You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@lucene.apache.org by GitBox <gi...@apache.org> on 2021/09/15 02:08:35 UTC
[GitHub] [lucene] dnhatn opened a new pull request #300: LUCENE-10106: Sort optimization wrongly skip first docs
dnhatn opened a new pull request #300:
URL: https://github.com/apache/lucene/pull/300
The first documents of subsequent segments are mistakenly skipped when sort optimization is enabled. We should initialize maxDocVisited in NumericComparator to -1 instead of 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-unsubscribe@lucene.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org
[GitHub] [lucene] mayya-sharipova commented on a change in pull request #300: LUCENE-10106: Sort optimization wrongly skip first docs
Posted by GitBox <gi...@apache.org>.
mayya-sharipova commented on a change in pull request #300:
URL: https://github.com/apache/lucene/pull/300#discussion_r709150904
##########
File path: lucene/core/src/test/org/apache/lucene/search/TestSortOptimization.java
##########
@@ -633,4 +637,84 @@ public void testPointValidation() throws IOException {
reader.close();
dir.close();
}
+
+ public void testMaxDocVisited() throws IOException {
+ Directory dir = newDirectory();
+ IndexWriter writer = new IndexWriter(dir, new IndexWriterConfig());
+ int numDocs = atLeast(10000);
+ long offset = 100 + random().nextInt(100);
+ long smallestValue = 50 + random().nextInt(50);
+ boolean flushed = false;
+ for (int i = 0; i < numDocs; ++i) {
+ Document doc = new Document();
+ doc.add(new NumericDocValuesField("my_field", i + offset));
+ doc.add(new LongPoint("my_field", i + offset));
+ writer.addDocument(doc);
+ if (i >= 5000 && flushed == false) {
+ flushed = true;
+ writer.flush();
+ // Index the smallest value to the first slot of the second segment
+ doc = new Document();
+ doc.add(new NumericDocValuesField("my_field", smallestValue));
+ doc.add(new LongPoint("my_field", smallestValue));
+ writer.addDocument(doc);
+ }
+ }
+ IndexReader reader = DirectoryReader.open(writer);
+ writer.close();
+ IndexSearcher searcher = new IndexSearcher(reader);
+ SortField sortField = new SortField("my_field", SortField.Type.LONG);
+ TopFieldDocs topDocs =
+ searcher.search(new MatchAllDocsQuery(), 1 + random().nextInt(100), new Sort(sortField));
+ FieldDoc fieldDoc = (FieldDoc) topDocs.scoreDocs[0];
+ assertEquals(smallestValue, ((Long) fieldDoc.fields[0]).intValue());
+ reader.close();
+ dir.close();
+ }
+
+ public void testRandomLong() throws IOException {
Review comment:
great test!
--
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-unsubscribe@lucene.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org
[GitHub] [lucene] dnhatn merged pull request #300: LUCENE-10106: Sort optimization wrongly skip first docs
Posted by GitBox <gi...@apache.org>.
dnhatn merged pull request #300:
URL: https://github.com/apache/lucene/pull/300
--
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-unsubscribe@lucene.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org
[GitHub] [lucene] dnhatn commented on pull request #300: LUCENE-10106: Sort optimization wrongly skip first docs
Posted by GitBox <gi...@apache.org>.
dnhatn commented on pull request #300:
URL: https://github.com/apache/lucene/pull/300#issuecomment-920014590
@romseygeek @mayya-sharipova Thanks for review.
> We also should fix this bug on https://github.com/apache/lucene-solr branch 8.x.
> Also would be good to fix it on 8.10 if not so late.
Sure, I will do 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 above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@lucene.apache.org
For additional commands, e-mail: issues-help@lucene.apache.org