[05:10] <thumper> review needed: https://code.edge.launchpad.net/~thumper/launchpad/recipe-build-email-fix/+merge/32063
[05:10] <thumper> ^^ critical bug
[05:48] <thumper> lifeless: do you think it is worthwhile changing the mail to be a symbolic name?
[05:48] <lifeless> thumper: not a blocker, and I'm sure there are many places that would need touched.
[05:49] <lifeless> thumper: it was just reading it I went 'erk' - its tech debt waiting for use in future.
[05:49] <thumper> lifeless: actually it looks like it could be a config variable
[05:49] <lifeless> ahha! in which case that makes it rather more important to use that
[05:50] <thumper> config.canonical.noreply_from_address
[05:50] <thumper> changed
[05:52] <lifeless> sweet, thanks
[05:56] <thumper> I'll just run through ec2 to make sure there are no surprising breakages
[05:56] <lifeless> naturally ;)
[12:26]  * henninge lunches
[13:18] <salgado> bigjools, LP seems to be having trouble to generate a diff of https://code.edge.launchpad.net/~salgado/launchpad/bug-614923/+merge/32089, but I think it's worth of an RC so here's the diff: http://bazaar.launchpad.net/~salgado/launchpad/bug-614923/revision/9639 :)
[14:36] <bigjools> salgado: approved :)
[14:42] <salgado> thanks bigjools!
[14:47] <salgado> bigjools, should I submit it to devel or db-devel?
[14:51] <henninge> Hi abentley! ;)
[14:51] <abentley> henninge, Hi!
[14:52] <bigjools> salgado: devel is open
[14:53] <salgado> bigjools, I actually submitted it to db-devel before asking.  should I submit it to devel as well?
[14:53] <bigjools> salgado: might be a good idea
[14:54] <salgado> ok, I'll do that
[15:12] <henninge> jtv: the reason for putting translation_side_traits in the interface is so that it can be accessed on security proxied instances?
[15:12] <jtv> henninge: no, it's just for convenience.
[15:13] <jtv> As the cover letter says: potemplate.translation_side_traits is easier to spell than getUtility(ITranslationSideTraitsSet).getForTemplate(potemplate)
[15:14] <henninge> yes, I understand taht.
[15:21] <bac> bigjools:  rc ping
[15:21] <bigjools> bac: rc pong
[15:23] <bac> bigjools:  i have a db branch that i tried to get in on friday but due to some hardware problems i missed by 45 minutes
[15:23] <bac> https://code.edge.launchpad.net/~bac/launchpad/db-bug-613442/+merge/31955
[15:24] <bac> bigjools:  the patch modifies the db but does nothing operational.  it is prep for work scheduled for the next cycle.
[15:24] <bac> bigjools:  so i am requesting an RC
[15:25] <bigjools> bac: what are the risks?
[15:27] <bac> bigjools:  there are no call sites yet using the new properties.  i cannot see any risks.  i will land it through ec2 to ensure, again, that nothing breaks.
[15:28] <bac> bigjools:  there is ONE risk.  i have a self-assigned db-patch number as stub was not available to give me one
[15:28] <bigjools> ah
[15:28] <bac> bigjools:  after talking with jml i just picked the next one , 81
[15:28] <bigjools> did you let him know?
[15:28] <bac> bigjools:  that is a risk that you as the gatekeeper may need to manage.
[15:28] <bac> bigjools:  no i haven't but will right now
[15:29] <bigjools> FWIW I often ask him for numbers before a schema review
[15:30] <bigjools> bac: ok so rc=me.  It's probably the least risky time to pick the next number since there's not many branches going in.
[15:31] <bac> bigjools:  email just sent to stub and lp-dev
[15:31] <bigjools> bac: great, thankds
[15:54] <henninge> jtv: reviewed
[15:54] <jtv> henninge: yayyy!  Er… sneak preview?
[15:54] <henninge> r=me ;)
[15:54] <jtv> yayyy!
[15:54] <jtv> :-)
[15:54] <jtv> Thanks.
[15:55] <henninge> np but we will have to discuss the traits some more another time.
[15:55] <jtv> henninge: OK.  By the way, I'm not saying you did anything wrong w.r.t. the parentheses—it's just that the new lint doesn't like it that way and I try to keep it quiet.
[15:55] <henninge> jtv: all good. ;)
[15:58] <jtv> henninge: you're right that we might as well not expose the traits in the interface… I'll see if I can change that.
[15:58] <henninge> cool
[17:51] <rockstar> abentley, I have a rc candidate branch coming up.
[18:23] <rockstar> abentley, https://code.edge.launchpad.net/~rockstar/launchpad/fix-cancel-rescore-permissions/+merge/32126
[18:28] <abentley> rockstar, it's nice to block access to the pages, but AFAIK, this only prevents people from viewing the pages, not sending POST requests to rescore/cancel builds.
[18:29] <rockstar> abentley, I don't think that's true, because it's a request to the page, and the access SHOULD be on a request level.
[18:29] <rockstar> abentley, I guess I don't know, but I'm assuming it would DTRT.
[18:30] <abentley> rockstar, possibly.  Testing required.  wgrant has demonstrated that the model permissions are wrong, and I think that's the key thing to fix.
[18:30] <abentley> rockstar, but for these pages, shouldn't it be Admin, not Edit?  You can imagine that in the future, we'd want users to be able to cancel their own builds.
[18:31] <rockstar> abentley, I thought about that, but wanted to make a minimal change for rc.
[18:31] <abentley> rockstar, thought about model permissions or Admin vs Edit?
[18:31] <rockstar> abentley, the latter.
[18:32] <abentley> rockstar, why would Admin be a larger change?
[18:32] <rockstar> abentley, it's not a much larger change, but the current permissions are launchpad.Edit, so I'd have to change the links, etc.
[18:33] <rockstar> launchpad.Admin isn't currently defined.  We can make the current launchpad.Edit be launchpad.Admin, but it seems 6 of 1, half a dozen of the other.
[18:33] <abentley> rockstar, that's fine then.
[18:35] <abentley> rockstar, I'm thinking that eventually we will want it defined, and that it's clearer if lp.Edit isn't admin-only, but we can chalk that up as tech debt.
[18:36] <rockstar> abentley, okay.
[18:38] <abentley> r=me
[21:13] <rockstar> abentley, please to have a review: https://code.edge.launchpad.net/~rockstar/launchpad/fix-spr-builds-perms/+merge/32139
[21:14] <abentley> rockstar, r=me.  Thanks.
[21:14] <rockstar> abentley, no problem.  I'm fixing the model issues now.
[21:15] <abentley> rock!
[21:15] <rockstar> (because otherwise we'll expose it in the API one day without even thinking about it)
[21:50] <bac> sinzui:  https://code.launchpad.net/~bac/launchpad/bug-32618/+merge/32145
[21:53] <abentley> rockstar, I can has review?
[21:53] <abentley> rockstar, https://code.edge.launchpad.net/~abentley/launchpad/fix-builder-history/+merge/32150
[21:55] <rockstar> abentley, r=me
[21:55] <abentley> rockstar, thanks.
[23:36] <wgrant> abentley: Code seems to prefer model permissions and ignore view permissions. But then when the model permissions fail, a hole is left. There are a few other places like this in Code as well.
[23:36] <wgrant> Everyone else seems to use correct permissions for the views as well.