RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-14 07:54:58 +00:00
Code Issues Activity Actions

AK: Add ASCII fast path to Utf8CodePointIterator

Browse source 
Much of the UTF-8 data that we'll iterate over will be ASCII only,
and we can get a significant speed-up by simply having a fast path
when the iterator points at a byte that is obviously an ASCII character
(<= 0x7F).
This commit is contained in:
Andreas Kling 2023-12-29 17:36:06 +01:00
parent 75cecd19a5
commit a19d8a4a37
1 changed files with 12 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

12
AK/Utf8View.cpp
View file

@ -147,6 +147,13 @@ Utf8CodePointIterator& Utf8CodePointIterator::operator++()
{ {
VERIFY(m_length > 0); VERIFY(m_length > 0);
// OPTIMIZATION: Fast path for ASCII characters.
if (*m_ptr <= 0x7F) {
m_ptr += 1;
m_length -= 1;
return *this;
}
size_t code_point_length_in_bytes = underlying_code_point_length_in_bytes(); size_t code_point_length_in_bytes = underlying_code_point_length_in_bytes();
if (code_point_length_in_bytes > m_length) { if (code_point_length_in_bytes > m_length) {
// We don't have enough data for the next code point. Skip one character and try again. // We don't have enough data for the next code point. Skip one character and try again.
@ -190,6 +197,11 @@ ReadonlyBytes Utf8CodePointIterator::underlying_code_point_bytes() const
u32 Utf8CodePointIterator::operator*() const u32 Utf8CodePointIterator::operator*() const
{ {
VERIFY(m_length > 0); VERIFY(m_length > 0);
// OPTIMIZATION: Fast path for ASCII characters.
if (*m_ptr <= 0x7F)
return *m_ptr;
auto [code_point_length_in_bytes, code_point_value_so_far, first_byte_makes_sense] = Utf8View::decode_leading_byte(*m_ptr); auto [code_point_length_in_bytes, code_point_value_so_far, first_byte_makes_sense] = Utf8View::decode_leading_byte(*m_ptr);
if (!first_byte_makes_sense) { if (!first_byte_makes_sense) {