src: move all 1-byte encodings to native #61118
Open
+519
−160
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
|
Review requested:
|
nodejs-github-bot
added
c++
ChALkeR
changed the title
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
7 times, most recently
from
6130c12 to
4be8283
Compare
mcollina
reviewed
Dec 19, 2025
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
from
4be8283 to
6a46ae3
Compare
This was referenced Jan 17, 2026
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
from
6a46ae3 to
109aa87
Compare
Member
Author
|
#61093 landed, rebased. |
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
3 times, most recently
from
51adfc3 to
82a6ecc
Compare
src/encoding_singlebyte.h
Outdated
Comment on lines
370
to
387
| const uint16_t* const tSingleByteEncodings[28] = { | ||
| tIBM866, tKOI8_R, tKOI8_U, tMacintosh, tXMacCyrillic, | ||
| tISO_8859_2, tISO_8859_3, tISO_8859_4, tISO_8859_5, tISO_8859_6, tISO_8859_7, | ||
| tISO_8859_8, tISO_8859_8, // second time for for 8-i | ||
| tISO_8859_10, tISO_8859_13, tISO_8859_14, tISO_8859_15, tISO_8859_16, | ||
| tWindows874, tWindows1250, tWindows1251, tWindows1252, tWindows1253, | ||
| tWindows1254, tWindows1255, tWindows1256, tWindows1257, tWindows1258 | ||
| }; | ||
|
|
||
| // Matches the list of encoding sin single-byte.js | ||
| const bool fSingleByteEncodings[28] = { | ||
| fIBM866, fKOI8_R, fKOI8_U, fMacintosh, fXMacCyrillic, | ||
| fISO_8859_2, fISO_8859_3, fISO_8859_4, fISO_8859_5, fISO_8859_6, fISO_8859_7, | ||
| fISO_8859_8, fISO_8859_8, // second time for for 8-i | ||
| fISO_8859_10, fISO_8859_13, fISO_8859_14, fISO_8859_15, fISO_8859_16, | ||
| fWindows874, fWindows1250, fWindows1251, fWindows1252, fWindows1253, | ||
| fWindows1254, fWindows1255, fWindows1256, fWindows1257, fWindows1258 | ||
| }; |
Member
There was a problem hiding this comment.
If the header is included in more than one translation unit, it may cause ODR violations. We can either mark them inline or move their definitions to a .cc file.
Member
|
For the cpp & other lint errors, this might help: |
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
from
82a6ecc to
35b7960
Compare
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
7 times, most recently
from
b5d51ae to
ee8b680
Compare
Member
Author
|
@mertcanaltin lint was just not ready yet (as was noted in the PR description), fixed. |
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
from
ee8b680 to
0847010
Compare
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #61118 +/- ##
==========================================
- Coverage 88.52% 88.50% -0.03%
==========================================
Files 704 703 -1
Lines 208802 208785 -17
Branches 40318 40312 -6
==========================================
- Hits 184842 184784 -58
- Misses 15947 15973 +26
- Partials 8013 8028 +15
🚀 New features to boost your workflow:
|
Co-authored-by: Mert Can Altin <mertgold60@gmail.com>
ChALkeR
force-pushed
the
chalker/decoder/single-byte/1
branch
from
0847010 to
1d96511
Compare
gurgunday
reviewed
Jan 17, 2026
Member
gurgunday
left a comment
gurgunday
left a comment
There was a problem hiding this comment.
nice!
lgtm other than one comment
| return; | ||
| } | ||
|
|
||
| uint16_t* dst = node::UncheckedMalloc<uint16_t>(length); |
Member
There was a problem hiding this comment.
Here we can allocate a huge uint16_t[length] before any V8 string-length limit check, and if it ends up being higher than what V8 accepts, we would allocate for no reason
Maybe it would be better to guard this with sth like:
Suggested change
| uint16_t* dst = node::UncheckedMalloc<uint16_t>(length); | |
| if (length > static_cast<size_t>(v8::String::kMaxLength)) { | |
| // throw | |
| } | |
| uint16_t* dst = node::UncheckedMalloc<uint16_t>(length); |
Let me know if I'm missing something
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Tracking: #61041
This builds on top of #61093 and gives an additional ~2x improvement by moving the same logic to native
The old native fast path removed in #61093 had a bunch of nested ifs and converted data from utf16 to utf8 and then back to utf16
This instead constructs strings using direct maps as #61093 and returns them as raw buffers
windows-1252, main pre-#61093:windows-1252, main with #61093:windows-1252, this PR:This also similarly improves all other 1-byte encodings compared to #61093
For comparison, Bun:
v8/jsc is not an issue, unoptimal code is
cc @nodejs/performance