code review has no way to force regenerating Review diff

Bug #383346 reported by John A Meinel
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Launchpad itself
Triaged
Low
Unassigned

Bug Description

There doesn't seem to be a way to tell the code review page that the diff needs to be regenerated.

It is possible to get a small race condition for the patch previews, as evidenced by:
https://code.edge.launchpad.net/~jameinel/bzr/1.16-simple-win32/+merge/6984

My guess of what happened is that:

1) Vincent landed his patch about 'stderr' to http://bazaar-vcs.org/bzr/bzr.dev
2) I branched from that, and made my fix
3) I pushed up my fix, and requested a review
4) Code-review generated the "Review Diff"
5) Launchpad then mirrored http://bazaar-vcs.org/bzr/bzr.dev => lp:bzr

At that point the Review diff ends up including a patch which isn't actually special for my change.
The weirdest part is that the list of revisions being merged is "correct" in that it only shows the single revision that is actually my change. But it certainly has the wrong diff associated with that change.

I don't think there is an easy way to avoid the race condition, so adding a "regenerate the diff" option would be nice. Another option would be to have code that notices that *some* of the revisions proposed for merge have actually landed in the development focus, and have it automatically detect that the diff needs to be regenerated.

Aaron Bentley (abentley)
affects: launchpad → launchpad-code
Changed in launchpad-code:
importance: Undecided → Low
status: New → Triaged
Revision history for this message
Tim Penhey (thumper) wrote : Re: [Bug 383346] Re: code review has no way to force regenerating Review diff

We've been talking about how to do this.

Revision history for this message
John A Meinel (jameinel) wrote :

This isn't completely a duplicate of bug #338002, though it may be close enough.

That bug is more when the branch I've proposed has been updated, it doesn't "automatically" update the review.

This bug was opened more because when the review was generated it had an old version of the target branch. The target branch was then updated, which caused the list of revisions to be merged to be updated, but the *diff* was not updated.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.