| [07:56] <tomwardill> hmm, there's some doc test fallout from the code import changes, unsure as to correct way to fix | |
| [08:00] <ilasc> yep, saw that | |
| [08:01] <ilasc> and can't seem to reproduce locally | |
| [08:02] <ilasc> ok, got it locally | |
| [08:02] <ilasc> meaning reproduced | |
| [08:02] <wgrant> The diff on that MP is very weird | |
| [08:02] <ilasc> tomwardill: do you want me to look at it or are you ? | |
| [08:02] <tomwardill> `bin/test -vvct lib/lp/code/doc/codeimport.txt` on master will do it | |
| [08:02] <wgrant> <require | |
| [08:02] <wgrant> + permission="launchpad.Edit" | |
| [08:03] <wgrant> + set_attributes="url"/> | |
| [08:03] <wgrant> url has a separate setter method AFAIK which performs extra checks | |
| [08:04] <tomwardill> wgrant: the doctests imply it does, but I can't find any reference to it actuallye xisting | |
| [08:04] <wgrant> Oh, interesting. | |
| [08:05] <tomwardill> there is `updateFromData`, but there's no setters for the attributes | |
| [08:05] <wgrant> Oh, yeah, that's what I mean | |
| [08:05] <wgrant> And what it means. | |
| [08:05] <wgrant> Not as in a property.setter, but a general accessor. | |
| [08:06] <wgrant> The fix here is to remove that extra security declaration and fix any test fallout from that. | |
| [08:24] <tomwardill> poking now | |
| [09:05] <tomwardill> wgrant, ilasc: Doing this seemed a sensible way forward: https://code.launchpad.net/~twom/launchpad/+git/launchpad/+merge/386241 | |
| [09:06] <tomwardill> otherwise updateFromData grows a bunch of 'if' checks for permissions on certain attributes. | |
| [09:23] <tomwardill> thanks ilasc :) wgrant, can you just check we've not missed something easier with the zcml? I get suspicious of xml, it's sneaky. | |
| [09:41] <wgrant> Needs more gaml | |
| [09:41] <wgrant> tomwardill: Looks fine I think | |
| [09:41] <tomwardill> ta, landing | |
| [09:42] <wgrant> Only objection is that updateFromData could use updateURL, but that's hard with how the former is currently implemented | |
| [09:42] <wgrant> So this way around is good | |
| [09:42] <tomwardill> yeah, it's a little too wrapped up in the notification checks to extract easily | |
| [09:48] <tomwardill> I bet me landing the QuestionMessage changes breaks a bunch too | |
| [09:52] <ilasc> 🙂 | |
| [09:52] <ilasc> I guess we'll find out | |
| [10:12] <tomwardill> sigh, yep | |
| [10:15] <wgrant> Easy ones, at least. | |
| [10:18] <tomwardill> yeah, got a branch to fix, just waiting to see if there's any more | |
| [10:28] <tomwardill> https://code.launchpad.net/~twom/launchpad/+git/launchpad/+merge/386248 counting is hard | |
| [10:28] <ilasc> lol | |
| [13:06] <pappacena> Thanks for fixing the doctest, folks. Sorry about that! I didn't see the build bot failure. | |
| [13:09] <tomwardill> no worries :) | |
| [14:23] <tomwardill> hmm, this LXD transcription is going to have some 'interesting' features here... | |
| [14:25] <SpecialK|Canon> Obligatory reminder to please signpost all the known dragons... | |
| [14:25] <tomwardill> it's okay, I've named them all Errol | |
| [14:26] <SpecialK|Canon> I really must read Discworld some day | |
| [14:26] <tomwardill> start at Guard Guards. | |
| [14:26] <tomwardill> (this is a semi-controversial opinion) | |
| [15:43] <tomwardill> pylxd does not make things easy | |
| [15:44] <tomwardill> to the point where actually just shelling out and using bash is a lot easier | |
| [16:08] * tomwardill -> EOD for bicycle club | |
| [16:09] <SpecialK|Canon> That's disappointing :( | |
| [16:09] <SpecialK|Canon> Er, the pylxd thing not the bicycle club! | |
| [16:09] <SpecialK|Canon> Happy cycling! | |