Apache9 commented on pull request #3688:
URL: https://github.com/apache/hbase/pull/3688#issuecomment-925854591
I tried to move this block of code
```
if (storeLimit > -1 && this.countPerRow >= (storeLimit +
storeOffset)) {
// do what SEEK_NEXT_ROW does.
if (!matcher.moreRowsMayExistAfter(cell)) {
close(false);// Do all cleanup except heap.close()
return
scannerContext.setScannerState(NextState.NO_MORE_VALUES).hasMoreValues();
}
matcher.clearCurrentRow();
seekToNextRow(cell);
break LOOP;
}
```
Into the `if (this.countPerRow > storeOffset) {` block, at the last of this
if block.
And also, changed the condition from `this.countPerRow > (storeLimit +
storeOffset)` to `this.countPerRow >= (storeLimit + storeOffset)`. It could
also pass the added UT.
Not sure if there are other side effects... Let me try to run all the UTs
with this patch.
For me, it is a bit strange that why we need to read an extra cell and then
drop it, why not just break the LOOP when we get enough cells. That's why I
changed the condition from > to >=.
Thanks.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
