Releases: rapidfuzz/python-Levenshtein
Releases · rapidfuzz/python-Levenshtein
Release 0.21.0
Changed
- relax dependency requirement on
rapidfuzz
Release 0.20.9
Fixed
- fix function signature of
get_requires_for_build_wheel
Release 0.20.8
Fixed
- type hints for
editops/opcodes/matching_blocksdid not allow any
hashable sequence
Release 0.20.7
Fixed
- type hints did not get installed
Release 0.20.6
Fixed
- fix incorrect result normalization in
setratioandseqratio
Release 0.20.5
Fixed
- fix support for cmake versions below 3.17
- fix version requirement for
rapidfuzz-cppwhen building against a previously installed version
Release 0.20.4
This release converts the python-Levenshtein package to a metapackage, which simply installs the Levenshtein package which is API compatible. Below is a summary of the improvements:
Performance
- performance improved for the following functions:
distance,ratio,jaro,jaro_winkler,editops,opcodes
Fixed
- fixed incorrect results of
jaroandjaro_winkler - fix issues in multiple of the median functions which could lead to segmentation faults
Changes
- drop support for Python versions < 3.6
- replace usage of deprecated Unicode APIs, which will be removed in Python 3.12
Added
- add type hints
- add wheels for all common platforms