Lucene summary for this week:

   - A 5.3.1 bugfix release may be coming soon
   <http://lucene.markmail.org/thread/d356iwsseh25agoj>

   - See the confusion matrix
   <https://issues.apache.org/jira/browse/LUCENE-6479> for a classifier

   - Remove a nasty classloader hack that broke MorfologikFilter
   <https://issues.apache.org/jira/browse/LUCENE-6774>, and fix it correctly
   <https://issues.apache.org/jira/browse/LUCENE-6775> so you can pass the
   dictionary as a URI

   - The new BoostQuery decouples Query from boosting
   <https://issues.apache.org/jira/browse/LUCENE-6590>, but it was missing
   its rewrite method <https://issues.apache.org/jira/browse/LUCENE-6781>

   - How can we compress postings payloads
   <https://issues.apache.org/jira/browse/LUCENE-6764>?

   - Nested conjunctions should always be flattened
   <https://issues.apache.org/jira/browse/LUCENE-6773>

   - MultiCollector did not handle early termination properly
   <https://issues.apache.org/jira/browse/LUCENE-6772>

   - Add a point-within-distance query
   <https://issues.apache.org/jira/browse/LUCENE-6698> implemented with BKD
   trees

   - Speed up IndexSearcher.count
   <https://issues.apache.org/jira/browse/LUCENE-6754>when a query is so
   simple (match all, single term) that we can use index statistics instead

   - The integration of BKDTree and Geo3D
   <https://issues.apache.org/jira/browse/LUCENE-6699>is done (for Lucene
   5.4.0), providing accurate and fast earth-surface "point in shape" queries,
   but we need to make its randomized tests more evil by simulating planets
   more squashed than earth
   <https://issues.apache.org/jira/browse/LUCENE-6776>, requiring somecrazy
   math
   
<https://issues.apache.org/jira/browse/LUCENE-6776?focusedCommentId=14729991&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14729991>,
   including Lagrange multipliers
   <https://en.wikipedia.org/wiki/Lagrange_multiplier>

   - GeoPointDistanceQuery is buggy with large distances
   <https://issues.apache.org/jira/browse/LUCENE-6780>

   - Don't use approximations
   <https://issues.apache.org/jira/browse/LUCENE-6761> for MatchAllDocsQuery,
   and give it a dedicated BulkScorer
   <https://issues.apache.org/jira/browse/LUCENE-6756>

   - Dodge bugs in Java's collators
   <https://issues.apache.org/jira/browse/LUCENE-6206>

   - Windows NTFS pending delete state for a file
   <https://issues.apache.org/jira/browse/LUCENE-6684> causes assertion
   failures in Lucene; we should fix Lucene's WindowsFS to also simulate
   this state <https://issues.apache.org/jira/browse/LUCENE-6771>

   - Symlinks to an index directory continue to cause problems for users
   <https://issues.apache.org/jira/browse/LUCENE-6770>

   - CheckIndex cannot handle corrupt .si
   <https://issues.apache.org/jira/browse/LUCENE-6762>files

   - Reduce heap used by <https://issues.apache.org/jira/browse/LUCENE-6779>
    CompressingStoredFieldsWriter when writing large strings during
   indexing

   - Reduce heap used by
<https://issues.apache.org/jira/browse/LUCENE-6777> the
   new geo point queries by building the BytesRef on demand for sub-ranges

   - GeoPointDistanceRangeQuery will match points within a min/max distance
   range <https://issues.apache.org/jira/browse/LUCENE-6778>

   - When you incorrectly index nested documents the resulting error
   messages are very confusing
   <https://issues.apache.org/jira/browse/LUCENE-6660>

   - DisjunctionMaxQuery, BoostingQuery and BoostedQuery now use
   IndexSearcher to create sub-weights
   <https://issues.apache.org/jira/browse/LUCENE-6746> so caching can apply

Mike McCandless

Reply via email to