fergalmoran/ladybird
1
0
Fork 0
mirror of https://github.com/fergalmoran/ladybird.git synced 2026-01-06 08:36:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

LibUnicode: Update to Unicode version 15.1.0

Browse Source 
https://unicode.org/versions/Unicode15.1.0/

This update includes a new set of code point properties, Indic Conjunct
Break. These may have the values Consonant, Linker, or Extend. These are
used in text segmentation to prevent breaking on some extended grapheme
cluster sequences.
This commit is contained in:
Timothy Flynn
2023-09-15 09:23:56 -04:00
committed by Andreas Kling
parent ae15b68b79
commit 139c575cc9
4 changed files with 70 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
Userland/Libraries/LibUnicode/Segmentation.cpp
View File

@@ -57,6 +57,22 @@ static void for_each_grapheme_segmentation_boundary_impl([[maybe_unused]] ViewTy
return (code_point_has_grapheme_break_property(code_point, properties) || ...);
};
auto skip_incb_extend_linker_sequence = [&](auto& it) {
while (true) {
if (it == view.end() || !code_point_has_property(*it, Property::InCB_Extend))
return;
auto next_it = it;
++next_it;
if (next_it == view.end() || !code_point_has_property(*next_it, Property::InCB_Linker))
return;
it = next_it;
++it;
}
};
// GB1
if (callback(0) == IterationDecision::Break)
return;
@@ -70,6 +86,23 @@ static void for_each_grapheme_segmentation_boundary_impl([[maybe_unused]] ViewTy
for (++it; it != view.end(); ++it, code_point = next_code_point) {
next_code_point = *it;
// GB9c
if (code_point_has_property(code_point, Property::InCB_Consonant)) {
auto it_copy = it;
skip_incb_extend_linker_sequence(it_copy);
if (it_copy != view.end() && code_point_has_property(*it_copy, Property::InCB_Linker)) {
++it_copy;
skip_incb_extend_linker_sequence(it_copy);
if (it_copy != view.end() && code_point_has_property(*it_copy, Property::InCB_Consonant)) {
next_code_point = *it_copy;
it = it_copy;
continue;
}
}
}
// GB11
if (code_point_has_property(code_point, Property::Extended_Pictographic) && has_any_gbp(next_code_point, GBP::Extend, GBP::ZWJ)) {
auto it_copy = it;