mirrors/ladybird
0
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2025-08-29 13:46:31 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

AK+LibJS: Extract some UTF-16 helpers for use in an outside class

Browse source 
An upcoming Utf16String will need access to these helpers. Let's make
them publicly available.
This commit is contained in:
Timothy Flynn 2025-06-12 17:39:05 -04:00 • committed by Tim Flynn
commit 66006d3812
Notes: github-actions[bot] 2025-07-03 13:54:12 +00:00
10 changed files with 121 additions and 86 deletions
Download patch file Download diff file Expand all files Collapse all files

76
AK/UnicodeUtils.h
View file

@ -78,6 +78,82 @@ template<FallibleFunction<char> Callback>
return -1;
}
constexpr inline u16 HIGH_SURROGATE_MIN = 0xd800;
constexpr inline u16 HIGH_SURROGATE_MAX = 0xdbff;
constexpr inline u16 LOW_SURROGATE_MIN = 0xdc00;
constexpr inline u16 LOW_SURROGATE_MAX = 0xdfff;
constexpr inline u32 REPLACEMENT_CODE_POINT = 0xfffd;
constexpr inline u32 FIRST_SUPPLEMENTARY_PLANE_CODE_POINT = 0x10000;
[[nodiscard]] constexpr size_t code_unit_length_for_code_point(u32 code_point)
{
return code_point < FIRST_SUPPLEMENTARY_PLANE_CODE_POINT ? 1uz : 2uz;
}
[[nodiscard]] constexpr bool is_utf16_high_surrogate(u16 code_unit)
{
return (code_unit >= HIGH_SURROGATE_MIN) && (code_unit <= HIGH_SURROGATE_MAX);
}
[[nodiscard]] constexpr bool is_utf16_low_surrogate(u16 code_unit)
{
return (code_unit >= LOW_SURROGATE_MIN) && (code_unit <= LOW_SURROGATE_MAX);
}
[[nodiscard]] constexpr u32 decode_utf16_surrogate_pair(u16 high_surrogate, u16 low_surrogate)
{
VERIFY(is_utf16_high_surrogate(high_surrogate));
VERIFY(is_utf16_low_surrogate(low_surrogate));
return ((high_surrogate - HIGH_SURROGATE_MIN) << 10) + (low_surrogate - LOW_SURROGATE_MIN) + FIRST_SUPPLEMENTARY_PLANE_CODE_POINT;
}
template<typename Callback>
[[nodiscard]] constexpr size_t code_point_to_utf16(u32 code_point, Callback callback)
{
if (code_point < FIRST_SUPPLEMENTARY_PLANE_CODE_POINT) {
callback(static_cast<char16_t>(code_point));
return 1uz;
}
if (code_point <= 0x10ffff) {
code_point -= FIRST_SUPPLEMENTARY_PLANE_CODE_POINT;
auto code_unit = static_cast<u16>(HIGH_SURROGATE_MIN | (code_point >> 10));
callback(static_cast<char16_t>(code_unit));
code_unit = static_cast<u16>(LOW_SURROGATE_MIN | (code_point & 0x3ff));
callback(static_cast<char16_t>(code_unit));
return 2uz;
}
VERIFY_NOT_REACHED();
}
template<FallibleFunction<char16_t> Callback>
constexpr ErrorOr<size_t> try_code_point_to_utf16(u32 code_point, Callback callback)
{
if (code_point < FIRST_SUPPLEMENTARY_PLANE_CODE_POINT) {
TRY(callback(static_cast<char16_t>(code_point)));
return 1uz;
}
if (code_point <= 0x10ffff) {
code_point -= FIRST_SUPPLEMENTARY_PLANE_CODE_POINT;
auto code_unit = static_cast<u16>(HIGH_SURROGATE_MIN | (code_point >> 10));
TRY(callback(static_cast<char16_t>(code_unit)));
code_unit = static_cast<u16>(LOW_SURROGATE_MIN | (code_point & 0x3ff));
TRY(callback(static_cast<char16_t>(code_unit)));
return 2uz;
}
VERIFY_NOT_REACHED();
}
/**
* Compute the maximum number of UTF-8 bytes needed to store a given UTF-16 string, accounting for unmatched UTF-16 surrogates.
* This function will overcount by at most 33%; 2 bytes for every valid UTF-16 codepoint between U+100000 and U+10FFFF.