javanna commented on PR #11793:
URL: https://github.com/apache/lucene/pull/11793#issuecomment-1308433716
@jpountz would you have time to take another look at this please?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
mohamedniyaz1996 opened a new issue, #11908:
URL: https://github.com/apache/lucene/issues/11908
### Description
As we know, lucene spellcheck
[suggestsimilar()](https://lucene.apache.org/core/6_0_1/suggest/org/apache/lucene/search/spell/SpellChecker.html#suggestSimilar-java.lang.Strin
benwtrent commented on PR #11905:
URL: https://github.com/apache/lucene/pull/11905#issuecomment-1308664186
@rmuir i took your test, modified it slightly (changing number of vectors
and the assertion). It ran for 3.5 hours and failed on the old code in the
exactly correct spot (overflowing o
rmuir commented on PR #11905:
URL: https://github.com/apache/lucene/pull/11905#issuecomment-1308697281
yeah its your search at the end that triggers the issue, because checkindex
on vectors is currently too wimpy and doesn't ever call seek(). this is also an
issue that we must address here.
donnerpeter opened a new pull request, #11909:
URL: https://github.com/apache/lucene/pull/11909
add NGramFragmentChecker to quickly check whether insertions/replacements
produce strings that are even possible in the language
--
This is an automated message from the Apache Git Service.
To
rmuir opened a new issue, #11910:
URL: https://github.com/apache/lucene/issues/11910
### Description
As mentioned by @dweiss and @benwtrent in #11905, some static analysis could
help here. Maybe it can force us to do a pass reviewing other sketchy possible
bugs like this in the code
jpountz commented on PR #11875:
URL: https://github.com/apache/lucene/pull/11875#issuecomment-1309017859
I'm somewhat familiar with this code, I wonder if it could be refactored in
such a way that it would directly leverage Lucene's timeout support, e.g. can
the logic that uses live docs to
jpountz commented on issue #11702:
URL: https://github.com/apache/lucene/issues/11702#issuecomment-1309023675
> e.g., the ESQL query language/engine proposed by Elastic.
I don't think ESQL is going to be different from existing faceting support:
it will still want to use ordinals when
benwtrent commented on issue #11910:
URL: https://github.com/apache/lucene/issues/11910#issuecomment-1309035065
I think static analysis will be a significant help.
So, I did a quick check at turning on `IntLongMath` for `error-prone` and it
is noisy.
Its flagging things like t
rmuir commented on issue #11910:
URL: https://github.com/apache/lucene/issues/11910#issuecomment-1309038831
maybe we could even turn it on, and dump its output here as a one-time
thing? we can just take a pass through it, and filter out the noisy ones.
I realize it isn't ideal and doe
reta commented on PR #11875:
URL: https://github.com/apache/lucene/pull/11875#issuecomment-1309053182
> Then it should be possible to use the existing `IndexSearcher#search`
logic as-is?
Thanks @jpountz, I think that would be the best option for everyone (Lucene
/ OpenSearch / Elasti
reta closed pull request #11875: Usability improvements for timeout support in
IndexSearcher
URL: https://github.com/apache/lucene/pull/11875
--
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 spec
reta commented on issue #11874:
URL: https://github.com/apache/lucene/issues/11874#issuecomment-1309053848
Closing was "won't do"
--
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 comm
reta closed issue #11874: Usability improvements for timeout support in
IndexSearcher
URL: https://github.com/apache/lucene/issues/11874
--
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
dweiss commented on issue #11910:
URL: https://github.com/apache/lucene/issues/11910#issuecomment-1309095897
I can understand where it signals a problem but can't determine the domain
(the v - 2 can underflow if you pass v small enough)... The 2* 2 case is odd,
I'm surprised it doesn't see
dweiss commented on code in PR #11909:
URL: https://github.com/apache/lucene/pull/11909#discussion_r1018232665
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/FragmentChecker.java:
##
@@ -0,0 +1,35 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF
benwtrent commented on PR #11905:
URL: https://github.com/apache/lucene/pull/11905#issuecomment-1309129512
Updated the monster test.
Ran about 2hrs on my laptop (but I was working, in virtual meetings, etc.
during the entire run).
Confirmed it fails without this patch. This p
rmuir commented on PR #11905:
URL: https://github.com/apache/lucene/pull/11905#issuecomment-1309198710
nice. i'm fine with the changes. We can open another issue to fix the
checkindex stuff. I do really think we should do that before releasing to look
for more trouble. Also good if we can g
rmuir commented on PR #11905:
URL: https://github.com/apache/lucene/pull/11905#issuecomment-1309200544
and thanks for help battling the testing. it will get better!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
U
rmuir opened a new issue, #11911:
URL: https://github.com/apache/lucene/issues/11911
### Description
Currently checkindex only `next()'s` through the vectors. We should do some
seeking as well. It doesnt have to be intense, e.g. we could do 64 seeks (if
there's 20,000,000 docs in the
rmuir commented on issue #11702:
URL: https://github.com/apache/lucene/issues/11702#issuecomment-1309244025
> Curious why you think a document can't have multiple locations? Why
wouldn't the geo (wkt, json, wkb, protobuf) specification then not have Multi
geometry types? The reason they do
nknize commented on issue #11702:
URL: https://github.com/apache/lucene/issues/11702#issuecomment-1309269456
> because in the real world objects can only exist in one place a a time.
Except in geo search / analysis this depends on [spatial
resolution](https://en.wikipedia.org/wiki/Spa
nknize commented on issue #11702:
URL: https://github.com/apache/lucene/issues/11702#issuecomment-1309274884
> Therefore, we created a prototype that implements multi-valued binary
docvalues which works well. However, having some support for this use case
directly in Lucene is preferable, b
donnerpeter commented on code in PR #11909:
URL: https://github.com/apache/lucene/pull/11909#discussion_r1018356474
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/FragmentChecker.java:
##
@@ -0,0 +1,35 @@
+/*
+ * Licensed to the Apache Software Foundation
rmuir opened a new issue, #11912:
URL: https://github.com/apache/lucene/issues/11912
### Description
For #11905 bug, as an example, the user may get a generic exception "Seek
Past EOF".
Can we improve it to include the bogus position? For example, If you can see
that offset is
donnerpeter commented on code in PR #11909:
URL: https://github.com/apache/lucene/pull/11909#discussion_r1018359280
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/NGramFragmentChecker.java:
##
@@ -0,0 +1,181 @@
+/*
+ * Licensed to the Apache Software Foun
dweiss commented on code in PR #11909:
URL: https://github.com/apache/lucene/pull/11909#discussion_r1018368778
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/FragmentChecker.java:
##
@@ -0,0 +1,35 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF
donnerpeter commented on code in PR #11909:
URL: https://github.com/apache/lucene/pull/11909#discussion_r1018377196
##
lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/FragmentChecker.java:
##
@@ -0,0 +1,35 @@
+/*
+ * Licensed to the Apache Software Foundation
uschindler commented on issue #11912:
URL: https://github.com/apache/lucene/issues/11912#issuecomment-1309351422
As discussed in chat an hour ago:
- for MemorySegmentIndexInput it already has a reworked Exception code where
also those suppress unused warnings are gone due to some trick: w
uschindler commented on issue #11912:
URL: https://github.com/apache/lucene/issues/11912#issuecomment-1309352460
Maybe include that in 9.4.3 ?
--
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
uschindler commented on issue #11912:
URL: https://github.com/apache/lucene/issues/11912#issuecomment-1309354880
Will do that tomorrow, PR will come. We can decide to add it to 9.4 branch
to make later debugging.of broken code with recent vectors easier.
--
This is an automated message fr
stevenschlansker opened a new issue, #11913:
URL: https://github.com/apache/lucene/issues/11913
### Description
In the lucene-replicator module, the PrimaryNode does some initialization
work in the constructor. It starts with an IndexWriter provided by the
application author. At line
jtibshirani commented on issue #11911:
URL: https://github.com/apache/lucene/issues/11911#issuecomment-1309460506
This is a good idea! As a note -- in order to exercise all parts of the file
format, we'll have to perform a kNN search too through
`KnnVectorsReader#search`. If we only load `V
jpountz opened a new issue, #11914:
URL: https://github.com/apache/lucene/issues/11914
### Description
I don't understand well why `QueryTimeout` has a `isTimeoutEnabled` method.
If the timeout is not enabled, why configure a timeout at all?
### Version and environment details
jpountz commented on issue #11868:
URL: https://github.com/apache/lucene/issues/11868#issuecomment-1309896931
Let's do `FilterIndexInput` at the same time?
--
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
35 matches
Mail list logo
