Use Arrays.compareUnsigned in IDVersionSegmentTermsEnum and OrdsSegmentTermsEnum.

[vsop-479]

Description

Description
This change similars to #13252.

[vsop-479]

@uschindler
Please take a look when you get a chance.

[vsop-479]

@jpountz
I think this check failure relats to 5045d3c .

[jpountz]

Indeed it likely does, can you merge main back into your branch?

[uschindler]

Hi,
We had others like this in former PRs (e.g., #13252). As this old code looks like copy pasted over and over, can we find other occurrences using some regex searches?

[uschindler]

Backport or not?

[jpountz]

I would backport as it looks pretty safe, though I doubt we have anyone using this postings format, so it likely doesn't matter much in practice.

[vsop-479]

can you merge main back into your branch?

Merged.

can we find other occurrences using some regex searches?

I searched code with final int targetUptoMid = targetUpto which these iterating compare start with.
There are only two places remains in org.apache.lucene.backward_codecs.lucene40.blocktree.SegmentTermsEnum now, since this is a bakward class, I am not sure we should modify it.

[uschindler]

I would backport as it looks pretty safe, though I doubt we have anyone using this postings format, so it likely doesn't matter much in practice.

The we should move the changes entry. Do we have a corresponding 9.x section already?

[vsop-479]

The we should move the changes entry. Do we have a corresponding 9.x section already?

We already have a similar change entry for #13252 under 9.11.0, so we should move this change entry?

[uschindler]

The we should move the changes entry. Do we have a corresponding 9.x section already?

We already have a similar change entry for #13252 under 9.11.0, so we should move this change entry?

Yes, let's merge them an list both PRs.

[vsop-479]

Ok, I will do it.

[vsop-479]

let's merge them an list both PRs.

I merged them into one entry.

[vsop-479]

can we find other occurrences using some regex searches?

Hmm, I also find some suffix's loop comparing in IDVersionSegmentTermsEnumFrame and OrdsSegmentTermsEnumFrame, similar to SegmentTermsEnumFrame, these code maybe could replaced by Arrays#compareUnsigned too.

But, before that I find we attempt to load sub block when scanning a leaf block in IDVersionSegmentTermsEnumFrame#scanToTermLeaf: #13786.

[vsop-479]

can we find other occurrences using some regex searches?

I replaced loop compare suffixe with Arrays#compareUnsigned in IDVersionSegmentTermsEnumFrame and OrdsSegmentTermsEnumFrame.

[vsop-479]

I replaced loop compare suffixe with Arrays#compareUnsigned in IDVersionSegmentTermsEnumFrame and OrdsSegmentTermsEnumFrame.

@jpountz It seems you implemented the same thing in SegmentTermsEnumFrame. Please take a look when you get a chance.

[github-actions]

This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the [email protected] list. Thank you for your contribution!

[github-actions]

This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the [email protected] list. Thank you for your contribution!

[vsop-479]

Hello @jpountz , Please take a look when you get a chance.

[github-actions]

This PR has not had activity in the past 2 weeks, labeling it as stale. If the PR is waiting for review, notify the [email protected] list. Thank you for your contribution!