v2.0.1

• Updates Textmatch dependency, including fix for case plus regex ignores not being combining correctly

v2.0

• Ignore options are all now all behind a single flag, --ignore. For example, to ignore case and nonalpha characters you would previously have written --ignore-case --ignore-nonalpha, or -ia for short. Now you should write --ignore case nonalpha or -i c na. Some of the ignores have been renamed too -- see --help for details.
• The option to ignore letter order has been removed, it wasn't useful.
• Various deprecated aliases for ignore options have now been removed: --filter (use ignore regex), --filter-titles (use ignore titles), --as-latin (use ignore nonlatin), and --sort-words (use ignore words order).
• The fuzzy method flag has been renamed from --fuzzy to --method. It now defaults to an exact match.
• The short option for --threshold was previously -r, but it is now -t.
• Input is no longer accepted from STDIN.
• Output now defaults to including all fields from both input files, instead of only the fields used for the match.
• Added option to ignore leading and tailing words.
• Added support for blocking.
• Many operations will be faster.
• Improved remaining time display.

v1.24

• Fixes a bug where the wrong version number was reported

v1.23

• Updates dependencies, fixes issue #38

v1.22

• Adds some extra titles
• Metaphone now works by-word (split by space) rather than trying to encode the whole thing

v1.21

• Updates dependencies

v1.20

• Updates dependencies, fixing Python 3.8 compatibility
• Other minor fixes

v1.19

• Ignore parameter names are now more consistent (old names still work for now)
• Parameter to ignore order of letters
• Ignoring non-alphanumeric also ignores spaces now
• Levenshtein should be faster on Unicode text now
• Hopefully faster naive comparisons (ignores)
• Upgraded dependencies
• More tests
• Removed Python 2 support!

v1.18

• Fixes multi-field comparisons -- it's now only a match if all fields pass the threshold, rather than if any did
• Allow multiple algorithms and multiple thresholds to be specified, which are interpreted respective to the specified fields
• Prevents error messages when piping output
• Updated dependencies

v1.17

• Updated degree help text
• Better encoding detection
• Better encoding errors
• Add common installation problems to readme