vsop-479 commented on PR #12846:
URL: https://github.com/apache/lucene/pull/12846#issuecomment-1842384044
> is this new approach helping cover more cases maybe?
Yes, Previous patch only apply copy for level 0 's acc in
``Lucene99SkipWriter.bufferSkip``, Current patch apply copy for
uschindler commented on issue #12863:
URL: https://github.com/apache/lucene/issues/12863#issuecomment-1842448366
> Thank you very much @rmuir, @uschindler. Yes, there are many threads
running at the same time and we have a managed reference count to close the
reader, maybe there are some is
dweiss commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1416950035
##
lucene/core/src/java/org/apache/lucene/util/fst/ReadWriteDataOutput.java:
##
@@ -0,0 +1,82 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or m
dungba88 commented on code in PR #12624:
URL: https://github.com/apache/lucene/pull/12624#discussion_r1416980241
##
lucene/core/src/java/org/apache/lucene/util/fst/ReadWriteDataOutput.java:
##
@@ -0,0 +1,82 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or
shubhamvishu opened a new pull request, #12880:
URL: https://github.com/apache/lucene/pull/12880
### Description
This PR removes(or cleans up) some of the redundant modifiers from the code.
Addresses the below warning in my IDE :
- Modifier ‘static’ is redundant for inner inter
slow-J commented on PR #12876:
URL: https://github.com/apache/lucene/pull/12876#issuecomment-1842557831
> I'm not sure about the removal of these two functions, they're quite nice,
only taking a CodecReader and an InfoStream. Removing them is forcing users to
call the more complex functions
uschindler merged PR #12878:
URL: https://github.com/apache/lucene/pull/12878
--
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: issues-unsubscr...@lucene.
dweiss opened a new pull request, #12881:
URL: https://github.com/apache/lucene/pull/12881
This patch provides a number of small improvements aimed at improving
performance of MatchHighlighter (and MatchRegionRetriever), especially in
corner cases like:
* queries that result in a lar
benwtrent commented on issue #12091:
URL: https://github.com/apache/lucene/issues/12091#issuecomment-1843036132
This has been added. A MR-JAR for vector operations was added and Lucene can
now take advantage of the Panama Vector API for vector similarity functions.
--
This is an automated
benwtrent closed issue #12091: Speeding up Lucene Vector Similarity through the
Java Vector API
URL: https://github.com/apache/lucene/issues/12091
--
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
benwtrent commented on issue #11758:
URL: https://github.com/apache/lucene/issues/11758#issuecomment-1843041904
There are multiple simplifications now in the quantized byte operations.
One key thing is that the graph searcher now takes a "vector scorer" object
and doesn't care about
benwtrent closed issue #11758: Follow-up refactors to 8-bit quantization change
URL: https://github.com/apache/lucene/issues/11758
--
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
benwtrent commented on issue #11440:
URL: https://github.com/apache/lucene/issues/11440#issuecomment-1843050638
While exploring other things, I noticed this:
https://github.com/apache/lucene/pull/12789
There is definitely room for improvement here. We should test:
- when to use Sp
benwtrent commented on issue #12219:
URL: https://github.com/apache/lucene/issues/12219#issuecomment-1843052002
We can close this, we added panama vector API to Lucene directly, that was
my main concern with this issue.
--
This is an automated message from the Apache Git Service.
To respo
benwtrent closed issue #12219: Making vector similarity functions pluggable
URL: https://github.com/apache/lucene/issues/12219
--
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.
T
benwtrent closed issue #12497: Add Scalar Quantization codec for Vectors
URL: https://github.com/apache/lucene/issues/12497
--
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 u
benwtrent commented on issue #12497:
URL: https://github.com/apache/lucene/issues/12497#issuecomment-1843057160
This was shipped in Lucene 9.9 as the
`Lucene99HnswScalarQuantizedVectorsFormat` codec format!
--
This is an automated message from the Apache Git Service.
To respond to the mes
romseygeek commented on code in PR #12881:
URL: https://github.com/apache/lucene/pull/12881#discussion_r1417479284
##
lucene/highlighter/src/java/org/apache/lucene/search/matchhighlight/MatchRegionRetriever.java:
##
@@ -199,34 +343,95 @@ public void highlightDocument(
Lea
dweiss commented on code in PR #12881:
URL: https://github.com/apache/lucene/pull/12881#discussion_r1417504333
##
lucene/highlighter/src/java/org/apache/lucene/search/matchhighlight/MatchRegionRetriever.java:
##
@@ -53,99 +58,173 @@
public class MatchRegionRetriever {
privat
dweiss commented on code in PR #12881:
URL: https://github.com/apache/lucene/pull/12881#discussion_r1417504745
##
lucene/highlighter/src/java/org/apache/lucene/search/matchhighlight/MatchRegionRetriever.java:
##
@@ -199,34 +343,95 @@ public void highlightDocument(
LeafRea
hurutoriya opened a new pull request, #12882:
URL: https://github.com/apache/lucene/pull/12882
### Description
https://github.com/apache/lucene/issues/11948
>>>smoketester leaks a GPG agent on my computer everytime it runs. @risdenk
pointed out this fix from solr:
https://
hurutoriya commented on issue #9705:
URL: https://github.com/apache/lucene/issues/9705#issuecomment-1843259962
Hello @tteofili . It seems OpenNLP is upgraded to 1.9.1+.
https://github.com/apache/lucene/commit/c228e4bb66ca73c8150d8eaebe2bb999bcc6c9b1
Can we close this issue to clean up
cpoerschke closed issue #9705: Upgrade OpenNLP to 1.9.1 [LUCENE-8659]
URL: https://github.com/apache/lucene/issues/9705
--
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 unsub
cpoerschke commented on issue #9705:
URL: https://github.com/apache/lucene/issues/9705#issuecomment-1843296060
Thanks @hurutoriya for re-surfacing this issue!
Closing as superseded by #11277 upgrading to 1.9.4 version.
--
This is an automated message from the Apache Git Service.
To
gsmiller commented on PR #12880:
URL: https://github.com/apache/lucene/pull/12880#issuecomment-1843430023
Thanks @shubhamvishu! I personally like this cleanup, but I'd be curious if
others have some reasons I'm not aware of why they may prefer differently?
Maybe more importantly though, I w
dweiss opened a new issue, #12883:
URL: https://github.com/apache/lucene/issues/12883
### Description
I came across this one by stress-testing the package with -Dtests.iters=200.
Occasional hiccups in unified highlighter tests, for example:
```
gradlew :lucene:highlighter:test -
dweiss commented on code in PR #12881:
URL: https://github.com/apache/lucene/pull/12881#discussion_r1417928882
##
lucene/highlighter/src/java/org/apache/lucene/search/matchhighlight/MatchRegionRetriever.java:
##
@@ -199,34 +343,95 @@ public void highlightDocument(
LeafRea
javanna commented on PR #240:
URL: https://github.com/apache/lucene/pull/240#issuecomment-1843624155
I think that it would be even better to get the deprecation out with 9x,
regardless of whether we will effectively remove with 10. At least we let users
know that they should use a different
dweiss commented on code in PR #12881:
URL: https://github.com/apache/lucene/pull/12881#discussion_r1417931314
##
lucene/highlighter/src/java/org/apache/lucene/search/matchhighlight/MatchRegionRetriever.java:
##
@@ -53,99 +58,173 @@
public class MatchRegionRetriever {
privat
dweiss commented on PR #12880:
URL: https://github.com/apache/lucene/pull/12880#issuecomment-1843774013
I don't think we have any conventions for this. And I don't know if tools
can automate this.
I remember reading a thread somewhere on the openjdk mailing list that
touched on a si
uschindler commented on PR #12841:
URL: https://github.com/apache/lucene/pull/12841#issuecomment-1843848755
Hi Adrien,
Thanks for the more insight!
> @uschindler FYI this is what I'm getting:
https://gist.github.com/jpountz/be81b1eb93c6118aac65c3679911f1d8. There are two
files
uschindler commented on PR #12841:
URL: https://github.com/apache/lucene/pull/12841#issuecomment-1843857763
The best idea that I have instead of VarHandles: Create an implementation
for ByteBuffer (using the methods available there). This implementation would
work with:
- byte arrays: Us
sosohu commented on issue #12863:
URL: https://github.com/apache/lucene/issues/12863#issuecomment-1844190152
Thanks @uschindler for your reply, really apprecicate your help!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
easyice commented on PR #12841:
URL: https://github.com/apache/lucene/pull/12841#issuecomment-1844310878
@uschindler Thanks for review!
> With unaligned random reads you mean that you read with positional reads
from the area where the buffer is saved? With default DataInput you can on
dweiss commented on PR #12881:
URL: https://github.com/apache/lucene/pull/12881#issuecomment-1844803975
I'll commit this later today, if there are no objections.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
shubhamvishu commented on PR #12880:
URL: https://github.com/apache/lucene/pull/12880#issuecomment-1844826409
Thanks @gsmiller @dweiss for taking a look. I see some great points raised
here by both of you(and I agree to all). A couple of points why I think we
should go with this change :
36 matches
Mail list logo
