[09:34] <blr> wgrant: that seems to be working better with the expanders (less code too), although I seem to have introduced a validation regression - will get to that tomorrow.
[09:34] <wgrant> blr: Great.
[09:34] <wgrant> Let me know if you run into any trouble.
[09:34] <blr> thanks - I'll be in town for an hour and a bit in the morning to pickup a monitor, but back before lunch.
[09:35] <blr> looking forward to no longer hunching over my laptop :P
[09:35] <wgrant> You've been single-monitor all this time?
[09:35] <wgrant> My condolences.
[09:36] <cjwatson> I'm not *totally* sure that my dual-monitor setup is exactly ergonomic, with the RH one being above and to the right of my laptop
[09:36] <cjwatson> But it's better than being single-monitor ...
[09:36] <blr> made sure the panel is vesa mountable - will be good to have two eventually.
[09:38] <cjwatson> My T450s is on the right continent now, so hopefully running 1920x1080 + 1600x1200 won't be too hideously confusing
[09:45] <wgrant> Yay
[10:25] <cjwatson> wgrant: You're right (git-merged-revno), it's much simpler to put merged_revision in the model, thanks.
[10:26] <wgrant> cjwatson: I'd call it something displayish if it were expected, but merged_revision is fine internally.
[10:26] <wgrant> ...
[10:26] <wgrant> s/expected/exported/
[10:27] <cjwatson> Yeah, I didn't export it
[10:28] <wgrant> Bah
[10:28] <wgrant> Goddammit, branchrevision
[10:28] <cjwatson> ?
[10:28] <wgrant> You can't have been that cold.
[10:28] <wgrant> Just a failed scan despite a few pushes during the day.
[10:29] <wgrant> cjwatson: Have you heard anything about git support in mojo?
[10:29] <cjwatson> Nope
[10:30] <wgrant> Tempted to switch something like turnip git-ward now that things are generally functional.
[10:31] <cjwatson> But mojo is definitely an issue there.
[10:31] <cjwatson> I guess we could run an import.
[10:31] <wgrant> Yeah, as long as we don't use gpgsig the import will be fine.
[10:31] <wgrant> We should dogfood at least *something* ASAP now it's reasonable.
[10:32] <wgrant> And the lack of CI on turnip makes it an easy target :)
[10:32] <wgrant> Also it'
[10:32] <cjwatson> I suspect the firewall will need a tweak
[10:32] <wgrant> s a git server hosted on bzr.
[10:32] <cjwatson> For self-imports
[10:32] <wgrant> Not just a firewall tweak, but yeah.
[10:32] <wgrant> Anything under .launchpad.net is blacklisted.
[10:33] <cjwatson> I wanted to finish merge detection before we started dogfooding.
[10:33] <cjwatson> Otherwise it'll just be annoying
[10:33] <wgrant> True.
[10:33] <wgrant> I was thinking that distro MPs didn't do it, so it couldn't be too annoying.
[10:33] <wgrant> But it's distro *branches* that don't.
[10:35] <cjwatson> Yeah, there are XXX comments about that
[10:35] <wgrant> (also I assumed something based on UDD being reasonable, which is itself unreasonable)
[10:35] <cjwatson> Much easier to have sensible logic without branch lifecycles.
[10:36] <wgrant> Quite.
[10:36] <wgrant> Deletable branches are cool.
[11:06] <cjwatson> wgrant: One reason I included target_default on the form for now is that otherwise if you retarget a repository to personal and back then the target defaultness goes away and you don't know why and have no UI way to restore it.
[11:07] <cjwatson> But I agree that the permission limit there is problematic.
[11:08] <cjwatson> Perhaps it should simply not be possible to retarget a target default.
[11:15] <wgrant> cjwatson: The lack of UI way to restore it will go away once the product config page is reworked.
[11:16] <cjwatson> Yeah.  Retargeting a target default repository is a slightly weird thing to do anyway, though, and has lots of intersecting permissions.
[11:17] <wgrant> Yep.
[11:17] <wgrant> It's inconsistent with the owner-target default checkbox.
[11:17] <wgrant> But the permissions for that are much less intractable.
[11:17] <cjwatson> Exactly.
[11:17] <cjwatson> That's just edit plus checking that there's no conflict.
[11:18] <wgrant> Yep
[11:19] <cjwatson> I'll work on something to make the target uneditable with an explanatory message in this case.
[16:57] <cjwatson> wgrant: What's the modern equivalent of the onKeyPress thing you called out in https://code.launchpad.net/~cjwatson/launchpad/git-repository-ui-edit-target/+merge/261232 ?
[16:57] <cjwatson> I can't find anything in LP that's using anything different
[20:51] <blr> cjwatson: I'm not certain, but perhaps he was suggesting adding an event handler to the JS rather than using the widget event?
[21:58] <cjwatson> blr: I'm still at the cargo-culting level with LP JS :-/
[21:58] <cjwatson> Or JS in general really
[21:59] <cjwatson> Are you thinking of something to do with YUI.Event?
[22:00] <cjwatson> Oh, there are some .on('keypress', blah) calls in our JS
[22:00] <cjwatson> OK, that I can work with, looks straightforward enough
[22:01] <blr> cjwatson: you might want to clarify with wgrant, but at a guess that's what he meant.
[22:02] <cjwatson> Yeah, it makes sense now that I've found it
[22:02] <blr> cool :)
[22:02] <cjwatson> Ideally I'd figure out how to make the JS live with the widget, mind
[22:03] <cjwatson> But I guess it's already separate