Re: [PR] Make `static final Map` constants immutable [lucene]
uschindler commented on PR #13092: URL: https://github.com/apache/lucene/pull/13092#issuecomment-1943350917 Hi this won't go into 9.10 because the branch is already cut and the release started. Please move the entry to 9.11. I don't think we will have a 9.11 release anymore, as we want to go for 10.0, so I am not sure if it is worth to backport the changes. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Make `static final Map` constants immutable [lucene]
dweiss commented on PR #13092: URL: https://github.com/apache/lucene/pull/13092#issuecomment-1943480107 If the backport has been done already, I think it's good to merge it in. I'm sure something will pop up to warrant 9.11... -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Make `static final Map` constants immutable [lucene]
dweiss commented on PR #13092: URL: https://github.com/apache/lucene/pull/13092#issuecomment-1943481155 I'll move the changes entry, my bad. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Backport: Make `static final Map` constants immutable [lucene]
dweiss commented on PR #13100: URL: https://github.com/apache/lucene/pull/13100#issuecomment-1943484848 The changes.txt entry should be consistent with what's on main (same section). -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix too many open files Exception for some tests [lucene]
mikemccand commented on PR #13035: URL: https://github.com/apache/lucene/pull/13035#issuecomment-1943499386 This change looks great, and it looks like @rmuir's concern was addressed. Thank you for focusing on test reliability @easyice! I'll merge this soon 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 above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Backport: Make `static final Map` constants immutable [lucene]
dweiss merged PR #13100: URL: https://github.com/apache/lucene/pull/13100 -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
[PR] Fix test failure in #13057 [lucene]
benwtrent opened a new pull request, #13102: URL: https://github.com/apache/lucene/pull/13102 The original tests assume particular document orders & scores. To make the test more resilient to random flushes & merges, I adjusted the assertion conditions. Particularly, we verify matching ids -> score instead of relying on docIds. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
[I] UnifiedHighlighter incorrectly returns field 'X' was indexed without offsets [lucene]
mayya-sharipova opened a new issue, #13103:
URL: https://github.com/apache/lucene/issues/13103
### Description
UnifiedHighlighter based on matches incorrectly returns field 'X' was
indexed without offsets, cannot highlight
Test to reproduce:
```java
static final FieldType textType = new FieldType(TextField.TYPE_STORED);
static {
textType.setStoreTermVectors(true);
textType.setStoreTermVectorPositions(true);
textType.setStoreTermVectorOffsets(true);
textType.freeze();
}
public void testHighlgiht() {
String indexPath = "../lucene-test-indices/index1";
Path path = Paths.get(indexPath);
try {
Directory directory = NIOFSDirectory.open(path);
Analyzer analyzer = new ClassicAnalyzer();
IndexWriterConfig config = new IndexWriterConfig(analyzer);
try (IndexWriter writer = new IndexWriter(directory, config)) {
addDoc(writer, "The quick brown fox jumps over the lazy
dog");
}
try (IndexReader reader = DirectoryReader.open(directory)) {
IndexSearcher searcher = new IndexSearcher(reader);
Query query = new IntervalQuery("content",
Intervals.analyzedText("quick brown fox jumps over
the lazy dog", analyzer, "content", 0, true));
TopDocs topDocs = searcher.search(query, 10);
UnifiedHighlighter.Builder uhBuilder = new
UnifiedHighlighter.Builder(searcher, analyzer)
.withWeightMatches(true);
UnifiedHighlighter highlighter = new
UnifiedHighlighter(uhBuilder);
String[] highlights = highlighter.highlight("content",
query, topDocs, 1);
System.out.println(Arrays.toString(highlights));
}
} catch (IOException e) {
e.printStackTrace();
}
}
private static void addDoc(IndexWriter writer, String content) throws
IOException {
Document doc = new Document();
doc.add(new Field("content", content, textType));
writer.addDocument(doc);
}
```
produces an error:
```
java.lang.IllegalArgumentException: field 'content' was indexed without
offsets, cannot highlight
at
org.apache.lucene.search.uhighlight.FieldHighlighter.highlightOffsetsEnums(FieldHighlighter.java:157)
at
org.apache.lucene.search.uhighlight.FieldHighlighter.highlightFieldForDoc(FieldHighlighter.java:83)
at
org.apache.lucene.search.uhighlight.UnifiedHighlighter.highlightFieldsAsObjects(UnifiedHighlighter.java:944)
at
org.apache.lucene.search.uhighlight.UnifiedHighlighter.highlightFields(UnifiedHighlighter.java:814)
at
org.apache.lucene.search.uhighlight.UnifiedHighlighter.highlightFields(UnifiedHighlighter.java:792)
at
org.apache.lucene.search.uhighlight.UnifiedHighlighter.highlight(UnifiedHighlighter.java:725)
```
A workaround to disable highlighting based on matches:
```java
UnifiedHighlighter.Builder uhBuilder = new
UnifiedHighlighter.Builder(searcher, analyzer)
.withWeightMatches(false);
```
This happens because of `ClassicAnalyzer` that removes stop words, and
because of it usage of `ExtendedIntervalsSource` that returns -1 offsets.
### Version and environment details
Lucene v 9.9.1
--
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.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix too many open files Exception for some tests [lucene]
easyice commented on PR #13035: URL: https://github.com/apache/lucene/pull/13035#issuecomment-1944279470 @rmuir @mikemccand Thank you for reviewing! -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
[PR] Fix NPE for TestRandomChains.testRandomChainsWithLargeStrings [lucene]
easyice opened a new pull request, #13104: URL: https://github.com/apache/lucene/pull/13104 Related to LUCENE-10353 ``` ./gradlew test --tests TestRandomChains.testRandomChainsWithLargeStrings -Dtests.seed=1B915D6C476F9BFE -Dtests.nightly=true -Dtests.locale=mn -Dtests.timezone=Canada/Mountain -Dtests.asserts=true -Dtests.file.encoding=UTF-8 ``` ``` org.apache.lucene.analysis.tests.TestRandomChains > testRandomChainsWithLargeStrings FAILED java.lang.NullPointerException: Cannot invoke "String.equals(Object)" because the return value of "org.apache.lucene.analysis.tokenattributes.TypeAttribute.type()" is null at __randomizedtesting.SeedInfo.seed([1B915D6C476F9BFE:71CAE27D1E21BB0D]:0) at org.apache.lucene.analysis.common@10.0.0-SNAPSHOT/org.apache.lucene.analysis.payloads.NumericPayloadTokenFilter.incrementToken(NumericPayloadTokenFilter.java:49) at org.apache.lucene.analysis.common@10.0.0-SNAPSHOT/org.apache.lucene.analysis.miscellaneous.ConditionalTokenFilter.incrementToken(ConditionalTokenFilter.java:185) at org.apache.lucene.test_framework@10.0.0-SNAPSHOT/org.apache.lucene.tests.analysis.ValidatingTokenFilter.incrementToken(ValidatingTokenFilter.java:81) at org.apache.lucene.analysis.common@10.0.0-SNAPSHOT/org.apache.lucene.analysis.miscellaneous.ConditionalTokenFilter$OneTimeWrapper.incrementToken(ConditionalTokenFilter.java:65) at org.apache.lucene.analysis.icu@10.0.0-SNAPSHOT/org.apache.lucene.analysis.icu.ICUNormalizer2Filter.incrementToken(ICUNormalizer2Filter.java:80) at org.apache.lucene.analysis.common@10.0.0-SNAPSHOT/org.apache.lucene.analysis.miscellaneous.ConditionalTokenFilter.incrementToken(ConditionalTokenFilter.java:203) at org.apache.lucene.test_framework@10.0.0-SNAPSHOT/org.apache.lucene.tests.analysis.ValidatingTokenFilter.incrementToken(ValidatingTokenFilter.java:81) at org.apache.lucene.test_framework@10.0.0-SNAPSHOT/org.apache.lucene.tests.analysis.BaseTokenStreamTestCase.checkResetException(BaseTokenStreamTestCase.java:926) at org.apache.lucene.test_framework@10.0.0-SNAPSHOT/org.apache.lucene.tests.analysis.BaseTokenStreamTestCase.checkRandomData(BaseTokenStreamTestCase.java:1068) at org.apache.lucene.analysis.tests@10.0.0-SNAPSHOT/org.apache.lucene.analysis.tests.TestRandomChains.testRandomChainsWithLargeStrings(TestRandomChains.java:978) at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ``` -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
[I] (Byte|Float)KnnVectorFieldSource unusable if segment has no vector values [lucene]
hossman opened a new issue, #13105:
URL: https://github.com/apache/lucene/issues/13105
### Description
Like most field types in lucene, `Knn(Byte|Float)VectorField` classes do not
require a field value for every document -- they support use cases where some
documents are indexed w/o a field value.
The `(Byte|Float)KnnVectorFieldSource` implementations _generally_ follow
the pattern of most field based `ValueSource` impls in terms of how they behave
when some documents do not have a value in the specified field
(`TestKnnVectorSimilarityFunctions` even checks for this in various tests using
the `knnFloatField3` and `knnByteField3` fields which only exist in one
document).
**_But_** these value sources break if/when a user gets unlucky and a
segment exists where _none_ of the documents have value in the vector field
(and by extension: when querying an empty index)
This is due to the following check...
```
final ByteVectorValues vectorValues =
readerContext.reader().getByteVectorValues(fieldName);
if (vectorValues == null) {
throw new IllegalArgumentException(
"no byte vector value is indexed for field '" + fieldName + "'");
...
```
This problem is trivial to reproduce reliably by adding an `iw.commit();`
call to `TestKnnVectorSimilarityFunctions` between indexing the two documents.
(The `RandomIndexWriter` should eventually trigger it in this test even w/o
modifications .. not sure if it ever has in jenkins?)
The appropriate way to implement this type of check is to follow the pattern
in the `DocValues.getFoo()` methods, that use a `checkField(...)` method to
inspect the `FieldInfos` from the `LeafReader` for consistency w/expectations
if `reader.getFooDocValues()` returns null. `checkField(...)` assumes
everything is fine if no `FieldInfo` exist for this field in this reader --
using an "empty" impl in it's place.
I'm attaching a patch that fixes these `ValueSource` classes, by returning a
`VectorFieldFunction` that wraps `DocIdSetIterator.empty()` when `vectorValues`
is null, as long as a new `VectorEncoding` method `checkField(reader,
fieldName)` does not throw an exception (Adding `checkField` directly to
`VectorEncoding` seemed like the most straightforward place to put this ... not
sure if there is a more appropriate class?)
I also made the new `checkField` method throw `IllegalStateException`
instead of `IllegalArgumentException` to be consistent with other "is the index
field type consistent with the type of query you are trying to do?" checks in
the code base (like `DocValues.checkField(...)` and updated
`TestKnnVectorSimilarityFunctions` accordingly.
Patch:
[VectorFieldSource.fix-missing-check.patch.txt](https://github.com/apache/lucene/files/14284577/VectorFieldSource.fix-missing-check.patch.txt)
### Version and environment details
_No response_
--
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.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix NPE for TestRandomChains.testRandomChainsWithLargeStrings [lucene]
rmuir commented on PR #13104: URL: https://github.com/apache/lucene/pull/13104#issuecomment-1944453765 good catch! should we move this check to `setTokenType()`? Since it is a `public` method, it seems like a better place to add the check? -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Cleanup TokenizedPhraseQueryNode code [lucene]
stefanvodita commented on code in PR #13041:
URL: https://github.com/apache/lucene/pull/13041#discussion_r1489984421
##
lucene/queryparser/src/java/org/apache/lucene/queryparser/flexible/core/nodes/TokenizedPhraseQueryNode.java:
##
@@ -70,10 +70,8 @@ public QueryNode cloneTree() throws
CloneNotSupportedException {
@Override
public CharSequence getField() {
List children = getChildren();
-
-if (children == null || children.size() == 0) {
+if (children == null || children.isEmpty()) {
return null;
-
} else {
return ((FieldableNode) children.get(0)).getField();
Review Comment:
I think having the two methods aligned, the way they are in your latest
commit, makes sense.
--
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.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Cleanup TokenizedPhraseQueryNode code [lucene]
stefanvodita commented on code in PR #13041: URL: https://github.com/apache/lucene/pull/13041#discussion_r1489985337 ## lucene/CHANGES.txt: ## @@ -258,6 +258,8 @@ Improvements Tests are running with random byte order to ensure that the order does not affect correctness of code. Native order was enabled for LZ4 compression. (Uwe Schindler) +* GITHUB#13041: TokenizedPhraseQueryNode code cleanup (Dmitry Cherniachenko) Review Comment: This would have to move up to version 9.11 because 9.10 is already in the release process. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Remove redundant field initializers [lucene]
stefanvodita commented on code in PR #13060:
URL: https://github.com/apache/lucene/pull/13060#discussion_r1490019625
##
lucene/analysis/phonetic/src/java/org/apache/lucene/analysis/phonetic/DaitchMokotoffSoundexFilter.java:
##
@@ -32,10 +32,10 @@
*/
public final class DaitchMokotoffSoundexFilter extends TokenFilter {
/** true if encoded tokens should be added as synonyms */
- protected boolean inject = true;
+ final boolean inject;
Review Comment:
Since this class is marked with `@lucene.experimental`, we can make the
change.
##
lucene/analysis/phonetic/src/java/org/apache/lucene/analysis/phonetic/PhoneticFilter.java:
##
@@ -32,13 +32,13 @@
*/
public final class PhoneticFilter extends TokenFilter {
/** true if encoded tokens should be added as synonyms */
- protected boolean inject = true;
+ final boolean inject;
Review Comment:
It's a strange case. We're not really breaking backwards compatibility,
since no one could have accessed this field anyway with the class being final.
--
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.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix addBackcompatIndexes.py to properly generate missing versions [lucene]
s1monw commented on PR #13095: URL: https://github.com/apache/lucene/pull/13095#issuecomment-1944559092 > P.S.: I don't know if we still need those 4 special case indexes. They seem to be a relic from earlier times. As we only check that the indexes throw an Exception on loading, the already existing 5.x and 4.x indexes are enough. true, I will merge this one and open a new pr to remove these -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix addBackcompatIndexes.py to properly generate missing versions [lucene]
s1monw merged PR #13095: URL: https://github.com/apache/lucene/pull/13095 -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix addBackcompatIndexes.py to properly generate missing versions [lucene]
uschindler commented on PR #13095: URL: https://github.com/apache/lucene/pull/13095#issuecomment-1944578996 This was merged to 9.x only. Should we not have it in main, too? -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] Python backward compat scripts need updating [lucene]
s1monw closed issue #13094: Python backward compat scripts need updating URL: https://github.com/apache/lucene/issues/13094 -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix NPE for TestRandomChains.testRandomChainsWithLargeStrings [lucene]
uschindler commented on PR #13104: URL: https://github.com/apache/lucene/pull/13104#issuecomment-1944583551 I wonder why the token stream has a setter at all. The field should be final. I thought we want to have those settings all final and only set on construction. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix addBackcompatIndexes.py to properly generate missing versions [lucene]
s1monw commented on PR #13095: URL: https://github.com/apache/lucene/pull/13095#issuecomment-1944584231 @uschindler dude I need to press buttons it took me 14 mins to port I guess that's ok? -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Fix addBackcompatIndexes.py to properly generate missing versions [lucene]
uschindler commented on PR #13095: URL: https://github.com/apache/lucene/pull/13095#issuecomment-1944599169 Sorry, I was just irritated, because the original PR was opened against stable 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 to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] No static method version() crash in android project [lucene]
uschindler commented on issue #13078: URL: https://github.com/apache/lucene/issues/13078#issuecomment-1944674586 Android is not a supported platform. The official `java.lang.Runtime` class has a static `version()` method. So your setup does not follow the Java specification and we won't change this. Reference: https://docs.oracle.com/en/java/javase/17/docs/api/java.base/java/lang/Runtime.html#version() Please report this to the Dalvik runtime. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] No static method version() crash in android project [lucene]
uschindler closed issue #13078: No static method version() crash in android project URL: https://github.com/apache/lucene/issues/13078 -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] AnalysisSPILoader should refresh cached service when a different classloader is used [lucene]
uschindler commented on issue #13101: URL: https://github.com/apache/lucene/issues/13101#issuecomment-1944717416 This is a general problem of Service providers in Java. As Lucene only has a single instance in the parent classloader (Lucene is part of Solr's classloader) it only has a single lookup to find analysis components. Your proposal will cause the same problem if other cores. If you unload a core and it removes it's analysis components, the anal is in another core will fail from that point on. To fix this correctly, Solr would need to have a separate Lucene impl per core. This is undoable. In general the per core classpaths in Solr are strange. Codecs, PostingsFormats, DocvaluesFormats and analysis components should be loaded in same classloader like Solr and Lucene itself. There is no solution to this well known problem. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Move `brToString(BytesRef)` to `ToStringUtils` [lucene]
uschindler commented on PR #13068: URL: https://github.com/apache/lucene/pull/13068#issuecomment-1944770945 Please move entry to 9.11.0 -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [PR] Move `brToString(BytesRef)` to `ToStringUtils` [lucene]
uschindler commented on PR #13068: URL: https://github.com/apache/lucene/pull/13068#issuecomment-1944773941 Will take care of this tomorrow. Sorry for delay. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] (Byte|Float)KnnVectorFieldSource unusable if segment has no vector values [lucene]
hossman commented on issue #13105: URL: https://github.com/apache/lucene/issues/13105#issuecomment-1945211363 > ... as long as a new VectorEncoding method checkField(reader, fieldName) does not throw an exception (Adding checkField directly to VectorEncoding seemed like the most straightforward place to put this ... not sure if there is a more appropriate class?) The fact that this new method (in an otherwise bare bones `enum` class) had to be public kept gnawing at me -- so instead I refactored it into a package protected static method in `VectorFieldFunction`. I also added a simple test of the "field exists, but was not indexed with vectors" situation, and was surprised to discover that `FieldInfo.getVectorEncoding()` seems to default to `FLOAT32` (instead of null) when a field has no vectors -- so I added an explicit check of `FieldInfo.hasVectorValues()` Updated Patch: [VectorFieldSource.fix-missing-check.patch.txt](https://github.com/apache/lucene/files/14290130/VectorFieldSource.fix-missing-check.patch.txt) -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] AnalysisSPILoader should refresh cached service when a different classloader is used [lucene]
AndreyBozhko commented on issue #13101: URL: https://github.com/apache/lucene/issues/13101#issuecomment-1945225900 Thanks for taking a look @uschindler - I have to admit that I didn't fully grasp your comment about having a separate Lucene per core. My original understanding was that for the example I described, the loading and reloading of SPI occurs during the Solr node (and not core) startup. So very early in Solr node's lifecycle, it would load all Lucene classes (makes total sense that these are per-node), and also load other libraries (user-provided plugins) shared at the node level. But even if we only deal with shared classes, users can still encounter a `ClassNotFoundException` similar to the example above. @HoustonPutman Please correct me if there's something that I misinterpreted about the issue from our initial discussion. -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
Re: [I] AnalysisSPILoader should refresh cached service when a different classloader is used [lucene]
AndreyBozhko commented on issue #13101: URL: https://github.com/apache/lucene/issues/13101#issuecomment-1945226818 I do realize that the issue title may have sounded too prescriptive - I certainly didn't mean that. But I would definitely like to have a discussion on what path forward (if any) makes the most sense. I realize that the feedback may be along the lines of _"The behavior of Lucene's AnalysisSPILoader is set in stone, and Solr should just work around that"_. Or, as a hypothetical alternative, Lucene could expose a way to invalidate the SPI cache, and applications like Solr may choose to use that (having full control, but also full responsibility over what they're doing). Or perhaps there are other ways to go about this. What are your thoughts? -- 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.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
