Fix bug in scientific notation parsing #711
Merged
Conversation
This file contains 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
Codecov Report
@@ Coverage Diff @@
## main #711 +/- ##
==========================================
+ Coverage 84.97% 85.36% +0.38%
==========================================
Files 54 54
Lines 5858 5855 -3
Branches 1338 1337 -1
==========================================
+ Hits 4978 4998 +20
+ Misses 597 576 -21
+ Partials 283 281 -2
📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more |
Benchmark resultsBefore this PR: 353.1 thousand lines per second Measured change: 0.06% faster (1.12% slower to 0.54% faster) |
Fixes #676 Scientific notation parsing was using the shared `readInt` code path, which I had made more flexible to handle hex digits, but that means that literals like `1e2` were being treated as a complete number (with the hex digit `e`) rather than as scientific notation. Normally this wouldn't be a problem since it's all a number token anyway, but it caused trouble when using dot-style property access like `1e2.toString()`. Expressions like `1.toString()` are expected to fail because `.` is interpreted as a decimal point, but when using scientific notation, the parser needs to be smart enough see that a decimal point wouldn't be valid, and therefore not include it as part of the token. The fix was to change `readInt` to only handle decimal digits, not hex digits, and to inline the hex case into `readRadixNumber`. This should make number parsing slightly faster because it's not checking for hex digits anymore.
alangpierce
force-pushed
the
fix-scientific-notation-bug
branch
from
f9da2bc
to
4a04c49
Compare
1Lighty
pushed a commit
to Astra-mod/sucrase
that referenced
this pull request
Aug 14, 2022
Fixes alangpierce#676 Scientific notation parsing was using the shared `readInt` code path, which I had made more flexible to handle hex digits, but that means that literals like `1e2` were being treated as a complete number (with the hex digit `e`) rather than as scientific notation. Normally this wouldn't be a problem since it's all a number token anyway, but it caused trouble when using dot-style property access like `1e2.toString()`. Expressions like `1.toString()` are expected to fail because `.` is interpreted as a decimal point, but when using scientific notation, the parser needs to be smart enough see that a decimal point wouldn't be valid, and therefore not include it as part of the token. The fix was to change `readInt` to only handle decimal digits, not hex digits, and to inline the hex case into `readRadixNumber`. This should make number parsing slightly faster because it's not checking for hex digits anymore.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Fixes #676
Scientific notation parsing was using the shared
readIntcode path, which Ihad made more flexible to handle hex digits, but that means that literals like
1e2were being treated as a complete number (with the hex digite) ratherthan as scientific notation. Normally this wouldn't be a problem since it's all
a number token anyway, but it caused trouble when using dot-style property
access like
1e2.toString(). Expressions like1.toString()are expected tofail because
.is interpreted as a decimal point, but when using scientificnotation, the parser needs to be smart enough see that a decimal point wouldn't
be valid, and therefore not include it as part of the token.
The fix was to change
readIntto only handle decimal digits, not hex digits,and to inline the hex case into
readRadixNumber. This should make numberparsing slightly faster because it's not checking for hex digits anymore.
I also added some tests since some of this code wasn't well covered by test
coverage.