[
https://issues.apache.org/jira/browse/SOLR-14384?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17126935#comment-17126935
]
David Smiley commented on SOLR-14384:
-------------------------------------
Finally I think the PR is ready.
There is a slight doubt in my mind as to wether to tweak anything for the 8x
backport. The PR makes SolrRequestInfo rather strict in needing to always
clear after a set, and never set too many times or set too little, and can't
set null. If you violate these rules, then a ERROR is logged. If assertions
are enabled, it'll trip. Maybe just an upgrade note in the ref guide pointing
out the strict-ness? Or maybe ERROR is too harsh (do WARN instead)? Or it's
all totally fine as-is.
> Stack SolrRequestInfo
> ---------------------
>
> Key: SOLR-14384
> URL: https://issues.apache.org/jira/browse/SOLR-14384
> Project: Solr
> Issue Type: Improvement
> Reporter: David Smiley
> Assignee: David Smiley
> Priority: Minor
> Time Spent: 3h
> Remaining Estimate: 0h
>
> Sometimes SolrRequestInfo need to be suspended/overridden with a new one that
> is used temporarily. Examples are in the {{[subquery]}} transformer, and in
> warm of caches, and in QuerySenderListener (another type of warming), maybe
> others. This can be annoying to do correctly, and in at least one place it
> isn't done correctly. SolrRequestInfoSuspender shows some complexity. In
> this issue, [~dsmiley] proposes using a stack internally to SolrRequestInfo
> that is push'ed and pop'ed. It's not the only way to solve this but it's one
> way.
> See linked issues for the context and discussion.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]