RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 13:38:11 +00:00
Code Issues Activity Actions

LibRegex: Support property escapes of Unicode script extensions

Browse source
This commit is contained in:
Timothy Flynn 2021-08-04 07:26:25 -04:00 • committed by Linus Groh
parent 5edd458420
commit 484ccfadc3
5 changed files with 51 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

20
Userland/Libraries/LibRegex/RegexByteCode.cpp
View file

@ -573,6 +573,10 @@ ALWAYS_INLINE ExecutionResult OpCode_Compare::execute(MatchInput const& input, M
auto script = static_cast<Unicode::Script>(m_bytecode->at(offset++));
compare_script(input, state, script, current_inversion_state(), inverse_matched);
} else if (compare_type == CharacterCompareType::ScriptExtension) {
auto script = static_cast<Unicode::Script>(m_bytecode->at(offset++));
compare_script_extension(input, state, script, current_inversion_state(), inverse_matched);
} else {
warnln("Undefined comparison: {}", (int)compare_type);
VERIFY_NOT_REACHED();
@ -810,6 +814,22 @@ ALWAYS_INLINE void OpCode_Compare::compare_script(MatchInput const& input, Match
}
}
ALWAYS_INLINE void OpCode_Compare::compare_script_extension(MatchInput const& input, MatchState& state, Unicode::Script script, bool inverse, bool& inverse_matched)
{
if (state.string_position == input.view.length())
return;
u32 code_point = input.view[state.string_position_in_code_units];
bool equal = Unicode::code_point_has_script_extension(code_point, script);
if (equal) {
if (inverse)
inverse_matched = true;
else
advance_string_position(state, input.view, code_point);
}
}
String const OpCode_Compare::arguments_string() const
{
return String::formatted("argc={}, args={} ", arguments_count(), arguments_size());