RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-29 10:55:10 +00:00
Code Issues Activity Actions

LibJS: Skip ID_{Start,Continue} property lookup for any ASCII characters

Browse source 
Before this change, Lexer::is_identifier_{start,middle}() would do a
Unicode property lookup via Unicode::code_point_has_property() quite
frequently, especially for common characters like .,;{}[]() etc.

Since these and any other ASCII characters not covered by the alpha /
alphanumeric check are known to not have the ID_Start / ID_Continue
(except '_', which is special-cased now) properties, we can easily
avoid this function call.
This commit is contained in:
Linus Groh 2021-09-13 23:20:58 +01:00 • committed by Andreas Kling
parent e4d97add3d
commit a50e33abe3
1 changed files with 12 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

12
Userland/Libraries/LibJS/Lexer.cpp
View file

@ -408,6 +408,11 @@ Optional<u32> Lexer::is_identifier_start(size_t& identifier_length) const
if (is_ascii_alpha(code_point) || code_point == '_' || code_point == '$')
return code_point;
// Optimization: the first codepoint with the ID_Start property after A-Za-z is outside the
// ASCII range (0x00AA), so we can skip code_point_has_property() for any ASCII characters.
if (is_ascii(code_point))
return {};
static auto id_start_category = Unicode::property_from_string("ID_Start"sv);
if (id_start_category.has_value() && Unicode::code_point_has_property(code_point, *id_start_category))
return code_point;
@ -436,6 +441,13 @@ Optional<u32> Lexer::is_identifier_middle(size_t& identifier_length) const
if (is_ascii_alphanumeric(code_point) || (code_point == '$') || (code_point == ZERO_WIDTH_NON_JOINER) || (code_point == ZERO_WIDTH_JOINER))
return code_point;
// Optimization: the first codepoint with the ID_Continue property after A-Za-z0-9_ is outside the
// ASCII range (0x00AA), so we can skip code_point_has_property() for any ASCII characters.
if (code_point == '_')
return code_point;
if (is_ascii(code_point))
return {};
static auto id_continue_category = Unicode::property_from_string("ID_Continue"sv);
if (id_continue_category.has_value() && Unicode::code_point_has_property(code_point, *id_continue_category))
return code_point;