Commits

Show all
Author Commit Message Labels Comments Date
Matt Chaput
Updated release notes, bumped version number to 2.3.0.
Tags
2.3
Matt Chaput
Added hyphen to the list of word chars in the wildcard regex to make the result more predictable. A true fix will have to wait for a more advanced parser.
Matt Chaput
Merging branches.
Matt Chaput
Added functions for merging Schema objects.
Matt Chaput
Moved base85 implementation into separate module, added arbitrary-length encoding and decoding functions.
Matt Chaput
Fixed outdated references to fields.Format in fields.py. Fixes issue #196.
Matt Chaput
Fixed stress test script to use a test directory name ignored by hg.
Matt Chaput
Fixed bad attribute reference. Fixed fieldcache saving. Added test for persistent field caching. Fixes issue #200.
Matt Chaput
Expander was caching the collection frequency of every term in memory, which was very slow for large indexes. Added a guard against calling expanded_terms() on an "empty" Expander object. Fixes issue #195.
Matt Chaput
Added test for vector with a single term.
Matt Chaput
Made update_document() with no unique fields equivalent to add_document().
Matt Chaput
Fixed a bug where the "inlining" of postings was not disabled for vectors, so vectors with a single term were thrown away because the terminfo object with the inlined postings was not used. The quick fix is to write vectors with inlinelimit=0.
Matt Chaput
Added FacetMap objects and the "maptype" argument to Searcher.search() and FacetType constructors, which allows much greater control over how facet groups are recorded. Cleaned up and updated documentation. Fixes issue #199.
Matt Chaput
Save the score/sort keys in the group lists and then sort the group lists by the scores/keys so docs in the group lists are in the same order as in the full results. See issue #197.
Matt Chaput
Updated docstrings for Searcher.documents() and Searcher.document_numbers().
Matt Chaput
Calling Searcher.documents()/.document_numbers() with no arguments will now yield all documents/numbers. Fixes issue #198.
Matt Chaput
Fixed Regex query type, added RegexPlugin and PseudoFieldPlugin. Fixes bug #189 and #190.
Matt Chaput
Removed another import print_function. Sorry Ross! Fixes issue #187.
Matt Chaput
ListMatcher.max_quality() failed if created without a scorer. Changed to forward to block_max_weight(). Fixes issue #191.
Matt Chaput
Fixed unfinished paragraph about StoredFieldFacet. Fixes issue #192.
Matt Chaput
Fixed unit test to work with spelling correction bugfix.
Matt Chaput
Changed float("nan") to a struct-based method that works on Python 2.5. Fixes issue #188.
Matt Chaput
Merging home and work branches.
Matt Chaput
Merging to get Regex query change I accidentally checked into the logging branch. See issue #189.
Matt Chaput
Added untested Regex query type. Refactored shared Wildcard/Regex functionality up to the PatternQuery class. See issue #189.
Branches
logging
Matt Chaput
Starting 'logging' branch
Branches
logging
Matt Chaput
Fixed sorting error. Fixed bug where original word could appear in suggestions. Updated docstring to explicitly state the original word will not be in the suggestions. Fixes #174.
Matt Chaput
Removed unused imports that were breaking under Python 2.5. Fixes issue #187.
Matt Chaput
Added tag 2.2.2 for changeset 9a84c1a1d557
Matt Chaput
Fixed FieldFacet to work with readers that don't support field caches. Fixes issue #177. Bumped version number to 2.2.2.
Tags
2.2.2
  1. Prev
  2. Next