Commits

Author Commit Message Labels Comments Date
Matt Chaput
Fixed tests on Python 3.2. Added py2.5 and py3.2 to tox.ini.
Matt Chaput
Fixed Python 2.5 compatibility problems: can't use plaintext codec, tuple doesn't have index method.
Matt Chaput
Fixed missing compat import.
Matt Chaput
Fixed references to _active in classes that weren't calling superclass.__init__.
Matt Chaput
Fixed loading of old-style DATETIME fields.
lakin.wecker
fixing the issue of using filters with search_page which causes a few type errors.
Branches
2.4x
lakin.wecker
fixing the ListMatcher.skip_to_quality. There are two issues here. 1 - it ALWAYS skips at least one element - which I think is incorrect. Next, it refers to .quality() an attribute which does not exist. I believe that should be .block_quality().
Branches
2.4x
Matt Chaput
Added COLUMN field type for unindexed, unstored field with a column. Added query types for searching column values instead of the inverted index. Added go_inactive() method to matchers.
Matt Chaput
Added CoordMatcher wrapper, which boosts/penalizes document scores based on how many query terms match in the document. See issue #271.
Matt Chaput
Moved spans module under whoosh.query.
Matt Chaput
Added ability to get column values using Hit[]. Added column reader cache to SegmentReader.
Matt Chaput
Changed max_quality() to a method in anticipation of doing it more dynamically. Moved/renamed FilePostingMatcher intermediate class to mcore.LeafMatcher. Renamed W2/W3 PostingMatcher classes to LeafMatcher. Integrated BlockReader into W3LeafMatcher. Added ability for UnionMatcher to decay into AndMaybeMatcher. Fixed logic error in SpanOrMatcher.
Matt Chaput
Checking in changes to make Whoosh 3 codec pass tests. Added memory codec, replaced use of ramindex in BufferedWrtier.
Matt Chaput
Fixed error in super initialization call.
Matt Chaput
Fixed all_stored_fields() to return deleted + undeleted documents. Added Searcher.iter_docs() method to allow iterating only undeleted stored fields. Fixed references to codec in SegmentReader.__init__().
Matt Chaput
Added StructFile wrapper that keeps track of checksum on file contents.
Matt Chaput
All tests pass with whoosh3 codec.
Matt Chaput
Converted codec interface to "per_document_reader" instead of 3 separate mini-readers. The separate readers don't make sense for Whoosh 3 which uses columns for all per-doc info.
Matt Chaput
Close the per-doc writer before trying to read lengths from it, to allow non-RAM lengths.
Matt Chaput
Fixed tests after reorganization. Moved legacy support into whoosh2 module. More work on columns implementation.
Matt Chaput
Copying fix for issue 267 from 2.4x branch.
Matt Chaput
Fixed broken import in legacy. Reformatted multi-line imports.
Matt Chaput
Moved code from fileindex, filewriting, and filewriting into index, writing, and reading. The original idea of filedb was that it could be one of many possible backends. However, it's unlikely that Whoosh will ever support a non-file-based backend. From now on, the filedb package will just be for disk I/O related code.
Matt Chaput
Removed RamIndex and commented out BufferedWriter pending replacement.
Matt Chaput
Initial commit of column writing/reading code.
Matt Chaput
Reorganized modules. Changed util module to a package. Moved several modules from support package to util package to return support to a place for 3rd party code. Moved and renamed several modules to the top level package.
Thomas Waldmann
import with_statement from __future__, fixes #268
Branches
2.4x
lakin.wecker
fixing the issue of using filters with search_page which causes a few type errors.
Comments 2
lakin.wecker
fixing the ListMatcher.skip_to_quality. There are two issues here. 1 - it ALWAYS skips at least one element - which I think is incorrect. Next, it refers to .quality() an attribute which does not exist. I believe that should be .block_quality().
Comments 2
Matt Chaput
Removed uses of uncommitted functionality.
Branches
2.4x
  1. Prev
  2. Next