MatchingDirectoryReader should not use a threaded searcher #100527
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
elasticsearchmachine
added
needs:triage
romseygeek
added
>test
|
Pinging @elastic/es-search (Team:Search) |
benwtrent
approved these changes
Oct 9, 2023
romseygeek
added a commit
to romseygeek/elasticsearch
that referenced
this pull request
Oct 10, 2023
…00527) MatchingDirectoryReader is a test reader wrapper that filters out documents matching a particular query. For each leaf, we create an IndexSearcher, execute the query against it and then use that as a filter for the leaf. This searcher is created using LuceneTestCase.newSearcher() and as such may be multi- threaded, which triggers extra index checks. For tests that are expecting certain methods to be called against internal readers a given number of times, these extra checks can add additional calls which then lead to a failure of test assumptions. Because this IndexSearcher is only executed against a single leaf it will only ever use a single thread, and so we can explicitly disable threading here. Fixes elastic#100487 Fixes elastic#99916
romseygeek
added a commit
that referenced
this pull request
Oct 10, 2023
…100597) MatchingDirectoryReader is a test reader wrapper that filters out documents matching a particular query. For each leaf, we create an IndexSearcher, execute the query against it and then use that as a filter for the leaf. This searcher is created using LuceneTestCase.newSearcher() and as such may be multi- threaded, which triggers extra index checks. For tests that are expecting certain methods to be called against internal readers a given number of times, these extra checks can add additional calls which then lead to a failure of test assumptions. Because this IndexSearcher is only executed against a single leaf it will only ever use a single thread, and so we can explicitly disable threading here. Fixes #100487 Fixes #99916 Fixes #100460
|
Thanks for fixing this! |
romseygeek
added a commit
that referenced
this pull request
Oct 11, 2023
…00668) Follow up to #100527 We are not testing anything to do with searching with this searcher, and so there is no point in using LuceneTestCase.newSearcher() which will wrap it with all sorts of extra checks that may access the underlying reader in ways that are not anticipated by tests. Fixes #100460 Fixes #99024
romseygeek
added a commit
to romseygeek/elasticsearch
that referenced
this pull request
Oct 11, 2023
…astic#100668) Follow up to elastic#100527 We are not testing anything to do with searching with this searcher, and so there is no point in using LuceneTestCase.newSearcher() which will wrap it with all sorts of extra checks that may access the underlying reader in ways that are not anticipated by tests. Fixes elastic#100460 Fixes elastic#99024
elasticsearchmachine
pushed a commit
that referenced
this pull request
Oct 11, 2023
…00668) (#100685) Follow up to #100527 We are not testing anything to do with searching with this searcher, and so there is no point in using LuceneTestCase.newSearcher() which will wrap it with all sorts of extra checks that may access the underlying reader in ways that are not anticipated by tests. Fixes #100460 Fixes #99024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
MatchingDirectoryReader is a test reader wrapper that filters out documents
matching a particular query. For each leaf, we create an IndexSearcher,
execute the query against it and then use that as a filter for the leaf. This searcher
is created using
LuceneTestCase.newSearcher()and as such may be multi-threaded, which triggers extra index checks. For tests that are expecting certain
methods to be called against internal readers a given number of times, these extra
checks can add additional calls which then lead to a failure of test assumptions.
Because this IndexSearcher is only executed against a single leaf it will only ever use
a single thread, and so we can explicitly disable threading here.
Fixes #100487
Fixes #99916