lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Harwood (JIRA)" <>
Subject [jira] Commented: (LUCENE-2454) Nested Document query support
Date Fri, 16 Jul 2010 16:13:52 GMT


Mark Harwood commented on LUCENE-2454:

bq. I made a minor modification your approch by making it do a "Forward-scan" instead of reverse

Interesting, but I'm not sure what guarantees Lucene will make about:

* Sequencing of calls on scorer.nextDoc (i.e. are calls to all scorers involved guaranteed
to be in doc-insert order ?)
* Index-time merging of segments (i.e. are all segments merged together in an order that keeps
the parent doc in one segment next to the child doc from the next segment?)

That seems like a fragile set of dependencies. 
Also, don't things get tricky when reporting matches from NestedDocumentQuery and PerParentLimitedQuery
back to the collector? During the query process the IndexSearcher resets the docId context
(Collector.setNextReader) as it moves from one Scorer/segment to another. If we are delaying
the assessment/reporting of matches until we've crossed a segment boundary it is too late
to report a match on a child doc id from a previous segment as the collector has already changed
context. Unfortunately "PerParentLimitedQuery" needs to do this when selecting the "best"
children for a single parent.

> Nested Document query support
> -----------------------------
>                 Key: LUCENE-2454
>                 URL:
>             Project: Lucene - Java
>          Issue Type: New Feature
>          Components: Search
>    Affects Versions: 3.0.2
>            Reporter: Mark Harwood
>            Assignee: Mark Harwood
>            Priority: Minor
>         Attachments:
> A facility for querying nested documents in a Lucene index as outlined in

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message