Changelog for python310-Whoosh-2.7.4-qubes.bp155.6.9.noarch.rpm :

* Mon Jun 21 2021 Matej Cepl - Add NullMatcherClass-hashable.patch which adds __hash__ method to NullMatcherClass, so that it works with Sphinx4 (gh#whoosh-community/whoosh#570).
* Thu Jul 18 2019 Tomáš Chvátal - Add patch to fix build with pytest4+ (both in git of upstream):
* pytest4.patch
* py2encoding.patch- Use fdupes- Update URL- Make sure py2 tests are run too
* Tue Dec 04 2018 Matej Cepl - Remove superfluous devel dependency for noarch package
* Mon Jul 10 2017 jengelhAATTinai.de- Ensure neutrality of description.
* Mon Jun 26 2017 tbechtoldAATTsuse.com- convert to singlespec- split -doc package- use files.pythonhosted.org as Source url- update to 2.7.4:
* Changed version number to 2.7.0.
* Added tag 2.7.0 for changeset 8abeb09483b9
* Results.extend() didn\'t update the cached length.
* Fixed VarBytesListColumn and FixedBytesListColumn.
* Handle an empty queue exception in mpwriter.
* Merging pull request with local commit.
* Added test for storing tuples.
* Remove caching decorator from VarBytesColumn reader.
* Added test for has_deletions() after an optimization.
* Try to avoid an UnboundLocalError in the DFA when MultiTerm.matcher() has an empty string.
* Don\'t try to write a vector if the list of items is empty.
* Implemented use of score_fn argument to NestedParent.
* Don\'t cache query -> bitset, since this somehow prevents the Searcher from being collected.
* Changed README docs link to point to readthedocs.org.
* Change uses of dump[s]() to always use protocol 2 instead of -1 (\"use latest protocol\") so indices created with Python 3.x are backwards compatible.
* Fix the analyzer in test_vector_unicode() to not lowercase, since this makes the test fail on some Python versions.
* Actually use the split_fn passed by the user. Not sure what happened here.
* Handle pickled Schema objects that don\'t have a _subfields attribute.
* Added fix and tests for calling expand_prefix() with unicode.
* Fix short code for Swedish in language aliases.
* Fix missing a consecutive token during highlighting
* Added estimate_size() methods to SpanNear2.
* Removed reference to \"allfields\" keyword argument.
* Added tests/english-words.10.gz to manifest so it\'s included in the distribution.
* Fix warnings in docs build.
* BufferedWriter should call close() instead of commit() when exiting a context.
* Reinstate fields.IDLIST\'s analyzer.
* Change the version of my fork to be distinct from the base repo.
* Backed out changeset 96255fc8ff17
* A typo in `whoosh.fields.KEYWORD` documentation
* Created new branch romanian-stemmer
* Skip ISO-8859-1 suffixes on Unicode strings
* Created new branch spanish-tokenizer
* Add sanity check for Spanish stemmer
* Fix sample highlight class
* Reorder the self._tempstorage.destroy() in SegmentWriter._finish to before the lock is released
* IOError from rmdir is ok if the error was ENOENT
* Fix reporting of total count in FilterCollector, based on PR #63 by Jannon Frank.
* Added a test for pickling a schema with a stemming analyzer.
* Initial unfinished, massive checkin of next-gen architecture.
* Fix forward-compatibility issue for Python 3.x.
* Write offsets in VarBytesColumn when there are more than a certain number of rows.
* Replaced porter stemming algorithm implementation with one based on the one in NLTK.
* Add IDEA\'s .cache directory to hgignore.
* Add IDEA\'s .cache directory to hgignore.
* Remove docstring chickenscratch.
* Fix test accidentally left with assert False at the end.
* Removed accidentally committed debug prints.
* Bumped version number for bugfix release.
* Added tag 2.7.1 for changeset 1bd4b9792eed
* Merging Bitbucket and local heads.
* Bumped version number.
* Added tag 2.7.2 for changeset f6bf123ac708
* Fixed logic for checking if floats are all whole in encode_weights.
* Change posting storage so the data from individual posts can be pulled out as raw bytes.
* Fixed error in CommaSeparatedTokenizer.
* Fixed import error.
* Minor changes.
* Updated VarBytesColumn to store offsets as well as lengths.
* Added context protocol to Index class. Store TOC file name in object.
* Switched to lighter-weight posting block format.
* Forgot to check in pfor.py.
* Reduced size of and references to whoosh.compat, in anticipation of using 3to2 for most things.
* Remove (object) from class definitions, assuming 3to2 will take care of it.
* Fixed problems from compat refactoring.
* Fixed problem with calling leaf matcher methods on a MultiMatcher.
* Don\'t encode bytes in _kws_to_query.
* Removed typing annotations to prevent circular import.
* Cleaned up filename generation/checking, added GC step to saving TOC.
* Added header checks in terms and postings files.
* Cleaned up storage cleanup, call it in SegmentWriter.cancel().
* Added another except clause to catch TypeError, because that\'s what\'s raised if you try to pickle a function.
* Fixed error that discarded vector metadata when merging segments.
* Added ability to specify an analyzer in KEYWORD field constructor.
* Bumped version number to 2.7.3.
* Added tag 2.7.3 for changeset f5a1508e93fd
* Added HyphenFilter to collapse hyphenated words into single tokens.
* Have OrGroup.factory() return a real class instead of a function.
* Bumped version number.
* Fri Apr 24 2015 cdenicoloAATTsuse.com- license update: BSD-2-Clause correct license ist BSD-2-Clause
* Wed Apr 22 2015 mciharAATTsuse.cz- Update to 2.7.0:
* Removed on-disk word graph implementation of spell checking in favor of much simpler and faster FSA implementation over the term file.
* Many bug fixes.
* Removed backwards compatibility with indexes created by versions prior to 2.5. You may need to re-index if you are using an old index that hasn\'t been updated.
* This is the last 2.x release before a major overhaul that will break backwards compatibility.
* Wed Jul 23 2014 mciharAATTsuse.cz- Upgrate to 2.5.7 (no changelog provided)
* Thu Oct 24 2013 speilickeAATTsuse.com- Require python-setuptools instead of distribute (upstreams merged)
* Mon Oct 07 2013 mciharAATTsuse.cz- build documentation during build