OSSForks/ladybird
1
0
Fork 0
mirror of https://github.com/zebrajr/ladybird.git synced 2025-12-06 00:19:53 +01:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

AK: Add fast paths for Utf8View::*_offset_of() methods

Browse Source 
If all code points in the string are represented by a single byte, we
can simply take the fast path of returning the input for these methods.
This commit is contained in:
Jelle Raaijmakers 2025-06-12 20:26:25 +02:00 committed by Jelle Raaijmakers
parent 7beecaa43d
commit 01ede6cc58
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
AK/Utf8View.cpp
View File

@ -33,6 +33,11 @@ Utf8CodePointIterator Utf8View::iterator_at_byte_offset_without_validation(size_
size_t Utf8View::code_point_offset_of(size_t byte_offset) const
{
VERIFY(byte_offset < byte_length());
// Fast path: each code point is represented by a single byte.
if (length() == byte_length())
return byte_offset;
size_t code_point_offset = 0;
for (auto it = begin(); !it.done(); ++it) {
if (it.m_ptr > begin_ptr() + byte_offset)
@ -44,8 +49,13 @@ size_t Utf8View::code_point_offset_of(size_t byte_offset) const
size_t Utf8View::byte_offset_of(size_t code_point_offset) const
{
size_t byte_offset = 0;
VERIFY(code_point_offset < length());
// Fast path: each code point is represented by a single byte.
if (length() == byte_length())
return code_point_offset;
size_t byte_offset = 0;
for (auto it = begin(); !it.done(); ++it) {
if (code_point_offset == 0)
return byte_offset;
@ -53,7 +63,6 @@ size_t Utf8View::byte_offset_of(size_t code_point_offset) const
byte_offset += it.underlying_code_point_length_in_bytes();
--code_point_offset;
}
return byte_offset;
}