cpoerschke commented on a change in pull request #300: SOLR-11831: Skip second
grouping step if group.limit is 1 (aka Las Vegas Patch)
URL: https://github.com/apache/lucene-solr/pull/300#discussion_r272682603
##########
File path:
solr/core/src/java/org/apache/solr/handler/component/QueryComponent.java
##########
@@ -288,6 +328,13 @@ protected void prepareGrouping(ResponseBuilder rb) throws
IOException {
groupingSpec.setMain(params.getBool(GroupParams.GROUP_MAIN, false));
groupingSpec.setNeedScore((rb.getFieldFlags() &
SolrIndexSearcher.GET_SCORES) != 0);
groupingSpec.setTruncateGroups(params.getBool(GroupParams.GROUP_TRUNCATE,
false));
+
+
groupingSpec.setSkipSecondGroupingStep(params.getBool(GroupParams.GROUP_SKIP_DISTRIBUTED_SECOND,
false));
+ boolean isReranking = (rb.getRankQuery() != null);
+ if (groupingSpec.isSkipSecondGroupingStep() &
!allowSkipSecondGroupingStep(groupingSpec.getWithinGroupSortSpec(),
groupingSpec.getGroupSortSpec(), isReranking)){
Review comment:
observation: `allowSkipSecondGroupingStep` currently does 'error' logging in
response to illegal grouping specifications in a request. this probably should
not be 'error' or 'warn' logging (since there is nothing 'wrong' with Solr
itself) but 'info' logging at most or no logging at all with the details
returned in the `Illegal grouping specification` that is thrown here. An
alternative could also be to 'info' log and _not_ throw but to just ignore the
`group.skip.second.step=true` parameter if it's invalid.
edit: a little further below in the class there's a
`cmd.setSegmentTerminateEarly(false); // not supported, silently ignore any
segmentTerminateEarly flag` for example
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
