1101524 - Shouldn't be necessary to publish new commits to an existing review

Status: RESOLVED INVALID

(MozReview Graveyard :: General, defect, P3)

Description

[James Graham [:jgraham]]

When updating a review by pushing new commits, it shouldn't be necessary to push "publish"; this is easy to forget and disruptive as it requires switching into the browser.

Comment 1

[Steven MacLeod [:smacleod]]

I think this would be best fixed by introducing some sort of flag on push to indicate publish should happen immediately, rather than publishing by default.

Having the draft state is nice because it allows you to check your push, and correct it, without ever making things public. I actually use this to do self review in the diff viewer before requesting other reviewers.

Comment 2

[Gregory Szorc [:gps]]

Mercurial has an internal "is ui interactive" flag. We can prompt to publish if that is enabled. That will add minimal overhead while preventing publishing of accidents.

This is arguably orthogonal to a --publish flag. But I think we should start here.

Comment 3

[Mark Côté [:mcote]]

We have too many P1s, so I'm spreading out the priorities.  P3 -> P4, P2 -> P3, and some portion of P1s will become P2.

Comment 4

[BMO Automation]

MozReview is now obsolete. Please use Phabricator instead. Closing this bug.