[
https://issues.apache.org/jira/browse/SOLR-15048?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chris M. Hostetter updated SOLR-15048:
--------------------------------------
Attachment: SOLR-15048.patch
Status: Open (was: Open)
There are twe distinct but interconnected problems that were contributing to
the symptoms i was seeing before (which were only a subset of the actual
problems involved, due to the natural order of the docs in my test)
* the abstract base "strategy" classes expect that (in the nullPolicy=expand)
case _if_ a null doc was boosted, then it would not *also* be listed in the
'nullScores' array (and the nullScoreIndex would not be increased in the "null
group but boosted" code path of the finish method()) -- but not all of the
concrete strategy impls were returning immediately when detecting a boosted
doc, meaning that "extra" scores would be in the nullScores array, leading to
using the wrong index to try nad find the score of (non-boosted) null docs
* the finish method in the abstract base "strategy" class takes care of purging
any "group heads" that have the same group value as a known boosted doc, but
this code wasn't also cleaning the "nullDoc" group head of the null group (ie:
it assumed nullPolicy=ignore or nullPolicy=expand)
I'm attaching a patch with more beefed up testing, and the minimal changes to
get all of hte various strategies/collectors in agreement of how to deal with
null groups -- i plan to do a lot of deduplication refactoring prior to
committing (copy/past coding appears to be how we got in this mess) but i
wanted to post the patch in this state since it makes it more evident where the
"wrong" code was.
> collapse + query elevation behaves inconsistenty w/ 'null group' docs
> depending on group head selector
> ------------------------------------------------------------------------------------------------------
>
> Key: SOLR-15048
> URL: https://issues.apache.org/jira/browse/SOLR-15048
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Chris M. Hostetter
> Assignee: Chris M. Hostetter
> Priority: Major
> Attachments: SOLR-15048.patch, SOLR-15048.patch, SOLR-15048.patch
>
>
> while working on SOLR-15047, I realized I wasn't really clear on what the
> _expected_ semantics of were suppose to be when "boosting"
> docs that had null values in the collapse field.
> I expanded on my test from that jira, to demonstrate the logic i (thought) i
> understood from the Ord based collector - but then discovered that depending
> on the group head selector used (ie: OrdScoreCollector vs
> OrdFieldValueCollector+OrdIntStrategy vs
> OrdFieldValueCollector+OrdValueSourceStrategy , etc...) you get different
> behavior - not just in what group head is selected, but even when the
> behavior should be functionally equivilent, you can get different sets of
> groups. (even for simple string field collapsing, independent of the bugs in
> numeric field collapsing).
>
> I have not dug into WTF is happening here, but I'll attach my WIP test
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
