HBASE-28634 Fix FuzzyRowFilter may not return data on reverse scans #6457
Conversation
This file contains hidden or 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
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
PDavid
force-pushed
the
HBASE-28634-FuzzyRowFilter-reverse-no-data
branch
from
9abb459 to
ed2e4a5
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
|
In the latest PR build two unit tests failed:
For me, these look unrelated and I was able to successfully run these tests locally. |
PDavid
force-pushed
the
HBASE-28634-FuzzyRowFilter-reverse-no-data
branch
from
ed2e4a5 to
c03e416
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Reproduced the issue with a mini cluster test.
Some expectations were not correct. Also extracted PrefixFilter.increaseLastNonMaxByte() method to PrivateCellUtil so that it can be reused in FuzzyRowFilter.
…() when the hint should contain a 0xFF As a start, when we have a reverse scan, we not only need trailing 0xff's instead of trailing 0x00's, but it seems we always need 0xff's instead of the 0x00's.
…GetNextForFuzzyRuleForward
to not start another minicluster needlessly.
PDavid
force-pushed
the
HBASE-28634-FuzzyRowFilter-reverse-no-data
branch
from
c03e416 to
348b364
Compare
stoty
approved these changes
Nov 18, 2024
hbase-server/src/test/java/org/apache/hadoop/hbase/filter/TestFuzzyRowFilter.java
Outdated
Show resolved
Hide resolved
| @Override | ||
| public ReturnCode filterCell(final Cell c) { | ||
| final int startIndex = lastFoundIndex >= 0 ? lastFoundIndex : 0; | ||
| final int startIndex = Math.max(lastFoundIndex, 0); |
There was a problem hiding this comment.
I have checked, this compiles to the same inline, so it's not a performance problem.
There was a problem hiding this comment.
Many thanks. 👍
Yes, this line change was not because of performance. Actually here just IntelliJ IDEA suggested to replace this to make it simpler to understand. It can be reverted if needed.
|
Thank you. |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
As the comment did not make much sense without the mask and current row.
|
🎊 +1 overall
This message was automatically generated. |
|
Prepared a backport PR for |
stoty
pushed a commit
that referenced
this pull request
Nov 25, 2024
…6457) Signed-off-by: Istvan Toth <[email protected]> (cherry picked from commit a735eff)
gvprathyusha6
pushed a commit
to gvprathyusha6/hbase
that referenced
this pull request
Dec 19, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
The FuzzyRowFilter reverse scan related code was added under HBASE-12183.
My Changes:
TestFuzzyRowFilteras some expectations were not correct.FuzzyRowFilter.getNextForFuzzyRule()in the reverse case:PrefixFilter.increaseLastNonMaxByte()method toPrivateCellUtilso that it can be reused inFuzzyRowFilter.0xFF. As a start, when we have a reverse scan, we not only need trailing0xff's instead of trailing0x00's, but it seems we always need0xff's instead of the0x00's.FuzzyRowFilterwhich IntelliJ IDEA suggested (final fields, simplifications).