magibney commented on pull request #624:
URL: https://github.com/apache/solr/pull/624#issuecomment-1045163521
Sorry, I misunderstood what you meant in your [comment
above](#pullrequestreview-887690496). But in any event I still think the
concern is not warranted:
1. `fq=filter({!cache=false}foo:bar)` would currently cache the FilterQuery,
but not the inner query, so subsequent requests for `fq=foo:bar` would not be a
cache hit.
2. `fq=filter(foo:bar)` currently caches twice, this PR would cause to only
cache `foo:bar`
3. `fq={!cache=false}filter(foo:bar)`, oddly, would currently only cache the
inner query (behavior equivalent to what this PR introduces as the default for
`fq=filter(foo:bar)`
4. `fq={!cache=true}filter(foo:bar)` -- i.e., explicitly requesting that the
FilterQuery _per se_ be cached, would throw an IllegalArgumentException under
the new PR, and on current main the behavior of this would be equivalent to
`fq=filter(foo:bar)`: double-caching. I mean, we could skip the IAE, but this
is really really wrong -- I can't think of any circumstance where you'd
actually want this.
5. And yes, you'd be better off with `(foo:bar)^=1` for an explicit
ConstantScoreQuery; currently in order to support the use case you mention
(using FilterQuery to filter but _not_ cache), iiuc you'd have to do this:
`filter({!cache=false}foo:bar)`!
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
