mocobeta commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146600229
As for backward compatibility, my thought on that was slightly changed -
let's commit this into `main` without adding extra configuration parameters if
there are no objections, it's a major
rmuir commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146608421
i can try to do some tests this weekend (using query-expansion approach).
i've just been busy with work and it is a bit time-consuming (need to find some
external hard drives and so on)
--
rmuir commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146609887
> I'll add a flag to opt-in for `branch_9x` when backporting (we would need
to preserve the current behavior as default in 9.x).
why add flag? The similarity is already pluggable. So I t
mocobeta commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146612939
Thanks @rmuir for your suggestions.
I have a deadline myself this month so can't help in trying extensive
experiments right now, but I'll try to be responsive here.
--
This is an auto
mocobeta commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146613684
I haven't closely looked at the `SimpleQQParser` yet, but I imagine this
measures something like the performance of pseudo relevance feedback?
--
This is an automated message from the Apa
rmuir commented on PR #940:
URL: https://github.com/apache/lucene/pull/940#issuecomment-1146616934
It does not do this by default, but can be easily modified to do so... I
have done it before. It is one way to measure how well the mlt is working
relatively easily, albeit indirectly. At leas
[
https://issues.apache.org/jira/browse/LUCENE-10589?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tomoko Uchida reopened LUCENE-10589:
The same failure happened with another seed and multiplier.
{code}
./gradlew test --tests Tes
[
https://issues.apache.org/jira/browse/LUCENE-10583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17549375#comment-17549375
]
Thomas Hoffmann commented on LUCENE-10583:
--
Hello Vigya,
thanks for reporting
[
https://issues.apache.org/jira/browse/LUCENE-10583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17550072#comment-17550072
]
Vigya Sharma commented on LUCENE-10583:
---
Thanks for confirming, Thomas.
{quote}no
[
https://issues.apache.org/jira/browse/LUCENE-10583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17550172#comment-17550172
]
Thomas Hoffmann commented on LUCENE-10583:
--
Thanks for your answer. The main p
[
https://issues.apache.org/jira/browse/LUCENE-10583?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thomas Hoffmann updated LUCENE-10583:
-
Issue Type: Improvement (was: Bug)
> Deadlock with MMapDirectory while waitForMerges
>
[
https://issues.apache.org/jira/browse/LUCENE-10583?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Thomas Hoffmann updated LUCENE-10583:
-
Priority: Minor (was: Critical)
> Deadlock with MMapDirectory while waitForMerges
> --
12 matches
Mail list logo
