[12:09] <gmb> danilos: Are you OCRing today?
[12:10] <danilos> gmb, oh, indeed! :)
[12:11] <gmb> Heh.
[12:11] <gmb> danilos: An easy one to start with, then, if you could: https://code.launchpad.net/~gmb/launchpad/no-mute-for-non-subscribers-bug-741821/+merge/55512
[12:11] <danilos> gmb, and I added a card for me to remember it
[12:11] <danilos> gmb, that's like more than a hundred lines of diff, find someone else!
[12:12] <gmb> :D
[12:14] <danilos> gmb, comment on lines 12-13 confuses me a bit: isMuted() is a subset of isSubscribed() in that isSubscribed() will be true if it's a NOTHING ("mute") subscription, right?
[12:14] <danilos> gmb, or, put more simply, whenever isMuted() is true, isSubscribed() is true as well?
[12:15] <gmb> danilos: You're right.
[12:15] <danilos> ack
[12:15] <gmb> (The second version is more succinct than my comment, so I'll change it)
[12:15] <danilos> gmb, heh, cool
[12:16] <danilos> gmb, also, if you are losing the facet/layer in configure.zcml change (indentation suggests you might :), I suppose that's intentional
[12:18] <gmb> danilos: I'm not losing the facet, am I? The page definition is still within the <facet facet="bugs">...</facet> block.
[12:22] <danilos> gmb, ah, ok, thanks for checking so I don't have to :)
[12:22] <gmb> danilos: Heh. I had to check because my ZCML changes are often done pretty much blind anyway :)
[12:22] <danilos> gmb, in general, I am pretty sure facet is not used anymore for anything, but just making sure if it's happening that you know you are doing it :)
[12:22] <gmb> Right.
[12:24] <danilos> gmb, anyway, looks good; I am assuming that "mute link" becomes "unmute link" when you have it already muted, since that has nothing to do with this branch :)
[12:24] <danilos> gmb, r=me
[12:24] <gmb> Exactly.
[12:24] <gmb> danilos: Thanks.
[13:04] <gary_poster> joy.  gmb, danilos, have either or both of you already investigated wgrant's email and reverting of our branch?
[13:05] <gmb> gary_poster: I saw the email but haven't investigated further yet.
[13:06] <gary_poster> cool, just curious where we were
[13:11] <danilos> gary_poster, I haven't seen it so far
[13:11] <gary_poster> ack danilos thanks
[13:11] <danilos> gary_poster, looking at it now, I am unsure what wgrant means so I'll have to ask for clarification
[13:12] <gary_poster> yeah.  His approach seems unhelpful and unnecessary to me, and his decree of inappropriate approaches seems heavy-handed at best.
[13:12] <gary_poster> approach of roling back a feature-flag protected bit of code, I mean
[13:13] <gary_poster> rolling
[13:17] <bac> gary_poster: it turns out when a bug supervisor for ubuntu tries to subscribe one of his other teams there is a failure.  i don't think it is very noticeable, though:  http://people.canonical.com/~bac/subscription-boom.png
[13:17] <gary_poster> heh
[13:18] <bac> WONTFIX for now?
[13:18] <gary_poster> no, can't see it at all
[13:18] <danilos> gary_poster, we have agreed very early on in our pre-imp that this was not the "right" way to do it, but that it was the only one we could afford to do
[13:18] <gary_poster> danilos, yes, though what you have done was not inherently problematic either.
[13:19] <gary_poster> danilos, I'm replying to your email taking responsibility for the pre-imp.
[13:19] <gary_poster> bac, is it going to be too annoying to fix?
[13:19] <bac> gary_poster: no, it should be easy
[13:20] <bac> sorry, i should've used my <sarcasm> tags.  :)
[13:20] <gary_poster> oh ok :-)
[13:20] <danilos> bac, sorry, they got escaped by our XSS fix :)
[13:21] <danilos> gary_poster, also, I wouldn't worry about the reply, it's not about the blame, it's about the approach
[13:21] <danilos> gary_poster, my main question would be "are we supposed to halt any AJAX feature development until we fix our AJAX infrastructure"
[13:22] <gary_poster> bac, ok...how about make a bug and a backlog card?
[13:22] <gary_poster> we'll get to it if we get to it
[13:22] <bac> gary_poster: i already have a card
[13:22] <bac> it was my next task
[13:22] <gary_poster> bac, well, ok, file the bug, associate it with the card, and move the card to the backlog then.
[13:23] <gary_poster> bac, also I made cards for the issues you brought up before
[13:23] <gary_poster> as I suspect you saw
[13:23] <gary_poster> I'm not sure what you think we ought to do with them
[13:23] <gary_poster> but I din't want them to be lost
[13:27] <bac> bug 745660
[13:27] <_mup_> Bug #745660: Attempt to subscribe unassociated team to distribution fails <Launchpad itself:New> < https://launchpad.net/bugs/745660 >
[13:28] <gary_poster> cool thanks
[13:35] <danilos> gary_poster, call time? :)
[13:35] <gary_poster> oh meh, thank you
[13:35] <gary_poster> was looking at email
[13:35] <gary_poster> bac benji danilos gmb, mumble/kanban now-ish
[14:08] <benji> hi
[14:13] <gary_poster> benji hi :-)
[14:13] <gary_poster> yeah, moving card is good
[14:13] <gary_poster> FW2 I guess
[14:13] <benji> k
[14:13] <gary_poster> but not really super important which one
[14:14] <benji> I guess the first thing to do is to determine the actual merge state (whether the code is in devel or not)
[14:54] <benji> gary_poster: from the logs and your email I gather that the code is actually on stable... so my task is now to address the way html is constructed?
[14:55] <gary_poster> right, benji--or at least devel.  (We said that on the call--bac said that his branch had inadvertently put the code back on devel)
[14:56] <gary_poster> sorry if that was not clear.  Should we have a pre-imp to make sure we are on the same page?
[14:56] <benji> oh, I was under the impression that we thought it might be back but weren't sure
[14:56] <benji> sure, mumble?
[14:56] <gary_poster> sure
[15:38] <benji> danilos: I don't want to undo or step on any changes you've made to HTML generation while reworking the HTML generation for structural subscription JS, is there a way for me to do that?
[15:42] <bac> from wgrant's email, what is the difference in
[15:42] <bac> structured("<some>%(bar)s</literal>", bar='<untrusted>')  (Good)  vs
[15:42] <bac> structured("<some>%s</literal>" % '<untrusted')  (Bad) ?
[15:42] <bac> other than the named interpolation?
[15:43] <gary_poster> bac, I assumed that the first uses the code in "structured" to makes the replacement
[15:43] <gary_poster> the second uses Python
[15:43] <bac> ah, right
[15:57] <gary_poster> benji, fwiw and iirc, danilos said his changes were primarily in the drawing of the first, recipients part of the overlay.  In trunk, we decide when we first draw the overlay whether or not to draw the recipient selection or the recipient display.  Since he is adding the ability to add a subscription on the edit page, he had to make that choice happen later.
[15:57] <gary_poster> So, that's at least the area of conflict
[15:58] <gary_poster> depending on how far along he is, you might be able to merge his branch in and then work from it.
[15:58] <benji> thanks
[15:58] <gary_poster> sure
[16:08] <danilos> benji, sorry, got myself immersed in a JS review
[16:09] <benji> danilos: is there a branch I should work from instead of the trunk or particular areas I should stay away from?
[16:09] <danilos> benji, my branch that reworks those bits is pretty simple: https://pastebin.canonical.com/45453/
[16:10] <danilos> benji, I am pretty sure you can't stay away from this, but you can perhaps construct it in a similar way
[16:10] <benji> danilos: when do you expect your branch to land?
[16:11] <benji> I suppose I could apply that patch and work from there.  I /think/ that would keep things sane when we merge.
[16:11] <benji> when do you expect to merge your branch?  I expect mine to be pretty soon.
[16:12] <gary_poster> if you apply the patch, someone will almost certainly still get conflicts, but they should hopefully be easier to resolve...maybe?
[16:12] <danilos> benji, yeah, that should be fine
[16:12] <benji> k
[16:12] <danilos> benji, it's not a big change at all, so I don't mind redoing it as long as you don't go to the extra effort to make this harder :)
[16:12] <gary_poster> heh
[16:13] <danilos> benji, and I am sure it's only going to be easier with the changes you are doing anyway
[16:13] <benji> It's a good thing today's not Make Things Hard on Your Coworkers day in America.
[16:13] <gary_poster> heh
[16:14] <danilos> benji, btw, you don't need the lib/lp/bugs/templates/bugtarget-subscription-list.pt part of the patch
[16:14] <danilos> benji, that's strictly my branch material
[16:14] <benji> k; I'll just use the JS stuff
[16:15] <danilos> benji, excellent
[16:15] <danilos> gary_poster, I forgot to mention that I was OCR today as well :)
[16:15] <gary_poster> :-) heh, ok danilos, just let your clone handle that bit
[16:32] <danilos> done for the day, haven't been able to finish the reversion branch either (it was supposed to be enough at lp:~danilo/launchpad/revert-xss-workaround, but I hit some weirdness in LP behaviour not generating sprites properly and such), but I've got to run
[16:33] <gary_poster> danilos, ack.  we should take the reversion from you then
[16:33] <gary_poster> can you publish--do you have the time?
[16:33] <gary_poster> oh duh
[16:33] <gary_poster> you just said the branch :-)
[16:33] <gary_poster> thanks ttyl
[18:04] <benji> oh the humanity! the windmill tests fail for me on virgin devel
[18:08] <gary_poster> :-(
[19:34] <bac> gmb: you around?
[19:35] <gmb> bac: Yes-ish. What's up?
[19:35] <bac> gmb: just curious, you reviewed huw's private bug marking branch.  in it he deleted icon-sprites.  any good reason for that?
[19:35] <bac> lp is pretty ugly without them.  :(
[19:36] <gmb> bac: He did? I must've missed that bit.
[19:36] <gmb> I don't recall seeing that when I did the review.
[19:36] <bac> yeah, easy to overlook...until they are gone.
[19:36] <bac> i'll put them back.
[19:37] <gmb> Thanks.
[21:01] <bac> benji: do you have two seconds to review this makefile fix?  https://code.launchpad.net/~bac/launchpad/restore-sprites/+merge/55619
[21:01] <benji> bac: sure (perhaps it will distract me long enough to forget that JavaScript exists)
[21:02] <bac> benji: it is really short
[21:02] <benji> darn
[21:02] <bac> perhaps this review plus a belt of scotch
[21:02] <gary_poster> heh
[21:03] <bac> gary_poster: did you see the thread on trizpug yesterday?
[21:03] <gary_poster> no
[21:04] <bac> josh asked for opinions of django vs pyramid and there were some very thoughtful replies...until cc chimed in
[21:04] <gary_poster> heh
[21:07] <gary_poster> I just read the thread, bac.  I thought the Django one was interestingly...without experience of the other alternatives, but it still had some fair points.
[21:07] <benji> bac: done
[21:08] <bac> thanks benji.  what do you mean "tried to keep it from happening again?"
[21:08] <bac> IMO, the way to do that is to not check in derived files
[21:09] <benji> bac: something like a web page difference reporter for a long time, but with projects like http://www.phantomjs.org/ it might be reasonable to consider nowadays
[21:10] <benji> re. derrived files: absolutely; I was thinking slightly bigger in scope: keeping pages from chaning visually in ways that are otherwise hard to test
[21:10] <bac> benji: ah, ok
[21:10] <bac> your heads in the clouds and mine is in the gutter that are makefile rules
[21:13] <benji> heh
[22:42] <bac> gary_poster: i've finished my branch and pushed a MP to https://code.launchpad.net/~bac/launchpad/bug-745660/+merge/55636 .  if the elves don't review it overnight i'll get a yellow reviewer tomorrow
[22:42]  * gary_poster looks
[22:42] <bac> gary_poster: most of it is drive by cruft from deleting a file
[22:43] <bac> gary_poster: and that really wasn't a passive-aggressive request for a review this late in the afternoon.  :)
[22:44] <gary_poster> :-) great bac.  That said, I will be happy to look at it tomorrow if the elves don't get it, as you say (or if anyone else doesn't beat me to it)
[22:44]  * bac eods
[22:44] <gary_poster> bye
[22:44] <bac> chao