[
https://issues.apache.org/jira/browse/LUCENE-1483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12655642#action_12655642
]
Mark Miller commented on LUCENE-1483:
-------------------------------------
bq. Is it TestSort.testTopDocScores that you see failing (that's what I see)?
right
bq. Unfortunately, the filter in that test is assuming that the IndexReader
it's passed in is equal to the "full" IndexReader, because it references
docs1.scoreDocs[0].doc, which is the docID space of the full reader.
right, its using the full id space - isnt that legal though? It did work.
bq. I would say the test is buggy (it's making an assumption about the API that
happens to be true but was not guaranteed). However, this could mean similar
filters "out there" think they can grab docIDs from the top-level IndexReader,
cache them, and then assign them inside the getDocIdSet method.
That was my first thought - are they aloud to build the filter this way? But it
worked to do it right? I like your assumption thought though - suits my lazy
side :)
What if someone made a filter that is supposed to only allow the first doc? So
it is set for 0...all of the first docs of each sub reader would match. That
seemed to be supported before right?
> Change IndexSearcher to use MultiSearcher semantics for sorted searches
> -----------------------------------------------------------------------
>
> Key: LUCENE-1483
> URL: https://issues.apache.org/jira/browse/LUCENE-1483
> Project: Lucene - Java
> Issue Type: Improvement
> Affects Versions: 2.9
> Reporter: Mark Miller
> Priority: Minor
> Attachments: LUCENE-1483.patch, LUCENE-1483.patch, LUCENE-1483.patch,
> LUCENE-1483.patch, LUCENE-1483.patch, LUCENE-1483.patch
>
>
> Here is a quick test patch. FieldCache for sorting is done at the individual
> IndexReader level and reloading the fieldcache on reopen can be much faster
> as only changed segments need to be reloaded.
--
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: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
