable to format type '" ++ @typeName(T) ++ "'"),
}
}
fn formatValue(
value: var,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
if (comptime std.mem.eql(u8, fmt, "B")) {
return formatBytes(value, options, 1000, writer);
} else if (comptime std.mem.eql(u8, fmt, "Bi")) {
return formatBytes(value, options, 1024, writer);
}
const T = @TypeOf(value);
switch (@typeInfo(T)) {
.Float, .ComptimeFloat => return formatFloatValue(value, fmt, options, writer),
.Int, .ComptimeInt => return formatIntValue(value, fmt, options, writer),
.Bool => return formatBuf(if (value) "true" else "false", options, writer),
else => comptime unreachable,
}
}
pub fn formatIntValue(
value: var,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
comptime var radix = 10;
comptime var uppercase = false;
const int_value = if (@TypeOf(value) == comptime_int) blk: {
const Int = math.IntFittingRange(value, value);
break :blk @as(Int, value);
} else
value;
if (fmt.len == 0 or comptime std.mem.eql(u8, fmt, "d")) {
radix = 10;
uppercase = false;
} else if (comptime std.mem.eql(u8, fmt, "c")) {
if (@TypeOf(int_value).bit_count <= 8) {
return formatAsciiChar(@as(u8, int_value), options, writer);
} else {
@compileError("Cannot print integer that is larger than 8 bits as a ascii");
}
} else if (comptime std.mem.eql(u8, fmt, "b")) {
radix = 2;
uppercase = false;
} else if (comptime std.mem.eql(u8, fmt, "x")) {
radix = 16;
uppercase = false;
} else if (comptime std.mem.eql(u8, fmt, "X")) {
radix = 16;
uppercase = true;
} else {
@compileError("Unknown format string: '" ++ fmt ++ "'");
}
return formatInt(int_value, radix, uppercase, options, writer);
}
fn formatFloatValue(
value: var,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
if (fmt.len == 0 or comptime std.mem.eql(u8, fmt, "e")) {
return formatFloatScientific(value, options, writer);
} else if (comptime std.mem.eql(u8, fmt, "d")) {
return formatFloatDecimal(value, options, writer);
} else {
@compileError("Unknown format string: '" ++ fmt ++ "'");
}
}
pub fn formatText(
bytes: []const u8,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
if (comptime std.mem.eql(u8, fmt, "s") or (fmt.len == 0)) {
return formatBuf(bytes, options, writer);
} else if (comptime (std.mem.eql(u8, fmt, "x") or std.mem.eql(u8, fmt, "X"))) {
for (bytes) |c| {
try formatInt(c, 16, fmt[0] == 'X', FormatOptions{ .width = 2, .fill = '0' }, writer);
}
return;
} else {
@compileError("Unknown format string: '" ++ fmt ++ "'");
}
}
pub fn formatAsciiChar(
c: u8,
options: FormatOptions,
writer: var,
) !void {
return writer.writeAll(@as(*const [1]u8, &c));
}
pub fn formatBuf(
buf: []const u8,
options: FormatOptions,
writer: var,
) !void {
const width = options.width orelse buf.len;
var padding = if (width > buf.len) (width - buf.len) else 0;
const pad_byte = [1]u8{options.fill};
switch (options.alignment) {
.Left => {
try writer.writeAll(buf);
while (padding > 0) : (padding -= 1) {
try writer.writeAll(&pad_byte);
}
},
.Center => {
const padl = padding / 2;
var i: usize = 0;
while (i < padl) : (i += 1) try writer.writeAll(&pad_byte);
try writer.writeAll(buf);
while (i < padding) : (i += 1) try writer.writeAll(&pad_byte);
},
.Right => {
while (padding > 0) : (padding -= 1) {
try writer.writeAll(&pad_byte);
}
try writer.writeAll(buf);
},
}
}
// Print a float in scientific notation to the specified precision. Null uses full precision.
// It should be the case that every full precision, printed value can be re-parsed back to the
// same type unambiguously.
pub fn formatFloatScientific(
value: var,
options: FormatOptions,
writer: var,
) !void {
var x = @floatCast(f64, value);
// Errol doesn't handle these special cases.
if (math.signbit(x)) {
try writer.writeAll("-");
x = -x;
}
if (math.isNan(x)) {
return writer.writeAll("nan");
}
if (math.isPositiveInf(x)) {
return writer.writeAll("inf");
}
if (x == 0.0) {
try writer.writeAll("0");
if (options.precision) |precision| {
if (precision != 0) {
try writer.writeAll(".");
var i: usize = 0;
while (i < precision) : (i += 1) {
try writer.writeAll("0");
}
}
} else {
try writer.writeAll(".0");
}
try writer.writeAll("e+00");
return;
}
var buffer: [32]u8 = undefined;
var float_decimal = errol.errol3(x, buffer[0..]);
if (options.precision) |precision| {
errol.roundToPrecision(&float_decimal, precision, errol.RoundMode.Scientific);
try writer.writeAll(float_decimal.digits[0..1]);
// {e0} case prints no `.`
if (precision != 0) {
try writer.writeAll(".");
var printed: usize = 0;
if (float_decimal.digits.len > 1) {
const num_digits = math.min(float_decimal.digits.len, precision + 1);
try writer.writeAll(float_decimal.digits[1..num_digits]);
printed += num_digits - 1;
}
while (printed < precision) : (printed += 1) {
try writer.writeAll("0");
}
}
} else {
try writer.writeAll(float_decimal.digits[0..1]);
try writer.writeAll(".");
if (float_decimal.digits.len > 1) {
const num_digits = if (@TypeOf(value) == f32) math.min(@as(usize, 9), float_decimal.digits.len) else float_decimal.digits.len;
try writer.writeAll(float_decimal.digits[1..num_digits]);
} else {
try writer.writeAll("0");
}
}
try writer.writeAll("e");
const exp = float_decimal.exp - 1;
if (exp >= 0) {
try writer.writeAll("+");
if (exp > -10 and exp < 10) {
try writer.writeAll("0");
}
try formatInt(exp, 10, false, FormatOptions{ .width = 0 }, writer);
} else {
try writer.writeAll("-");
if (exp > -10 and exp < 10) {
try writer.writeAll("0");
}
try formatInt(-exp, 10, false, FormatOptions{ .width = 0 }, writer);
}
}
// Print a float of the format x.yyyyy where the number of y is specified by the precision argument.
// By default floats are printed at full precision (no rounding).
pub fn formatFloatDecimal(
value: var,
options: FormatOptions,
writer: var,
) !void {
var x = @as(f64, value);
// Errol doesn't handle these special cases.
if (math.signbit(x)) {
try writer.writeAll("-");
x = -x;
}
if (math.isNan(x)) {
return writer.writeAll("nan");
}
if (math.isPositiveInf(x)) {
return writer.writeAll("inf");
}
if (x == 0.0) {
try writer.writeAll("0");
if (options.precision) |precision| {
if (precision != 0) {
try writer.writeAll(".");
var i: usize = 0;
while (i < precision) : (i += 1) {
try writer.writeAll("0");
}
} else {
try writer.writeAll(".0");
}
}
return;
}
// non-special case, use errol3
var buffer: [32]u8 = undefined;
var float_decimal = errol.errol3(x, buffer[0..]);
if (options.precision) |precision| {
errol.roundToPrecision(&float_decimal, precision, errol.RoundMode.Decimal);
// exp < 0 means the leading is always 0 as errol result is normalized.
var num_digits_whole = if (float_decimal.exp > 0) @intCast(usize, float_decimal.exp) else 0;
// the actual slice into the buffer, we may need to zero-pad between num_digits_whole and this.
var num_digits_whole_no_pad = math.min(num_digits_whole, float_decimal.digits.len);
if (num_digits_whole > 0) {
// We may have to zero pad, for instance 1e4 requires zero padding.
try writer.writeAll(float_decimal.digits[0..num_digits_whole_no_pad]);
var i = num_digits_whole_no_pad;
while (i < num_digits_whole) : (i += 1) {
try writer.writeAll("0");
}
} else {
try writer.writeAll("0");
}
// {.0} special case doesn't want a trailing '.'
if (precision == 0) {
return;
}
try writer.writeAll(".");
// Keep track of fractional count printed for case where we pre-pad then post-pad with 0's.
var printed: usize = 0;
// Zero-fill until we reach significant digits or run out of precision.
if (float_decimal.exp <= 0) {
const zero_digit_count = @intCast(usize, -float_decimal.exp);
const zeros_to_print = math.min(zero_digit_count, precision);
var i: usize = 0;
while (i < zeros_to_print) : (i += 1) {
try writer.writeAll("0");
printed += 1;
}
if (printed >= precision) {
return;
}
}
// Remaining fractional portion, zero-padding if insufficient.
assert(precision >= printed);
if (num_digits_whole_no_pad + precision - printed < float_decimal.digits.len) {
try writer.writeAll(float_decimal.digits[num_digits_whole_no_pad .. num_digits_whole_no_pad + precision - printed]);
return;
} else {
try writer.writeAll(float_decimal.digits[num_digits_whole_no_pad..]);
printed += float_decimal.digits.len - num_digits_whole_no_pad;
while (printed < precision) : (printed += 1) {
try writer.writeAll("0");
}
}
} else {
// exp < 0 means the leading is always 0 as errol result is normalized.
var num_digits_whole = if (float_decimal.exp > 0) @intCast(usize, float_decimal.exp) else 0;
// the actual slice into the buffer, we may need to zero-pad between num_digits_whole and this.
var num_digits_whole_no_pad = math.min(num_digits_whole, float_decimal.digits.len);
if (num_digits_whole > 0) {
// We may have to zero pad, for instance 1e4 requires zero padding.
try writer.writeAll(float_decimal.digits[0..num_digits_whole_no_pad]);
var i = num_digits_whole_no_pad;
while (i < num_digits_whole) : (i += 1) {
try writer.writeAll("0");
}
} else {
try writer.writeAll("0");
}
// Omit `.` if no fractional portion
if (float_decimal.exp >= 0 and num_digits_whole_no_pad == float_decimal.digits.len) {
return;
}
try writer.writeAll(".");
// Zero-fill until we reach significant digits or run out of precision.
if (float_decimal.exp < 0) {
const zero_digit_count = @intCast(usize, -float_decimal.exp);
var i: usize = 0;
while (i < zero_digit_count) : (i += 1) {
try writer.writeAll("0");
}
}
try writer.writeAll(float_decimal.digits[num_digits_whole_no_pad..]);
}
}
pub fn formatBytes(
value: var,
options: FormatOptions,
comptime radix: usize,
writer: var,
) !void {
if (value == 0) {
return writer.writeAll("0B");
}
const is_float = comptime std.meta.trait.is(.Float)(@TypeOf(value));
const mags_si = " kMGTPEZY";
const mags_iec = " KMGTPEZY";
const log2 = if (is_float) @floatToInt(usize, math.log2(value)) else math.log2(value);
const magnitude = switch (radix) {
1000 => math.min(log2 / comptime math.log2(1000), mags_si.len - 1),
1024 => math.min(log2 / 10, mags_iec.len - 1),
else => unreachable,
};
const new_value = lossyCast(f64, value) / math.pow(f64, lossyCast(f64, radix), lossyCast(f64, magnitude));
const suffix = switch (radix) {
1000 => mags_si[magnitude],
1024 => mags_iec[magnitude],
else => unreachable,
};
try formatFloatDecimal(new_value, options, writer);
if (suffix == ' ') {
return writer.writeAll("B");
}
const buf = switch (radix) {
1000 => &[_]u8{ suffix, 'B' },
1024 => &[_]u8{ suffix, 'i', 'B' },
else => unreachable,
};
return writer.writeAll(buf);
}
pub fn formatInt(
value: var,
base: u8,
uppercase: bool,
options: FormatOptions,
writer: var,
) !void {
const int_value = if (@TypeOf(value) == comptime_int) blk: {
const Int = math.IntFittingRange(value, value);
break :blk @as(Int, value);
} else
value;
if (@TypeOf(int_value).is_signed) {
return formatIntSigned(int_value, base, uppercase, options, writer);
} else {
return formatIntUnsigned(int_value, base, uppercase, options, writer);
}
}
fn formatIntSigned(
value: var,
base: u8,
uppercase: bool,
options: FormatOptions,
writer: var,
) !void {
const new_options = FormatOptions{
.width = if (options.width) |w| (if (w == 0) 0 else w - 1) else null,
.precision = options.precision,
.fill = options.fill,
};
const bit_count = @typeInfo(@TypeOf(value)).Int.bits;
const Uint = std.meta.Int(false, bit_count);
if (value < 0) {
try writer.writeAll("-");
const new_value = math.absCast(value);
return formatIntUnsigned(new_value, base, uppercase, new_options, writer);
} else if (options.width == null or options.width.? == 0) {
return formatIntUnsigned(@intCast(Uint, value), base, uppercase, options, writer);
} else {
try writer.writeAll("+");
const new_value = @intCast(Uint, value);
return formatIntUnsigned(new_value, base, uppercase, new_options, writer);
}
}
fn formatIntUnsigned(
value: var,
base: u8,
uppercase: bool,
options: FormatOptions,
writer: var,
) !void {
assert(base >= 2);
var buf: [math.max(@TypeOf(value).bit_count, 1)]u8 = undefined;
const min_int_bits = comptime math.max(@TypeOf(value).bit_count, @TypeOf(base).bit_count);
const MinInt = std.meta.Int(@TypeOf(value).is_signed, min_int_bits);
var a: MinInt = value;
var index: usize = buf.len;
while (true) {
const digit = a % base;
index -= 1;
buf[index] = digitToChar(@intCast(u8, digit), uppercase);
a /= base;
if (a == 0) break;
}
const digits_buf = buf[index..];
const width = options.width orelse 0;
const padding = if (width > digits_buf.len) (width - digits_buf.len) else 0;
if (padding > index) {
const zero_byte: u8 = options.fill;
var leftover_padding = padding - index;
while (true) {
try writer.writeAll(@as(*const [1]u8, &zero_byte)[0..]);
leftover_padding -= 1;
if (leftover_padding == 0) break;
}
mem.set(u8, buf[0..index], options.fill);
return writer.writeAll(&buf);
} else {
const padded_buf = buf[index - padding ..];
mem.set(u8, padded_buf[0..padding], options.fill);
return writer.writeAll(padded_buf);
}
}
pub fn formatIntBuf(out_buf: []u8, value: var, base: u8, uppercase: bool, options: FormatOptions) usize {
var fbs = std.io.fixedBufferStream(out_buf);
formatInt(value, base, uppercase, options, fbs.writer()) catch unreachable;
return fbs.pos;
}
pub const ParseIntError = error{
/// The result cannot fit in the type specified
Overflow,
/// The input was empty or had a byte that was not a digit
InvalidCharacter,
};
pub fn parseInt(comptime T: type, buf: []const u8, radix: u8) ParseIntError!T {
if (buf.len == 0) return error.InvalidCharacter;
if (buf[0] == '+') return parseWithSign(T, buf[1..], radix, .Pos);
if (buf[0] == '-') return parseWithSign(T, buf[1..], radix, .Neg);
return parseWithSign(T, buf, radix, .Pos);
}
test "parseInt" {
std.testing.expect((try parseInt(i32, "-10", 10)) == -10);
std.testing.expect((try parseInt(i32, "+10", 10)) == 10);
std.testing.expect((try parseInt(u32, "+10", 10)) == 10);
std.testing.expectError(error.Overflow, parseInt(u32, "-10", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(u32, " 10", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(u32, "10 ", 10));
std.testing.expect((try parseInt(u8, "255", 10)) == 255);
std.testing.expectError(error.Overflow, parseInt(u8, "256", 10));
// +0 and -0 should work for unsigned
std.testing.expect((try parseInt(u8, "-0", 10)) == 0);
std.testing.expect((try parseInt(u8, "+0", 10)) == 0);
// ensure minInt is parsed correctly
std.testing.expect((try parseInt(i8, "-128", 10)) == math.minInt(i8));
std.testing.expect((try parseInt(i43, "-4398046511104", 10)) == math.minInt(i43));
// empty string or bare +- is invalid
std.testing.expectError(error.InvalidCharacter, parseInt(u32, "", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(i32, "", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(u32, "+", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(i32, "+", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(u32, "-", 10));
std.testing.expectError(error.InvalidCharacter, parseInt(i32, "-", 10));
}
fn parseWithSign(
comptime T: type,
buf: []const u8,
radix: u8,
comptime sign: enum { Pos, Neg },
) ParseIntError!T {
if (buf.len == 0) return error.InvalidCharacter;
const add = switch (sign) {
.Pos => math.add,
.Neg => math.sub,
};
var x: T = 0;
for (buf) |c| {
const digit = try charToDigit(c, radix);
if (x != 0) x = try math.mul(T, x, try math.cast(T, radix));
x = try add(T, x, try math.cast(T, digit));
}
return x;
}
pub fn parseUnsigned(comptime T: type, buf: []const u8, radix: u8) ParseIntError!T {
return parseWithSign(T, buf, radix, .Pos);
}
test "parseUnsigned" {
std.testing.expect((try parseUnsigned(u16, "050124", 10)) == 50124);
std.testing.expect((try parseUnsigned(u16, "65535", 10)) == 65535);
std.testing.expectError(error.Overflow, parseUnsigned(u16, "65536", 10));
std.testing.expect((try parseUnsigned(u64, "0ffffffffffffffff", 16)) == 0xffffffffffffffff);
std.testing.expectError(error.Overflow, parseUnsigned(u64, "10000000000000000", 16));
std.testing.expect((try parseUnsigned(u32, "DeadBeef", 16)) == 0xDEADBEEF);
std.testing.expect((try parseUnsigned(u7, "1", 10)) == 1);
std.testing.expect((try parseUnsigned(u7, "1000", 2)) == 8);
std.testing.expectError(error.InvalidCharacter, parseUnsigned(u32, "f", 10));
std.testing.expectError(error.InvalidCharacter, parseUnsigned(u8, "109", 8));
std.testing.expect((try parseUnsigned(u32, "NUMBER", 36)) == 1442151747);
// these numbers should fit even though the radix itself doesn't fit in the destination type
std.testing.expect((try parseUnsigned(u1, "0", 10)) == 0);
std.testing.expect((try parseUnsigned(u1, "1", 10)) == 1);
std.testing.expectError(error.Overflow, parseUnsigned(u1, "2", 10));
std.testing.expect((try parseUnsigned(u1, "001", 16)) == 1);
std.testing.expect((try parseUnsigned(u2, "3", 16)) == 3);
std.testing.expectError(error.Overflow, parseUnsigned(u2, "4", 16));
// parseUnsigned does not expect a sign
std.testing.expectError(error.InvalidCharacter, parseUnsigned(u8, "+0", 10));
std.testing.expectError(error.InvalidCharacter, parseUnsigned(u8, "-0", 10));
// test empty string error
std.testing.expectError(error.InvalidCharacter, parseUnsigned(u8, "", 10));
}
pub const parseFloat = @import("fmt/parse_float.zig").parseFloat;
test "parseFloat" {
_ = @import("fmt/parse_float.zig");
}
pub fn charToDigit(c: u8, radix: u8) (error{InvalidCharacter}!u8) {
const value = switch (c) {
'0'...'9' => c - '0',
'A'...'Z' => c - 'A' + 10,
'a'...'z' => c - 'a' + 10,
else => return error.InvalidCharacter,
};
if (value >= radix) return error.InvalidCharacter;
return value;
}
pub fn digitToChar(digit: u8, uppercase: bool) u8 {
return switch (digit) {
0...9 => digit + '0',
10...35 => digit + ((if (uppercase) @as(u8, 'A') else @as(u8, 'a')) - 10),
else => unreachable,
};
}
pub const BufPrintError = error{
/// As much as possible was written to the buffer, but it was too small to fit all the printed bytes.
NoSpaceLeft,
};
pub fn bufPrint(buf: []u8, comptime fmt: []const u8, args: var) BufPrintError![]u8 {
var fbs = std.io.fixedBufferStream(buf);
try format(fbs.writer(), fmt, args);
return fbs.getWritten();
}
// Count the characters needed for format. Useful for preallocating memory
pub fn count(comptime fmt: []const u8, args: var) u64 {
var counting_writer = std.io.countingWriter(std.io.null_writer);
format(counting_writer.writer(), fmt, args) catch |err| switch (err) {};
return counting_writer.bytes_written;
}
pub const AllocPrintError = error{OutOfMemory};
pub fn allocPrint(allocator: *mem.Allocator, comptime fmt: []const u8, args: var) AllocPrintError![]u8 {
const size = math.cast(usize, count(fmt, args)) catch |err| switch (err) {
// Output too long. Can't possibly allocate enough memory to display it.
error.Overflow => return error.OutOfMemory,
};
const buf = try allocator.alloc(u8, size);
return bufPrint(buf, fmt, args) catch |err| switch (err) {
error.NoSpaceLeft => unreachable, // we just counted the size above
};
}
pub fn allocPrint0(allocator: *mem.Allocator, comptime fmt: []const u8, args: var) AllocPrintError![:0]u8 {
const result = try allocPrint(allocator, fmt ++ "\x00", args);
return result[0 .. result.len - 1 :0];
}
test "bufPrintInt" {
var buffer: [100]u8 = undefined;
const buf = buffer[0..];
std.testing.expectEqualSlices(u8, "-1", bufPrintIntToSlice(buf, @as(i1, -1), 10, false, FormatOptions{}));
std.testing.expectEqualSlices(u8, "-101111000110000101001110", bufPrintIntToSlice(buf, @as(i32, -12345678), 2, false, FormatOptions{}));
std.testing.expectEqualSlices(u8, "-12345678", bufPrintIntToSlice(buf, @as(i32, -12345678), 10, false, FormatOptions{}));
std.testing.expectEqualSlices(u8, "-bc614e", bufPrintIntToSlice(buf, @as(i32, -12345678), 16, false, FormatOptions{}));
std.testing.expectEqualSlices(u8, "-BC614E", bufPrintIntToSlice(buf, @as(i32, -12345678), 16, true, FormatOptions{}));
std.testing.expectEqualSlices(u8, "12345678", bufPrintIntToSlice(buf, @as(u32, 12345678), 10, true, FormatOptions{}));
std.testing.expectEqualSlices(u8, " 666", bufPrintIntToSlice(buf, @as(u32, 666), 10, false, FormatOptions{ .width = 6 }));
std.testing.expectEqualSlices(u8, " 1234", bufPrintIntToSlice(buf, @as(u32, 0x1234), 16, false, FormatOptions{ .width = 6 }));
std.testing.expectEqualSlices(u8, "1234", bufPrintIntToSlice(buf, @as(u32, 0x1234), 16, false, FormatOptions{ .width = 1 }));
std.testing.expectEqualSlices(u8, "+42", bufPrintIntToSlice(buf, @as(i32, 42), 10, false, FormatOptions{ .width = 3 }));
std.testing.expectEqualSlices(u8, "-42", bufPrintIntToSlice(buf, @as(i32, -42), 10, false, FormatOptions{ .width = 3 }));
}
fn bufPrintIntToSlice(buf: []u8, value: var, base: u8, uppercase: bool, options: FormatOptions) []u8 {
return buf[0..formatIntBuf(buf, value, base, uppercase, options)];
}
test "parse u64 digit too big" {
_ = parseUnsigned(u64, "123a", 10) catch |err| {
if (err == error.InvalidCharacter) return;
unreachable;
};
unreachable;
}
test "parse unsigned comptime" {
comptime {
std.testing.expect((try parseUnsigned(usize, "2", 10)) == 2);
}
}
test "optional" {
{
const value: ?i32 = 1234;
try testFmt("optional: 1234\n", "optional: {}\n", .{value});
}
{
const value: ?i32 = null;
try testFmt("optional: null\n", "optional: {}\n", .{value});
}
}
test "error" {
{
const value: anyerror!i32 = 1234;
try testFmt("error union: 1234\n", "error union: {}\n", .{value});
}
{
const value: anyerror!i32 = error.InvalidChar;
try testFmt("error union: error.InvalidChar\n", "error union: {}\n", .{value});
}
}
test "int.small" {
{
const value: u3 = 0b101;
try testFmt("u3: 5\n", "u3: {}\n", .{value});
}
}
test "int.specifier" {
{
const value: u8 = 'a';
try testFmt("u8: a\n", "u8: {c}\n", .{value});
}
{
const value: u8 = 0b1100;
try testFmt("u8: 0b1100\n", "u8: 0b{b}\n", .{value});
}
}
test "int.padded" {
try testFmt("u8: ' 1'", "u8: '{:4}'", .{@as(u8, 1)});
try testFmt("u8: 'xxx1'", "u8: '{:x<4}'", .{@as(u8, 1)});
}
test "buffer" {
{
var buf1: [32]u8 = undefined;
var fbs = std.io.fixedBufferStream(&buf1);
try formatType(1234, "", FormatOptions{}, fbs.writer(), default_max_depth);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "1234"));
fbs.reset();
try formatType('a', "c", FormatOptions{}, fbs.writer(), default_max_depth);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "a"));
fbs.reset();
try formatType(0b1100, "b", FormatOptions{}, fbs.writer(), default_max_depth);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "1100"));
}
}
test "array" {
{
const value: [3]u8 = "abc".*;
try testFmt("array: abc\n", "array: {}\n", .{value});
try testFmt("array: abc\n", "array: {}\n", .{&value});
var buf: [100]u8 = undefined;
try testFmt(
try bufPrint(buf[0..], "array: [3]u8@{x}\n", .{@ptrToInt(&value)}),
"array: {*}\n",
.{&value},
);
}
}
test "slice" {
{
const value: []const u8 = "abc";
try testFmt("slice: abc\n", "slice: {}\n", .{value});
}
{
var runtime_zero: usize = 0;
const value = @intToPtr([*]align(1) const []const u8, 0xdeadbeef)[runtime_zero..runtime_zero];
try testFmt("slice: []const u8@deadbeef\n", "slice: {}\n", .{value});
}
try testFmt("buf: Test \n", "buf: {s:5}\n", .{"Test"});
try testFmt("buf: Test\n Other text", "buf: {s}\n Other text", .{"Test"});
}
test "pointer" {
{
const value = @intToPtr(*align(1) i32, 0xdeadbeef);
try testFmt("pointer: i32@deadbeef\n", "pointer: {}\n", .{value});
try testFmt("pointer: i32@deadbeef\n", "pointer: {*}\n", .{value});
}
{
const value = @intToPtr(fn () void, 0xdeadbeef);
try testFmt("pointer: fn() void@deadbeef\n", "pointer: {}\n", .{value});
}
{
const value = @intToPtr(fn () void, 0xdeadbeef);
try testFmt("pointer: fn() void@deadbeef\n", "pointer: {}\n", .{value});
}
}
test "cstr" {
try testFmt(
"cstr: Test C\n",
"cstr: {s}\n",
.{@ptrCast([*c]const u8, "Test C")},
);
try testFmt(
"cstr: Test C \n",
"cstr: {s:10}\n",
.{@ptrCast([*c]const u8, "Test C")},
);
}
test "filesize" {
try testFmt("file size: 63MiB\n", "file size: {Bi}\n", .{@as(usize, 63 * 1024 * 1024)});
try testFmt("file size: 66.06MB\n", "file size: {B:.2}\n", .{@as(usize, 63 * 1024 * 1024)});
}
test "struct" {
{
const Struct = struct {
field: u8,
};
const value = Struct{ .field = 42 };
try testFmt("struct: Struct{ .field = 42 }\n", "struct: {}\n", .{value});
try testFmt("struct: Struct{ .field = 42 }\n", "struct: {}\n", .{&value});
}
{
const Struct = struct {
a: u0,
b: u1,
};
const value = Struct{ .a = 0, .b = 1 };
try testFmt("struct: Struct{ .a = 0, .b = 1 }\n", "struct: {}\n", .{value});
}
}
test "enum" {
const Enum = enum {
One,
Two,
};
const value = Enum.Two;
try testFmt("enum: Enum.Two\n", "enum: {}\n", .{value});
try testFmt("enum: Enum.Two\n", "enum: {}\n", .{&value});
try testFmt("enum: Enum.One\n", "enum: {x}\n", .{Enum.One});
try testFmt("enum: Enum.Two\n", "enum: {X}\n", .{Enum.Two});
// test very large enum to verify ct branch quota is large enough
try testFmt("enum: Win32Error.INVALID_FUNCTION\n", "enum: {}\n", .{std.os.windows.Win32Error.INVALID_FUNCTION});
}
test "non-exhaustive enum" {
const Enum = enum(u16) {
One = 0x000f,
Two = 0xbeef,
_,
};
try testFmt("enum: Enum.One\n", "enum: {}\n", .{Enum.One});
try testFmt("enum: Enum.Two\n", "enum: {}\n", .{Enum.Two});
try testFmt("enum: Enum(4660)\n", "enum: {}\n", .{@intToEnum(Enum, 0x1234)});
try testFmt("enum: Enum.One\n", "enum: {x}\n", .{Enum.One});
try testFmt("enum: Enum.Two\n", "enum: {x}\n", .{Enum.Two});
try testFmt("enum: Enum.Two\n", "enum: {X}\n", .{Enum.Two});
try testFmt("enum: Enum(1234)\n", "enum: {x}\n", .{@intToEnum(Enum, 0x1234)});
}
test "float.scientific" {
try testFmt("f32: 1.34000003e+00", "f32: {e}", .{@as(f32, 1.34)});
try testFmt("f32: 1.23400001e+01", "f32: {e}", .{@as(f32, 12.34)});
try testFmt("f64: -1.234e+11", "f64: {e}", .{@as(f64, -12.34e10)});
try testFmt("f64: 9.99996e-40", "f64: {e}", .{@as(f64, 9.999960e-40)});
}
test "float.scientific.precision" {
try testFmt("f64: 1.40971e-42", "f64: {e:.5}", .{@as(f64, 1.409706e-42)});
try testFmt("f64: 1.00000e-09", "f64: {e:.5}", .{@as(f64, @bitCast(f32, @as(u32, 814313563)))});
try testFmt("f64: 7.81250e-03", "f64: {e:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1006632960)))});
// libc rounds 1.000005e+05 to 1.00000e+05 but zig does 1.00001e+05.
// In fact, libc doesn't round a lot of 5 cases up when one past the precision point.
try testFmt("f64: 1.00001e+05", "f64: {e:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1203982400)))});
}
test "float.special" {
try testFmt("f64: nan", "f64: {}", .{math.nan_f64});
// negative nan is not defined by IEE 754,
// and ARM thus normalizes it to positive nan
if (builtin.arch != builtin.Arch.arm) {
try testFmt("f64: -nan", "f64: {}", .{-math.nan_f64});
}
try testFmt("f64: inf", "f64: {}", .{math.inf_f64});
try testFmt("f64: -inf", "f64: {}", .{-math.inf_f64});
}
test "float.decimal" {
try testFmt("f64: 152314000000000000000000000000", "f64: {d}", .{@as(f64, 1.52314e+29)});
try testFmt("f32: 0", "f32: {d}", .{@as(f32, 0.0)});
try testFmt("f32: 1.1", "f32: {d:.1}", .{@as(f32, 1.1234)});
try testFmt("f32: 1234.57", "f32: {d:.2}", .{@as(f32, 1234.567)});
// -11.1234 is converted to f64 -11.12339... internally (errol3() function takes f64).
// -11.12339... is rounded back up to -11.1234
try testFmt("f32: -11.1234", "f32: {d:.4}", .{@as(f32, -11.1234)});
try testFmt("f32: 91.12345", "f32: {d:.5}", .{@as(f32, 91.12345)});
try testFmt("f64: 91.1234567890", "f64: {d:.10}", .{@as(f64, 91.12345678901235)});
try testFmt("f64: 0.00000", "f64: {d:.5}", .{@as(f64, 0.0)});
try testFmt("f64: 6", "f64: {d:.0}", .{@as(f64, 5.700)});
try testFmt("f64: 10.0", "f64: {d:.1}", .{@as(f64, 9.999)});
try testFmt("f64: 1.000", "f64: {d:.3}", .{@as(f64, 1.0)});
try testFmt("f64: 0.00030000", "f64: {d:.8}", .{@as(f64, 0.0003)});
try testFmt("f64: 0.00000", "f64: {d:.5}", .{@as(f64, 1.40130e-45)});
try testFmt("f64: 0.00000", "f64: {d:.5}", .{@as(f64, 9.999960e-40)});
}
test "float.libc.sanity" {
try testFmt("f64: 0.00001", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 916964781)))});
try testFmt("f64: 0.00001", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 925353389)))});
try testFmt("f64: 0.10000", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1036831278)))});
try testFmt("f64: 1.00000", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1065353133)))});
try testFmt("f64: 10.00000", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1092616192)))});
// libc differences
//
// This is 0.015625 exactly according to gdb. We thus round down,
// however glibc rounds up for some reason. This occurs for all
// floats of the form x.yyyy25 on a precision point.
try testFmt("f64: 0.01563", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1015021568)))});
// errol3 rounds to ... 630 but libc rounds to ...632. Grisu3
// also rounds to 630 so I'm inclined to believe libc is not
// optimal here.
try testFmt("f64: 18014400656965630.00000", "f64: {d:.5}", .{@as(f64, @bitCast(f32, @as(u32, 1518338049)))});
}
test "custom" {
const Vec2 = struct {
const SelfType = @This();
x: f32,
y: f32,
pub fn format(
self: SelfType,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
if (fmt.len == 0 or comptime std.mem.eql(u8, fmt, "p")) {
return std.fmt.format(writer, "({d:.3},{d:.3})", .{ self.x, self.y });
} else if (comptime std.mem.eql(u8, fmt, "d")) {
return std.fmt.format(writer, "{d:.3}x{d:.3}", .{ self.x, self.y });
} else {
@compileError("Unknown format character: '" ++ fmt ++ "'");
}
}
};
var buf1: [32]u8 = undefined;
var value = Vec2{
.x = 10.2,
.y = 2.22,
};
try testFmt("point: (10.200,2.220)\n", "point: {}\n", .{&value});
try testFmt("dim: 10.200x2.220\n", "dim: {d}\n", .{&value});
// same thing but not passing a pointer
try testFmt("point: (10.200,2.220)\n", "point: {}\n", .{value});
try testFmt("dim: 10.200x2.220\n", "dim: {d}\n", .{value});
}
test "struct" {
const S = struct {
a: u32,
b: anyerror,
};
const inst = S{
.a = 456,
.b = error.Unused,
};
try testFmt("S{ .a = 456, .b = error.Unused }", "{}", .{inst});
}
test "union" {
const TU = union(enum) {
float: f32,
int: u32,
};
const UU = union {
float: f32,
int: u32,
};
const EU = extern union {
float: f32,
int: u32,
};
const tu_inst = TU{ .int = 123 };
const uu_inst = UU{ .int = 456 };
const eu_inst = EU{ .float = 321.123 };
try testFmt("TU{ .int = 123 }", "{}", .{tu_inst});
var buf: [100]u8 = undefined;
const uu_result = try bufPrint(buf[0..], "{}", .{uu_inst});
std.testing.expect(mem.eql(u8, uu_result[0..3], "UU@"));
const eu_result = try bufPrint(buf[0..], "{}", .{eu_inst});
std.testing.expect(mem.eql(u8, uu_result[0..3], "EU@"));
}
test "enum" {
const E = enum {
One,
Two,
Three,
};
const inst = E.Two;
try testFmt("E.Two", "{}", .{inst});
}
test "struct.self-referential" {
const S = struct {
const SelfType = @This();
a: ?*SelfType,
};
var inst = S{
.a = null,
};
inst.a = &inst;
try testFmt("S{ .a = S{ .a = S{ .a = S{ ... } } } }", "{}", .{inst});
}
test "struct.zero-size" {
const A = struct {
fn foo() void {}
};
const B = struct {
a: A,
c: i32,
};
const a = A{};
const b = B{ .a = a, .c = 0 };
try testFmt("B{ .a = A{ }, .c = 0 }", "{}", .{b});
}
test "bytes.hex" {
const some_bytes = "\xCA\xFE\xBA\xBE";
try testFmt("lowercase: cafebabe\n", "lowercase: {x}\n", .{some_bytes});
try testFmt("uppercase: CAFEBABE\n", "uppercase: {X}\n", .{some_bytes});
//Test Slices
try testFmt("uppercase: CAFE\n", "uppercase: {X}\n", .{some_bytes[0..2]});
try testFmt("lowercase: babe\n", "lowercase: {x}\n", .{some_bytes[2..]});
const bytes_with_zeros = "\x00\x0E\xBA\xBE";
try testFmt("lowercase: 000ebabe\n", "lowercase: {x}\n", .{bytes_with_zeros});
}
fn testFmt(expected: []const u8, comptime template: []const u8, args: var) !void {
var buf: [100]u8 = undefined;
const result = try bufPrint(buf[0..], template, args);
if (mem.eql(u8, result, expected)) return;
std.debug.warn("\n====== expected this output: =========\n", .{});
std.debug.warn("{}", .{expected});
std.debug.warn("\n======== instead found this: =========\n", .{});
std.debug.warn("{}", .{result});
std.debug.warn("\n======================================\n", .{});
return error.TestFailed;
}
pub fn trim(buf: []const u8) []const u8 {
var start: usize = 0;
while (start < buf.len and isWhiteSpace(buf[start])) : (start += 1) {}
var end: usize = buf.len;
while (true) {
if (end > start) {
const new_end = end - 1;
if (isWhiteSpace(buf[new_end])) {
end = new_end;
continue;
}
}
break;
}
return buf[start..end];
}
test "trim" {
std.testing.expect(mem.eql(u8, "abc", trim("\n abc \t")));
std.testing.expect(mem.eql(u8, "", trim(" ")));
std.testing.expect(mem.eql(u8, "", trim("")));
std.testing.expect(mem.eql(u8, "abc", trim(" abc")));
std.testing.expect(mem.eql(u8, "abc", trim("abc ")));
}
pub fn isWhiteSpace(byte: u8) bool {
return switch (byte) {
' ', '\t', '\n', '\r' => true,
else => false,
};
}
pub fn hexToBytes(out: []u8, input: []const u8) !void {
if (out.len * 2 < input.len)
return error.InvalidLength;
var in_i: usize = 0;
while (in_i != input.len) : (in_i += 2) {
const hi = try charToDigit(input[in_i], 16);
const lo = try charToDigit(input[in_i + 1], 16);
out[in_i / 2] = (hi << 4) | lo;
}
}
test "hexToBytes" {
const test_hex_str = "909A312BB12ED1F819B3521AC4C1E896F2160507FFC1C8381E3B07BB16BD1706";
var pb: [32]u8 = undefined;
try hexToBytes(pb[0..], test_hex_str);
try testFmt(test_hex_str, "{X}", .{pb});
}
test "formatIntValue with comptime_int" {
const value: comptime_int = 123456789123456789;
var buf: [20]u8 = undefined;
var fbs = std.io.fixedBufferStream(&buf);
try formatIntValue(value, "", FormatOptions{}, fbs.writer());
std.testing.expect(mem.eql(u8, fbs.getWritten(), "123456789123456789"));
}
test "formatFloatValue with comptime_float" {
const value: comptime_float = 1.0;
var buf: [20]u8 = undefined;
var fbs = std.io.fixedBufferStream(&buf);
try formatFloatValue(value, "", FormatOptions{}, fbs.writer());
std.testing.expect(mem.eql(u8, fbs.getWritten(), "1.0e+00"));
try testFmt("1.0e+00", "{}", .{value});
try testFmt("1.0e+00", "{}", .{1.0});
}
test "formatType max_depth" {
const Vec2 = struct {
const SelfType = @This();
x: f32,
y: f32,
pub fn format(
self: SelfType,
comptime fmt: []const u8,
options: FormatOptions,
writer: var,
) !void {
if (fmt.len == 0) {
return std.fmt.format(writer, "({d:.3},{d:.3})", .{ self.x, self.y });
} else {
@compileError("Unknown format string: '" ++ fmt ++ "'");
}
}
};
const E = enum {
One,
Two,
Three,
};
const TU = union(enum) {
const SelfType = @This();
float: f32,
int: u32,
ptr: ?*SelfType,
};
const S = struct {
const SelfType = @This();
a: ?*SelfType,
tu: TU,
e: E,
vec: Vec2,
};
var inst = S{
.a = null,
.tu = TU{ .ptr = null },
.e = E.Two,
.vec = Vec2{ .x = 10.2, .y = 2.22 },
};
inst.a = &inst;
inst.tu.ptr = &inst.tu;
var buf: [1000]u8 = undefined;
var fbs = std.io.fixedBufferStream(&buf);
try formatType(inst, "", FormatOptions{}, fbs.writer(), 0);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "S{ ... }"));
fbs.reset();
try formatType(inst, "", FormatOptions{}, fbs.writer(), 1);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "S{ .a = S{ ... }, .tu = TU{ ... }, .e = E.Two, .vec = (10.200,2.220) }"));
fbs.reset();
try formatType(inst, "", FormatOptions{}, fbs.writer(), 2);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "S{ .a = S{ .a = S{ ... }, .tu = TU{ ... }, .e = E.Two, .vec = (10.200,2.220) }, .tu = TU{ .ptr = TU{ ... } }, .e = E.Two, .vec = (10.200,2.220) }"));
fbs.reset();
try formatType(inst, "", FormatOptions{}, fbs.writer(), 3);
std.testing.expect(mem.eql(u8, fbs.getWritten(), "S{ .a = S{ .a = S{ .a = S{ ... }, .tu = TU{ ... }, .e = E.Two, .vec = (10.200,2.220) }, .tu = TU{ .ptr = TU{ ... } }, .e = E.Two, .vec = (10.200,2.220) }, .tu = TU{ .ptr = TU{ .ptr = TU{ ... } } }, .e = E.Two, .vec = (10.200,2.220) }"));
}
test "positional" {
try testFmt("2 1 0", "{2} {1} {0}", .{ @as(usize, 0), @as(usize, 1), @as(usize, 2) });
try testFmt("2 1 0", "{2} {1} {}", .{ @as(usize, 0), @as(usize, 1), @as(usize, 2) });
try testFmt("0 0", "{0} {0}", .{@as(usize, 0)});
try testFmt("0 1", "{} {1}", .{ @as(usize, 0), @as(usize, 1) });
try testFmt("1 0 0 1", "{1} {} {0} {}", .{ @as(usize, 0), @as(usize, 1) });
}
test "positional with specifier" {
try testFmt("10.0", "{0d:.1}", .{@as(f64, 9.999)});
}
test "positional/alignment/width/precision" {
try testFmt("10.0", "{0d: >3.1}", .{@as(f64, 9.999)});
}
test "vector" {
if (builtin.arch == .mipsel or builtin.arch == .mips) {
// https://github.com/ziglang/zig/issues/3317
return error.SkipZigTest;
}
if (builtin.arch == .riscv64) {
// https://github.com/ziglang/zig/issues/4486
return error.SkipZigTest;
}
if (builtin.arch == .wasm32) {
// https://github.com/ziglang/zig/issues/5339
return error.SkipZigTest;
}
const vbool: std.meta.Vector(4, bool) = [_]bool{ true, false, true, false };
const vi64: std.meta.Vector(4, i64) = [_]i64{ -2, -1, 0, 1 };
const vu64: std.meta.Vector(4, u64) = [_]u64{ 1000, 2000, 3000, 4000 };
try testFmt("{ true, false, true, false }", "{}", .{vbool});
try testFmt("{ -2, -1, 0, 1 }", "{}", .{vi64});
try testFmt("{ - 2, - 1, + 0, + 1 }", "{d:5}", .{vi64});
try testFmt("{ 1000, 2000, 3000, 4000 }", "{}", .{vu64});
try testFmt("{ 3e8, 7d0, bb8, fa0 }", "{x}", .{vu64});
try testFmt("{ 1kB, 2kB, 3kB, 4kB }", "{B}", .{vu64});
try testFmt("{ 1000B, 1.953125KiB, 2.9296875KiB, 3.90625KiB }", "{Bi}", .{vu64});
}
test "enum-literal" {
try testFmt(".hello_world", "{}", .{.hello_world});
}
test "padding" {
try testFmt("Simple", "{}", .{"Simple"});
try testFmt("true ", "{:10}", .{true});
try testFmt(" true", "{:>10}", .{true});
try testFmt("======true", "{:=>10}", .{true});
try testFmt("true======", "{:=<10}", .{true});
try testFmt(" true ", "{:^10}", .{true});
try testFmt("===true===", "{:=^10}", .{true});
try testFmt("Minimum width", "{:18} width", .{"Minimum"});
try testFmt("==================Filled", "{:=>24}", .{"Filled"});
try testFmt(" Centered ", "{:^24}", .{"Centered"});
}