Releases: ridiculousfish/regress
Releases · ridiculousfish/regress
v0.10.3
v0.10.2
What's Changed
- Fix a crash with empty brackets like
/[]*/
in #99 - Fix a false positive match for ASCII brackets against certain UTF-16 inputs in #100
- Update hashbrown requirement from 0.14.3 to 0.15.0 by @dependabot in #98
v0.10.0
v0.9.1
v0.9.0
v0.8.0
Version 0.8.0 of regress, REGex in Rust with EmcaScript Syntax.
- Fix a bug in case folding characters given by escapes (8b842f1)
- Correct case folding (56233d6)
- Correct first-byte computation for interval (4858a1b)
- Implement Unicode Character Classes in Brackets (#71)
- Fix multiple escape parsing issues (#76)
- Implement UTF-16 based matching (#75)
- Correctly reject characters in Atom PatternCharacter (#82)
- Conditionally parse named capture groups (#83)
- Fix invalid quantifier loop (#84)
- Implement unicode flag dependent case folding (#81)
- Fix parsing for invalid control characters in class atoms (#85)
- Allow quantifiable assertions in non unicode mode (#86)
- Fix remaining unicode script property tests (#87)
Full Changelog: v0.7.0...v0.8.0
v0.7.0
Version 0.7.0 of regress, REGex in Rust with EmcaScript Syntax.
- Corrected errors when parsing certain invalid quantifiers (@selfisekai #59)
- Added U+00A0 (NBSP) to whitespace (@HalidOdat #63)
- Capture groups with the same name now emit parse errors (@raskad #65)
- Lookbehinds now reject following invalid quantifiers (@raskad #66)
- Corrected handling of Unicode escapes in capture group names (@sirosen #67)
v0.6.0
Version 0.6.0 of regress, REGex in Rust with EmcaScript Syntax.
- Unpaired surrogates are now properly handled in accordance with the EcmaScript spec (@jedel1043 in #60)
- Invalid ranges like
[a-\s]
are now permitted in non-Unicode mode, in accordance with Annex B of EcmaScript spec (@HalidOdat in #62)
v0.5.0
Version 0.5.0 of regress, REGex in Rust with EmcaScript Syntax.
- Unicode property escape matching like
\p{Letter}
is implemented - Regex parsing may now use any u32 iterator, not simply strings
- The Unicode flag "u" is now recognized. Unicode is no longer the default; however non-Unicode regular expression support still has some known differences from JavaScript.