Replace direct mmaped buffer with Lucene abstractions in KnnVectorDict [LUCENE-10057]
Open
legacy-jira-priority:Major
type:bug
- Dominant language
- Java
- Stars
- 3.6k
- Forks
- 1.4k
- Avg merge
- 2d 11h
- Merged PRs (30d)
- 88
Description
---
Migrated from [LUCENE-10057](https://issues.apache.org/jira/browse/LUCENE-10057) by Dawid Weiss (@dweiss)
Attachments: [LUCENE-10057.patch](https://apache.github.io/lucene-jira-archive/attachments/LUCENE-10057/LUCENE-10057.patch)
Contributor guide
Research direction
Start with KnnVectorDict and the attached LUCENE-10057.patch, then trace how its direct memory-mapped buffer is used. Replace that access with the applicable Lucene abstractions; done means KnnVectorDict no longer relies on the direct mmaped buffer while preserving its vector-dictionary behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- search
- Issue type
- Refactor
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100