Page MenuHomePhabricator

[Regression 1.36.0-wmf.36] On enwiki, new edits are being marked as quality instead of "accepted" after manual pending changes review
Closed, ResolvedPublic

Description

All new manual pending changes reviews are resulting in them being marked as quality. For example here. The first revision to have been marked as quality appears to have been this.

Special:ProblemChanges no longer exists.

Event Timeline

DannyS712 triaged this task as Unbreak Now! priority.EditedMar 31 2021, 12:05 AM
DannyS712 subscribed.

The "Special:ProblemChanges no longer exists." part is expected and intentional, since the special page was removed, but the fact that new edits are being marked as quality instead of "accepted" is an issue

I've tagged .37 blockers, but this is really an issue with .36 that was rolled out to enwiki today

Aklapper renamed this task from Quality revisions are enabled on enwiki to On enwiki, new edits are being marked as quality instead of "accepted" after manual pending changes review.Mar 31 2021, 6:55 AM

I don't think it's appropriate to pretend this is a blocker to a train.

Ladsgroup lowered the priority of this task from Unbreak Now! to High.Mar 31 2021, 3:56 PM

I'm not sure if that's a big issue. most people can't even distinguish between quality and checked. quality tier will also go away soon. I'll take a look but I don't think this is a UBN nor a train blocker (wmf.37 has not introduced any errors)

Also it was rolled out yesterday.

Jdforrester-WMF renamed this task from On enwiki, new edits are being marked as quality instead of "accepted" after manual pending changes review to [Regression 1.36.0-wmf.36] On enwiki, new edits are being marked as quality instead of "accepted" after manual pending changes review.Apr 5 2021, 1:04 AM

enwiki local msg hack at MediaWiki:Revreview-hist-quality-user can be removed when this is solved

@Ladsgroup with the above patch merged is this issue solved?

Ladsgroup claimed this task.

Incomplete fix: With the default system message gone, the revisions marked as "quality" during the regression are now displayed as follows:

Screenshot_2021-04-30 Wikipedia Sandbox T279276 Revision history - Wikipedia(2).png (36×3 px, 29 KB)

Screenshot_2021-04-30 Wikipedia Sandbox T279276 Difference between revisions - Wikipedia(4).png (552×1 px, 64 KB)

Second screenshot is with safemode=1, just to make sure it's not an issue in my scripts.

I guess the review status of these revisions needs to be corrected in the database.