zig/lib/std/unicode.zig
2020-03-07 19:21:28 -05:00

700 lines
24 KiB
Zig

const std = @import("./std.zig");
const builtin = @import("builtin");
const assert = std.debug.assert;
const testing = std.testing;
const mem = std.mem;
/// Returns how many bytes the UTF-8 representation would require
/// for the given codepoint.
pub fn utf8CodepointSequenceLength(c: u21) !u3 {
if (c < 0x80) return @as(u3, 1);
if (c < 0x800) return @as(u3, 2);
if (c < 0x10000) return @as(u3, 3);
if (c < 0x110000) return @as(u3, 4);
return error.CodepointTooLarge;
}
/// Given the first byte of a UTF-8 codepoint,
/// returns a number 1-4 indicating the total length of the codepoint in bytes.
/// If this byte does not match the form of a UTF-8 start byte, returns Utf8InvalidStartByte.
pub fn utf8ByteSequenceLength(first_byte: u8) !u3 {
return switch (@clz(u8, ~first_byte)) {
0 => 1,
2 => 2,
3 => 3,
4 => 4,
else => error.Utf8InvalidStartByte,
};
}
/// Encodes the given codepoint into a UTF-8 byte sequence.
/// c: the codepoint.
/// out: the out buffer to write to. Must have a len >= utf8CodepointSequenceLength(c).
/// Errors: if c cannot be encoded in UTF-8.
/// Returns: the number of bytes written to out.
pub fn utf8Encode(c: u21, out: []u8) !u3 {
const length = try utf8CodepointSequenceLength(c);
assert(out.len >= length);
switch (length) {
// The pattern for each is the same
// - Increasing the initial shift by 6 each time
// - Each time after the first shorten the shifted
// value to a max of 0b111111 (63)
1 => out[0] = @intCast(u8, c), // Can just do 0 + codepoint for initial range
2 => {
out[0] = @intCast(u8, 0b11000000 | (c >> 6));
out[1] = @intCast(u8, 0b10000000 | (c & 0b111111));
},
3 => {
if (0xd800 <= c and c <= 0xdfff) return error.Utf8CannotEncodeSurrogateHalf;
out[0] = @intCast(u8, 0b11100000 | (c >> 12));
out[1] = @intCast(u8, 0b10000000 | ((c >> 6) & 0b111111));
out[2] = @intCast(u8, 0b10000000 | (c & 0b111111));
},
4 => {
out[0] = @intCast(u8, 0b11110000 | (c >> 18));
out[1] = @intCast(u8, 0b10000000 | ((c >> 12) & 0b111111));
out[2] = @intCast(u8, 0b10000000 | ((c >> 6) & 0b111111));
out[3] = @intCast(u8, 0b10000000 | (c & 0b111111));
},
else => unreachable,
}
return length;
}
const Utf8DecodeError = Utf8Decode2Error || Utf8Decode3Error || Utf8Decode4Error;
/// Decodes the UTF-8 codepoint encoded in the given slice of bytes.
/// bytes.len must be equal to utf8ByteSequenceLength(bytes[0]) catch unreachable.
/// If you already know the length at comptime, you can call one of
/// utf8Decode2,utf8Decode3,utf8Decode4 directly instead of this function.
pub fn utf8Decode(bytes: []const u8) Utf8DecodeError!u21 {
return switch (bytes.len) {
1 => @as(u21, bytes[0]),
2 => utf8Decode2(bytes),
3 => utf8Decode3(bytes),
4 => utf8Decode4(bytes),
else => unreachable,
};
}
const Utf8Decode2Error = error{
Utf8ExpectedContinuation,
Utf8OverlongEncoding,
};
pub fn utf8Decode2(bytes: []const u8) Utf8Decode2Error!u21 {
assert(bytes.len == 2);
assert(bytes[0] & 0b11100000 == 0b11000000);
var value: u21 = bytes[0] & 0b00011111;
if (bytes[1] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[1] & 0b00111111;
if (value < 0x80) return error.Utf8OverlongEncoding;
return value;
}
const Utf8Decode3Error = error{
Utf8ExpectedContinuation,
Utf8OverlongEncoding,
Utf8EncodesSurrogateHalf,
};
pub fn utf8Decode3(bytes: []const u8) Utf8Decode3Error!u21 {
assert(bytes.len == 3);
assert(bytes[0] & 0b11110000 == 0b11100000);
var value: u21 = bytes[0] & 0b00001111;
if (bytes[1] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[1] & 0b00111111;
if (bytes[2] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[2] & 0b00111111;
if (value < 0x800) return error.Utf8OverlongEncoding;
if (0xd800 <= value and value <= 0xdfff) return error.Utf8EncodesSurrogateHalf;
return value;
}
const Utf8Decode4Error = error{
Utf8ExpectedContinuation,
Utf8OverlongEncoding,
Utf8CodepointTooLarge,
};
pub fn utf8Decode4(bytes: []const u8) Utf8Decode4Error!u21 {
assert(bytes.len == 4);
assert(bytes[0] & 0b11111000 == 0b11110000);
var value: u21 = bytes[0] & 0b00000111;
if (bytes[1] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[1] & 0b00111111;
if (bytes[2] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[2] & 0b00111111;
if (bytes[3] & 0b11000000 != 0b10000000) return error.Utf8ExpectedContinuation;
value <<= 6;
value |= bytes[3] & 0b00111111;
if (value < 0x10000) return error.Utf8OverlongEncoding;
if (value > 0x10FFFF) return error.Utf8CodepointTooLarge;
return value;
}
pub fn utf8ValidateSlice(s: []const u8) bool {
var i: usize = 0;
while (i < s.len) {
if (utf8ByteSequenceLength(s[i])) |cp_len| {
if (i + cp_len > s.len) {
return false;
}
if (utf8Decode(s[i .. i + cp_len])) |_| {} else |_| {
return false;
}
i += cp_len;
} else |err| {
return false;
}
}
return true;
}
/// Utf8View iterates the code points of a utf-8 encoded string.
///
/// ```
/// var utf8 = (try std.unicode.Utf8View.init("hi there")).iterator();
/// while (utf8.nextCodepointSlice()) |codepoint| {
/// std.debug.warn("got codepoint {}\n", .{codepoint});
/// }
/// ```
pub const Utf8View = struct {
bytes: []const u8,
pub fn init(s: []const u8) !Utf8View {
if (!utf8ValidateSlice(s)) {
return error.InvalidUtf8;
}
return initUnchecked(s);
}
pub fn initUnchecked(s: []const u8) Utf8View {
return Utf8View{ .bytes = s };
}
/// TODO: https://github.com/ziglang/zig/issues/425
pub fn initComptime(comptime s: []const u8) Utf8View {
if (comptime init(s)) |r| {
return r;
} else |err| switch (err) {
error.InvalidUtf8 => {
@compileError("invalid utf8");
unreachable;
},
}
}
pub fn iterator(s: Utf8View) Utf8Iterator {
return Utf8Iterator{
.bytes = s.bytes,
.i = 0,
};
}
};
pub const Utf8Iterator = struct {
bytes: []const u8,
i: usize,
pub fn nextCodepointSlice(it: *Utf8Iterator) ?[]const u8 {
if (it.i >= it.bytes.len) {
return null;
}
const cp_len = utf8ByteSequenceLength(it.bytes[it.i]) catch unreachable;
it.i += cp_len;
return it.bytes[it.i - cp_len .. it.i];
}
pub fn nextCodepoint(it: *Utf8Iterator) ?u21 {
const slice = it.nextCodepointSlice() orelse return null;
switch (slice.len) {
1 => return @as(u21, slice[0]),
2 => return utf8Decode2(slice) catch unreachable,
3 => return utf8Decode3(slice) catch unreachable,
4 => return utf8Decode4(slice) catch unreachable,
else => unreachable,
}
}
};
pub const Utf16LeIterator = struct {
bytes: []const u8,
i: usize,
pub fn init(s: []const u16) Utf16LeIterator {
return Utf16LeIterator{
.bytes = mem.sliceAsBytes(s),
.i = 0,
};
}
pub fn nextCodepoint(it: *Utf16LeIterator) !?u21 {
assert(it.i <= it.bytes.len);
if (it.i == it.bytes.len) return null;
const c0: u21 = mem.readIntSliceLittle(u16, it.bytes[it.i .. it.i + 2]);
if (c0 & ~@as(u21, 0x03ff) == 0xd800) {
// surrogate pair
it.i += 2;
if (it.i >= it.bytes.len) return error.DanglingSurrogateHalf;
const c1: u21 = mem.readIntSliceLittle(u16, it.bytes[it.i .. it.i + 2]);
if (c1 & ~@as(u21, 0x03ff) != 0xdc00) return error.ExpectedSecondSurrogateHalf;
it.i += 2;
return 0x10000 + (((c0 & 0x03ff) << 10) | (c1 & 0x03ff));
} else if (c0 & ~@as(u21, 0x03ff) == 0xdc00) {
return error.UnexpectedSecondSurrogateHalf;
} else {
it.i += 2;
return c0;
}
}
};
test "utf8 encode" {
comptime testUtf8Encode() catch unreachable;
try testUtf8Encode();
}
fn testUtf8Encode() !void {
// A few taken from wikipedia a few taken elsewhere
var array: [4]u8 = undefined;
testing.expect((try utf8Encode(try utf8Decode(""), array[0..])) == 3);
testing.expect(array[0] == 0b11100010);
testing.expect(array[1] == 0b10000010);
testing.expect(array[2] == 0b10101100);
testing.expect((try utf8Encode(try utf8Decode("$"), array[0..])) == 1);
testing.expect(array[0] == 0b00100100);
testing.expect((try utf8Encode(try utf8Decode("¢"), array[0..])) == 2);
testing.expect(array[0] == 0b11000010);
testing.expect(array[1] == 0b10100010);
testing.expect((try utf8Encode(try utf8Decode("𐍈"), array[0..])) == 4);
testing.expect(array[0] == 0b11110000);
testing.expect(array[1] == 0b10010000);
testing.expect(array[2] == 0b10001101);
testing.expect(array[3] == 0b10001000);
}
test "utf8 encode error" {
comptime testUtf8EncodeError();
testUtf8EncodeError();
}
fn testUtf8EncodeError() void {
var array: [4]u8 = undefined;
testErrorEncode(0xd800, array[0..], error.Utf8CannotEncodeSurrogateHalf);
testErrorEncode(0xdfff, array[0..], error.Utf8CannotEncodeSurrogateHalf);
testErrorEncode(0x110000, array[0..], error.CodepointTooLarge);
testErrorEncode(0x1fffff, array[0..], error.CodepointTooLarge);
}
fn testErrorEncode(codePoint: u21, array: []u8, expectedErr: anyerror) void {
testing.expectError(expectedErr, utf8Encode(codePoint, array));
}
test "utf8 iterator on ascii" {
comptime testUtf8IteratorOnAscii();
testUtf8IteratorOnAscii();
}
fn testUtf8IteratorOnAscii() void {
const s = Utf8View.initComptime("abc");
var it1 = s.iterator();
testing.expect(std.mem.eql(u8, "a", it1.nextCodepointSlice().?));
testing.expect(std.mem.eql(u8, "b", it1.nextCodepointSlice().?));
testing.expect(std.mem.eql(u8, "c", it1.nextCodepointSlice().?));
testing.expect(it1.nextCodepointSlice() == null);
var it2 = s.iterator();
testing.expect(it2.nextCodepoint().? == 'a');
testing.expect(it2.nextCodepoint().? == 'b');
testing.expect(it2.nextCodepoint().? == 'c');
testing.expect(it2.nextCodepoint() == null);
}
test "utf8 view bad" {
comptime testUtf8ViewBad();
testUtf8ViewBad();
}
fn testUtf8ViewBad() void {
// Compile-time error.
// const s3 = Utf8View.initComptime("\xfe\xf2");
testing.expectError(error.InvalidUtf8, Utf8View.init("hel\xadlo"));
}
test "utf8 view ok" {
comptime testUtf8ViewOk();
testUtf8ViewOk();
}
fn testUtf8ViewOk() void {
const s = Utf8View.initComptime("東京市");
var it1 = s.iterator();
testing.expect(std.mem.eql(u8, "", it1.nextCodepointSlice().?));
testing.expect(std.mem.eql(u8, "", it1.nextCodepointSlice().?));
testing.expect(std.mem.eql(u8, "", it1.nextCodepointSlice().?));
testing.expect(it1.nextCodepointSlice() == null);
var it2 = s.iterator();
testing.expect(it2.nextCodepoint().? == 0x6771);
testing.expect(it2.nextCodepoint().? == 0x4eac);
testing.expect(it2.nextCodepoint().? == 0x5e02);
testing.expect(it2.nextCodepoint() == null);
}
test "bad utf8 slice" {
comptime testBadUtf8Slice();
testBadUtf8Slice();
}
fn testBadUtf8Slice() void {
testing.expect(utf8ValidateSlice("abc"));
testing.expect(!utf8ValidateSlice("abc\xc0"));
testing.expect(!utf8ValidateSlice("abc\xc0abc"));
testing.expect(utf8ValidateSlice("abc\xdf\xbf"));
}
test "valid utf8" {
comptime testValidUtf8();
testValidUtf8();
}
fn testValidUtf8() void {
testValid("\x00", 0x0);
testValid("\x20", 0x20);
testValid("\x7f", 0x7f);
testValid("\xc2\x80", 0x80);
testValid("\xdf\xbf", 0x7ff);
testValid("\xe0\xa0\x80", 0x800);
testValid("\xe1\x80\x80", 0x1000);
testValid("\xef\xbf\xbf", 0xffff);
testValid("\xf0\x90\x80\x80", 0x10000);
testValid("\xf1\x80\x80\x80", 0x40000);
testValid("\xf3\xbf\xbf\xbf", 0xfffff);
testValid("\xf4\x8f\xbf\xbf", 0x10ffff);
}
test "invalid utf8 continuation bytes" {
comptime testInvalidUtf8ContinuationBytes();
testInvalidUtf8ContinuationBytes();
}
fn testInvalidUtf8ContinuationBytes() void {
// unexpected continuation
testError("\x80", error.Utf8InvalidStartByte);
testError("\xbf", error.Utf8InvalidStartByte);
// too many leading 1's
testError("\xf8", error.Utf8InvalidStartByte);
testError("\xff", error.Utf8InvalidStartByte);
// expected continuation for 2 byte sequences
testError("\xc2", error.UnexpectedEof);
testError("\xc2\x00", error.Utf8ExpectedContinuation);
testError("\xc2\xc0", error.Utf8ExpectedContinuation);
// expected continuation for 3 byte sequences
testError("\xe0", error.UnexpectedEof);
testError("\xe0\x00", error.UnexpectedEof);
testError("\xe0\xc0", error.UnexpectedEof);
testError("\xe0\xa0", error.UnexpectedEof);
testError("\xe0\xa0\x00", error.Utf8ExpectedContinuation);
testError("\xe0\xa0\xc0", error.Utf8ExpectedContinuation);
// expected continuation for 4 byte sequences
testError("\xf0", error.UnexpectedEof);
testError("\xf0\x00", error.UnexpectedEof);
testError("\xf0\xc0", error.UnexpectedEof);
testError("\xf0\x90\x00", error.UnexpectedEof);
testError("\xf0\x90\xc0", error.UnexpectedEof);
testError("\xf0\x90\x80\x00", error.Utf8ExpectedContinuation);
testError("\xf0\x90\x80\xc0", error.Utf8ExpectedContinuation);
}
test "overlong utf8 codepoint" {
comptime testOverlongUtf8Codepoint();
testOverlongUtf8Codepoint();
}
fn testOverlongUtf8Codepoint() void {
testError("\xc0\x80", error.Utf8OverlongEncoding);
testError("\xc1\xbf", error.Utf8OverlongEncoding);
testError("\xe0\x80\x80", error.Utf8OverlongEncoding);
testError("\xe0\x9f\xbf", error.Utf8OverlongEncoding);
testError("\xf0\x80\x80\x80", error.Utf8OverlongEncoding);
testError("\xf0\x8f\xbf\xbf", error.Utf8OverlongEncoding);
}
test "misc invalid utf8" {
comptime testMiscInvalidUtf8();
testMiscInvalidUtf8();
}
fn testMiscInvalidUtf8() void {
// codepoint out of bounds
testError("\xf4\x90\x80\x80", error.Utf8CodepointTooLarge);
testError("\xf7\xbf\xbf\xbf", error.Utf8CodepointTooLarge);
// surrogate halves
testValid("\xed\x9f\xbf", 0xd7ff);
testError("\xed\xa0\x80", error.Utf8EncodesSurrogateHalf);
testError("\xed\xbf\xbf", error.Utf8EncodesSurrogateHalf);
testValid("\xee\x80\x80", 0xe000);
}
fn testError(bytes: []const u8, expected_err: anyerror) void {
testing.expectError(expected_err, testDecode(bytes));
}
fn testValid(bytes: []const u8, expected_codepoint: u21) void {
testing.expect((testDecode(bytes) catch unreachable) == expected_codepoint);
}
fn testDecode(bytes: []const u8) !u21 {
const length = try utf8ByteSequenceLength(bytes[0]);
if (bytes.len < length) return error.UnexpectedEof;
testing.expect(bytes.len == length);
return utf8Decode(bytes);
}
/// Caller must free returned memory.
pub fn utf16leToUtf8Alloc(allocator: *mem.Allocator, utf16le: []const u16) ![]u8 {
var result = std.ArrayList(u8).init(allocator);
// optimistically guess that it will all be ascii.
try result.ensureCapacity(utf16le.len);
var out_index: usize = 0;
var it = Utf16LeIterator.init(utf16le);
while (try it.nextCodepoint()) |codepoint| {
const utf8_len = utf8CodepointSequenceLength(codepoint) catch unreachable;
try result.resize(result.len + utf8_len);
assert((utf8Encode(codepoint, result.items[out_index..]) catch unreachable) == utf8_len);
out_index += utf8_len;
}
return result.toOwnedSlice();
}
/// Asserts that the output buffer is big enough.
/// Returns end byte index into utf8.
pub fn utf16leToUtf8(utf8: []u8, utf16le: []const u16) !usize {
var end_index: usize = 0;
var it = Utf16LeIterator.init(utf16le);
while (try it.nextCodepoint()) |codepoint| {
end_index += try utf8Encode(codepoint, utf8[end_index..]);
}
return end_index;
}
test "utf16leToUtf8" {
var utf16le: [2]u16 = undefined;
const utf16le_as_bytes = mem.sliceAsBytes(utf16le[0..]);
{
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 'A');
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 'a');
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "Aa"));
}
{
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 0x80);
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 0xffff);
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "\xc2\x80" ++ "\xef\xbf\xbf"));
}
{
// the values just outside the surrogate half range
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 0xd7ff);
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 0xe000);
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "\xed\x9f\xbf" ++ "\xee\x80\x80"));
}
{
// smallest surrogate pair
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 0xd800);
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 0xdc00);
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "\xf0\x90\x80\x80"));
}
{
// largest surrogate pair
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 0xdbff);
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 0xdfff);
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "\xf4\x8f\xbf\xbf"));
}
{
mem.writeIntSliceLittle(u16, utf16le_as_bytes[0..], 0xdbff);
mem.writeIntSliceLittle(u16, utf16le_as_bytes[2..], 0xdc00);
const utf8 = try utf16leToUtf8Alloc(std.testing.allocator, &utf16le);
defer std.testing.allocator.free(utf8);
testing.expect(mem.eql(u8, utf8, "\xf4\x8f\xb0\x80"));
}
}
pub fn utf8ToUtf16LeWithNull(allocator: *mem.Allocator, utf8: []const u8) ![:0]u16 {
var result = std.ArrayList(u16).init(allocator);
// optimistically guess that it will not require surrogate pairs
try result.ensureCapacity(utf8.len + 1);
const view = try Utf8View.init(utf8);
var it = view.iterator();
while (it.nextCodepoint()) |codepoint| {
if (codepoint < 0x10000) {
const short = @intCast(u16, codepoint);
try result.append(mem.nativeToLittle(u16, short));
} else {
const high = @intCast(u16, (codepoint - 0x10000) >> 10) + 0xD800;
const low = @intCast(u16, codepoint & 0x3FF) + 0xDC00;
var out: [2]u16 = undefined;
out[0] = mem.nativeToLittle(u16, high);
out[1] = mem.nativeToLittle(u16, low);
try result.appendSlice(out[0..]);
}
}
const len = result.len;
try result.append(0);
return result.toOwnedSlice()[0..len :0];
}
/// Returns index of next character. If exact fit, returned index equals output slice length.
/// Assumes there is enough space for the output.
pub fn utf8ToUtf16Le(utf16le: []u16, utf8: []const u8) !usize {
var dest_i: usize = 0;
var src_i: usize = 0;
while (src_i < utf8.len) {
const n = utf8ByteSequenceLength(utf8[src_i]) catch return error.InvalidUtf8;
const next_src_i = src_i + n;
const codepoint = utf8Decode(utf8[src_i..next_src_i]) catch return error.InvalidUtf8;
if (codepoint < 0x10000) {
const short = @intCast(u16, codepoint);
utf16le[dest_i] = mem.nativeToLittle(u16, short);
dest_i += 1;
} else {
const high = @intCast(u16, (codepoint - 0x10000) >> 10) + 0xD800;
const low = @intCast(u16, codepoint & 0x3FF) + 0xDC00;
utf16le[dest_i] = mem.nativeToLittle(u16, high);
utf16le[dest_i + 1] = mem.nativeToLittle(u16, low);
dest_i += 2;
}
src_i = next_src_i;
}
return dest_i;
}
test "utf8ToUtf16Le" {
var utf16le: [2]u16 = [_]u16{0} ** 2;
{
const length = try utf8ToUtf16Le(utf16le[0..], "𐐷");
testing.expectEqual(@as(usize, 2), length);
testing.expectEqualSlices(u8, "\x01\xd8\x37\xdc", mem.sliceAsBytes(utf16le[0..]));
}
{
const length = try utf8ToUtf16Le(utf16le[0..], "\u{10FFFF}");
testing.expectEqual(@as(usize, 2), length);
testing.expectEqualSlices(u8, "\xff\xdb\xff\xdf", mem.sliceAsBytes(utf16le[0..]));
}
}
test "utf8ToUtf16LeWithNull" {
{
const utf16 = try utf8ToUtf16LeWithNull(testing.allocator, "𐐷");
defer testing.allocator.free(utf16);
testing.expectEqualSlices(u8, "\x01\xd8\x37\xdc", mem.sliceAsBytes(utf16[0..]));
testing.expect(utf16[2] == 0);
}
{
const utf16 = try utf8ToUtf16LeWithNull(testing.allocator, "\u{10FFFF}");
defer testing.allocator.free(utf16);
testing.expectEqualSlices(u8, "\xff\xdb\xff\xdf", mem.sliceAsBytes(utf16[0..]));
testing.expect(utf16[2] == 0);
}
}
/// Converts a UTF-8 string literal into a UTF-16LE string literal.
pub fn utf8ToUtf16LeStringLiteral(comptime utf8: []const u8) *const [calcUtf16LeLen(utf8) :0] u16 {
comptime {
const len: usize = calcUtf16LeLen(utf8);
var utf16le: [len :0]u16 = [_ :0]u16{0} ** len;
const utf16le_len = utf8ToUtf16Le(&utf16le, utf8[0..]) catch |err| @compileError(err);
assert(len == utf16le_len);
return &utf16le;
}
}
/// Returns length of a supplied UTF-8 string literal. Asserts that the data is valid UTF-8.
fn calcUtf16LeLen(utf8: []const u8) usize {
var src_i: usize = 0;
var dest_len: usize = 0;
while (src_i < utf8.len) {
const n = utf8ByteSequenceLength(utf8[src_i]) catch unreachable;
const next_src_i = src_i + n;
const codepoint = utf8Decode(utf8[src_i..next_src_i]) catch unreachable;
if (codepoint < 0x10000) {
dest_len += 1;
} else {
dest_len += 2;
}
src_i = next_src_i;
}
return dest_len;
}
test "utf8ToUtf16LeStringLiteral" {
{
const bytes = [_:0]u16{ 0x41 };
const utf16 = utf8ToUtf16LeStringLiteral("A");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[1] == 0);
}
{
const bytes = [_:0]u16{ 0xD801, 0xDC37 };
const utf16 = utf8ToUtf16LeStringLiteral("𐐷");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[2] == 0);
}
{
const bytes = [_:0]u16{ 0x02FF };
const utf16 = utf8ToUtf16LeStringLiteral("\u{02FF}");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[1] == 0);
}
{
const bytes = [_:0]u16{ 0x7FF };
const utf16 = utf8ToUtf16LeStringLiteral("\u{7FF}");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[1] == 0);
}
{
const bytes = [_:0]u16{ 0x801 };
const utf16 = utf8ToUtf16LeStringLiteral("\u{801}");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[1] == 0);
}
{
const bytes = [_:0]u16{ 0xDBFF, 0xDFFF };
const utf16 = utf8ToUtf16LeStringLiteral("\u{10FFFF}");
testing.expectEqualSlices(u16, &bytes, utf16);
testing.expect(utf16[2] == 0);
}
}