Questionable/Underspecified behavior of DirectoryReader.getIndexCommit() on NRT Readers [LUCENE-9040]
- Dominant language
- Java
- Stars
- 3.6k
- Forks
- 1.4k
- Avg merge
- 2d 11h
- Merged PRs (30d)
- 88
Description
It seems like DirectoryReader.getIndexCommit() returns weird
results when using a "reopened" reader off of uncommited IW changes.
Even though 2 diff readers will expose diff views of the index, they will
claim to refer to the same IndexCommit (generation).
Original email thread (no replies): http://mail-archives.apache.org/mod_mbox/lucene-dev/201910.mbox/%3Calpine.DEB.2.21.1910301611190.8171%40slate%3E
---
Migrated from [LUCENE-9040](https://issues.apache.org/jira/browse/LUCENE-9040) by Chris M. Hostetter (@hossman), updated Nov 19 2019
Attachments: [LUCENE-9040.patch](https://apache.github.io/lucene-jira-archive/attachments/LUCENE-9040/LUCENE-9040.patch)
Linked issues:
- [SOLR-13908](https://issues.apache.org/jira/browse/SOLR-13908)
- [SOLR-13872](https://issues.apache.org/jira/browse/SOLR-13872)
Contributor guide
Research direction
Start with DirectoryReader.getIndexCommit() and the NRT reader behavior described for readers reopened from uncommitted IndexWriter changes. Reproduce the case with two readers exposing different index views, then inspect the LUCENE-9040.patch attachment and related SOLR issues. Done should establish and test the expected IndexCommit behavior for these readers.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- search
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100