When you have a lot of projects it's not at all obvious which repo
a given revision belongs to.
Details
Details
- Reviewers
epriestley chad - Group Reviewers
Blessed Reviewers
Viewed phabricator.lan/differential, saw the repo listed under each revision.
Screenshot:
Diff Detail
Diff Detail
- Repository
- rP Phabricator
- Branch
- master
- Lint
Lint Passed - Unit
Tests Passed - Build Status
Buildable 18526 Build 24951: Run Core Tests Build 24950: arc lint + arc unit
Event Timeline
Comment Actions
Sorry, not interested in bringing this upstream at this time. The code is fine if you want to patch it locally. See T4863 / T418 and linked tasks for some general discussion, I think. I think the current state of the world here is that this comes up every so often, but not meaningfully more than all the other "always show X on the list of Y" requests, and if we say "sure" to all of those requests the UI will be an unusuable mess of stuff that's only useful to 3% of users. T4863 / T418 talk about possibly supporting customization to deal with these cases.
Comment Actions
@epriestley I have no problem patching it locally (already have) it just seemed like an obvious thing to show on the differential screen. If it's not wanted upstream then it's no sweat off my back :)
Comment Actions
I'm very much interested in this change too — when you have a bunch of diffs applying to different repos it's really really really useful to see it at a glance which repos they belong to.
So, question: would you be willing to consider a version of this diff that makes this behavior optional, defaulting to false (i.e., current behavior) ?
Comment Actions
Well, the premise "It is usually much easier to get a change upstream if it does not add an option than if it does." doesn't seem true in this case, as you're opposed to accept the change without the optional behavior.
And, for what is worth, I totally agree with the general argument against options, but here you're opposing a proposed change that is really useful for people triaging diffs on a bunch of unrelated repositories.
I suspect that is why D9609 was accepted, but that solves the problem only for people using the email UI. The problem currently persists for people using the Web UI.