iverase commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1294716060
I will say it again, please move the factory methods to the docvalues
classes and make the constructor private. It makes no-sense to have them in
LatLonShape / XYShape.
--
This is an
rmuir commented on issue #11887:
URL: https://github.com/apache/lucene/issues/11887#issuecomment-1294755250
this test is slower for me quite often as well, tends to take around entire
minute in many cases.
--
This is an automated message from the Apache Git Service.
To respond to the mess
rmuir commented on PR #11886:
URL: https://github.com/apache/lucene/pull/11886#issuecomment-1294761677
i guess we could allow users to run it with java 18 now
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL ab
dweiss commented on PR #11886:
URL: https://github.com/apache/lucene/pull/11886#issuecomment-1294775595
I haven't had the time to test but it may indeed work out of the box.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and
vsop-479 commented on PR #11722:
URL: https://github.com/apache/lucene/pull/11722#issuecomment-1294803341
@jpountz I have fixed the test failures.
But I cant find the new commit in this PR (since it has been merged and
reverted?), do I need to open an new PR or some what to show the new
jpountz commented on PR #11722:
URL: https://github.com/apache/lucene/pull/11722#issuecomment-1294864473
A new PR would be great, thank you for looking into these failures!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and u
vsop-479 opened a new pull request, #11888:
URL: https://github.com/apache/lucene/pull/11888
Fix https://github.com/apache/lucene/pull/11722 's bug.
--
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
vsop-479 commented on PR #11888:
URL: https://github.com/apache/lucene/pull/11888#issuecomment-1294898679
@jpountz I have fixed the code and passed the tests, please have a review.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitH
nknize commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1294930300
> ... and make the constructor private. It makes no-sense to have them in
LatLonShape / XYShape.
The ctor is already protected. It does make sense in that LatLonShape and
XYSha
iverase commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1294972472
I strongly believe that LatLonShape /XYShape should only contain factory
methods that work on top of the BKD tree index. Factory methods that work on
doc values should be moved to the Do
rmuir commented on code in PR #11882:
URL: https://github.com/apache/lucene/pull/11882#discussion_r1008059292
##
lucene/build.gradle:
##
@@ -15,6 +15,8 @@
* limitations under the License.
*/
+apply plugin: 'java'
Review Comment:
> Yeah, I think it's broken somewhere an
rmuir opened a new pull request, #11889:
URL: https://github.com/apache/lucene/pull/11889
Since #11886, java 18 actually works, so let's allow it. unfortunately java
19 still does not work.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log
stefanvodita commented on PR #11815:
URL: https://github.com/apache/lucene/pull/11815#issuecomment-1295011134
Thank you for the review @mikemccand, I’ve incorporated your feedback into
the new commit.
--
This is an automated message from the Apache Git Service.
To respond to the message,
dweiss commented on code in PR #11882:
URL: https://github.com/apache/lucene/pull/11882#discussion_r1008085249
##
lucene/build.gradle:
##
@@ -15,6 +15,8 @@
* limitations under the License.
*/
+apply plugin: 'java'
Review Comment:
I'll be looking at it later today - cur
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295022832
> Based on the commit that added this check,
[c897aac](https://github.com/apache/lucene/commit/c897aac0773d7c14fedceefa26511b97866ad587)
gradlew and gradlew.bat should be updated as well.
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295024717
i'm gonna fix that too. but for the future i wonder if we need that
"generic" message anymore... or at least versions in it: that was kinda a
previous workaround for gradle's confusing err
dweiss commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295028109
>i'm gonna fix that too. but for the future i wonder if we need that
"generic" message anymore... or at least versions in it: that was kinda a
previous workaround for gradle's confusing e
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295032948
unrelated to this PR, does anyone know what happened with the "run
distribution test" check?
it hit connect timeout and failed, this isn't "our" downloading, there's no
`org.apache.
uschindler commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295033229
I restarted the Github actions. Was some timeout error.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
UR
uschindler commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295036904
> I restarted the Github actions. Was some timeout error.
Was a temporary problem with Gradle's sörvers. 2nd try worked.
--
This is an automated message from the Apache Git Se
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295038154
@uschindler i know, but in order to prevent flaky builds, we should have
backoff and retry to these type of requests.
--
This is an automated message from the Apache Git Service.
To res
uschindler commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295040553
> @uschindler i know, but in order to prevent flaky builds, we should have
backoff and retry to these type of requests.
Sometimes also Apache Git is down then Jenkins fails. Who
dweiss commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295041468
This is gradle bootstrapping itself (new version). It's the gradle-wrapper
that downloads the distribution. I'm not sure how we could retry it on network
failures, unfortunately. The down
nknize commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295043124
> I strongly believe that LatLonShape /XYShape should only contain factory
methods that work on top of the BKD tree index
I see what you're saying, and I really like that idea of se
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295046872
> This is gradle bootstrapping itself (new version). It's the gradle-wrapper
that downloads the distribution. I'm not sure how we could retry it on network
failures, unfortunately. The dow
uschindler commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295048862
> but if theres a way to add `try/catch` to some java code or even `download
|| (sleep 10 && download) || (sleep 10 && download)` to prevent builds from
having annoyingly noisy fails,
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295050135
> You would need to do that around each Maven Central call :-) This was just
a bad incidence that it failed. I don't care if the build fails from time to
time. It does it for other reasons
rmuir commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295051733
ok, ill ignore it. but i'm not happy about flaky builds because someone used
the network naively!
--
This is an automated message from the Apache Git Service.
To respond to the message,
nknize commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295061435
If we can't think of anything quickly I think a good compromise would be to
merge this long running PR so @navneet1v can use the field until we figure out
if there even needs to be a sepa
gf2121 commented on PR #602:
URL: https://github.com/apache/lucene/pull/602#issuecomment-1295063743
Here is the nightly benchmark result of 2022.10.24 ([commits
diff](https://github.com/apache/lucene/compare/0f525bfb14ca7106fe1fae916298b9eb4c755608...b1d1e488f2c126cef5067b3ffddb5dcb5db633aa)
iverase commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295082525
I have a similar discussion not long time ago.
At the moment, once you create a ShapeDocValues, you don't know if is geo or
cartesian because the specific implementations are hidd
nknize commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295095640
> once you create a ShapeDocValues, you don't know if is geo or cartesian
because the specific implementations are hidden from the user
I'm not following what you mean by this. The
iverase commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295101536
Then I don't understand why you cannot move those methods to that classes. I
understood from your last comment that they were hidden.
--
This is an automated message from the Apache Gi
iverase commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295112835
I lost myself here:
First:
> I see what you're saying, and I really like that idea of separating
DocValues and BKD factory methods to their own factory classes. Except
curr
dweiss commented on code in PR #11882:
URL: https://github.com/apache/lucene/pull/11882#discussion_r1008182151
##
lucene/build.gradle:
##
@@ -15,6 +15,8 @@
* limitations under the License.
*/
+apply plugin: 'java'
Review Comment:
Yeah - we were looking at the main bran
dweiss commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295155894
Now task dependencies seem to work as I'd expect them to. I don't know how
to read the coverage information at the top level when one only runs a set of
subprojects. For example:
```
dweiss commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295159626
Now task dependencies seem to work as I'd expect them to. I don't know how
to read the coverage information at the top level when one only runs a set of
subprojects. For example:
```
dweiss commented on PR #11889:
URL: https://github.com/apache/lucene/pull/11889#issuecomment-1295163562
> usually package managers have robust retry behavior like this already.
It's something that would have to be fixed in the gradle wrapper, I think.
:(
--
This is an automated m
rmuir commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295167465
> The "aggregated" view only makes sense when you actually run all the
tests I think this may be weird to just me though, I wouldn't worry about
it.
Maybe there is some way to d
nknize commented on code in PR #11753:
URL: https://github.com/apache/lucene/pull/11753#discussion_r1008219749
##
lucene/core/src/java/org/apache/lucene/document/XYShape.java:
##
@@ -269,4 +269,14 @@ public static Query newGeometryQuery(
}
return new XYShapeQuery(field
nknize merged PR #11753:
URL: https://github.com/apache/lucene/pull/11753
--
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.apac
navneet1v commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295347670
@nknize can we backport this to 9.x branch
--
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
nknize commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295348872
There's no automation for that. Just check out `branch_9x` and cherry-pick
the commit and open a backport PR.
--
This is an automated message from the Apache Git Service.
To respond to
navneet1v commented on PR #11753:
URL: https://github.com/apache/lucene/pull/11753#issuecomment-1295361055
> There's no automation for that. Just check out `branch_9x` and cherry-pick
the commit and open a backport PR.
wasn't aware of that.
--
This is an automated message from the
navneet1v opened a new pull request, #11890:
URL: https://github.com/apache/lucene/pull/11890
### Description
Backport of https://github.com/apache/lucene/pull/11753 to 9x
Added interface to relate a LatLonShape with another shape represented as
Component2D.
--
This
dweiss commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295377906
oh - it will work that way if you run coverage on a subproject (see my
explicit top-level task above). even then it can be a bit weird if you use the
filter (--tests) but I guess it's not
dweiss commented on PR #27:
URL: https://github.com/apache/lucene/pull/27#issuecomment-1295379045
you can move it back to benchmarks if it's used only from there.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
UR
dweiss commented on PR #27:
URL: https://github.com/apache/lucene/pull/27#issuecomment-1295385471
> I didn't see them where I expected them to be
Btw. you should use 'gradlew -p lucene/benchmark tasks' for such queries -
then the location of task definitions are less important? If yo
gf2121 opened a new issue, #11891:
URL: https://github.com/apache/lucene/issues/11891
### Description
In https://github.com/apache/lucene/issues/11202 we introduced a
MergeInstance of DocValuesProducer to speed up merge by bulk decoding. If i
understood correctly, MergeInstance shoul
gf2121 closed issue #11891: DocValuesProducer#getMergeInstance can not speed up
Browse*Facets
URL: https://github.com/apache/lucene/issues/11891
--
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 s
nknize merged PR #11890:
URL: https://github.com/apache/lucene/pull/11890
--
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.apac
rmuir commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295505148
Personally I don't think its confusing with the filter (`--tests`). When
writing new functionality/tests, I like to iterate and see what branches etc of
the code the new test is covering.
jtibshirani commented on code in PR #11860:
URL: https://github.com/apache/lucene/pull/11860#discussion_r1008594717
##
lucene/core/src/java/org/apache/lucene/codecs/lucene95/Lucene95HnswVectorsReader.java:
##
@@ -0,0 +1,505 @@
+/*
+ * Licensed to the Apache Software Foundation (
MarcusSorealheis commented on PR #874:
URL: https://github.com/apache/lucene/pull/874#issuecomment-1295674939
I think slow indexing throughput is a pain that customers ought to surface.
If they find that they mostly use vectors for use cases that don't have
nrt-scaling and replication requi
rmuir merged PR #11889:
URL: https://github.com/apache/lucene/pull/11889
--
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.apach
vigyasharma commented on PR #11875:
URL: https://github.com/apache/lucene/pull/11875#issuecomment-1295741599
> but I'd prefer not to make other implementation details like the custom
bulk scorer public, why do you need this?
I think they want to use they want to wrap their custom Bulk
vigyasharma commented on issue #11676:
URL: https://github.com/apache/lucene/issues/11676#issuecomment-1295746869
I would like to help out with this one..
> Could we grow the window of scored doc IDs exponentially, maybe with
guarantees such as making sure that the new window is at mo
dweiss commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295749455
That's fine for me. And I think this patch also fixes the problem of test
tasks running needlessly - I'm +1 to add it, I don't think I have anything else
to contribute here (?).
--
Thi
zhaih commented on PR #11882:
URL: https://github.com/apache/lucene/pull/11882#issuecomment-1295749859
Again, thank you @dweiss a lot for helping! I'll push it.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL a
zhaih merged PR #11882:
URL: https://github.com/apache/lucene/pull/11882
--
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.apach
zhaih closed issue #11839: gradle aggregate coverage report
URL: https://github.com/apache/lucene/issues/11839
--
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
61 matches
Mail list logo
