[00:00:02] second I'm saying that I think this should be part of the contact page rather than a separate screen [00:00:06] jdlrobson: also, it looks phil commented on the discussion page about the 'factual errors' thing: http://www.mediawiki.org/wiki/Talk:Mobile_design/Contact#Factual_Errors [00:00:07] e.g. a heading saying Article feedback [00:01:07] okay [00:01:34] that's more about the emailing though awjr - my main worry is the redirect [00:01:35] all of the sections lead to a "new page" or a it looks like it, as the title changes to what they selected. [00:02:21] unless we change it now based on what you are saying [00:02:56] let me know something up quickly [00:03:03] *knock [00:03:11] i.e. you choose article feedback you get http://www.mediawiki.org/wiki/File:Mobile_Contact_F2.png you choose technical problem you get http://www.mediawiki.org/wiki/File:Mobile_Contact_I1.png [00:03:21] awjr: let me know when your ready to talk about contact us [00:03:31] tfinc: we're already talking about it :) [00:04:07] spiffy [00:04:19] then grab me after you guys finish hashing things out [00:04:25] i want to get an idea of the scope [00:04:37] so that we understand how much time its going to take and wether its worth to do it now [00:04:54] tfinc: jdlrobson was proposing a significant scoping back for an initial release [00:05:47] rather than release with the emailing functionality, just release with the form writing to a wiki page (like it does now), but re-styled, being newly handled by a SpecialPage, and offering the different categories [00:06:56] awjr: i think the email functionality was pretty important for Philippe [00:07:04] he is the one who dictated most of the requirements here [00:07:42] philinje: for sure - we were just talking about an initial release. we would build the rest of the functionality in later/as we go [00:08:17] awjr has a good point and it fits nicely into what i've been saying. understanding how many iterations it'll take to get to our first release [00:08:20] ok, but how much later is later? [00:09:31] philinje: we'll be able to do an initial release much faster if we scope it back - it may take just as long to do an initial release with the functionality as it would to do a secondary release with it if we do an intial one that is scoped back. plus i think we'll have a better chance of getting it right and mitigating any other weirdness we discover as we build out these changes [00:10:12] if by release we are talking about a beta, and the final release is with the full functionality, that's fine [00:10:22] part of the problem here is that there is still ambiguity and uncertainty about some of the features in the spec [00:10:47] which we're getting hung up on and is slowing down getting this out, but if we can do it more piecemeal, i think we'll get something of value out quicker [00:10:52] in my experience the smaller chunks you do at a time the quicker you get to the end result [00:10:59] and usually the result is better [00:11:04] +1 [00:11:07] as it gets more feedback from earlier exposure [00:11:27] and i know heatherw will have lots of feedback when it goes live ;-) [00:11:48] i still have some from what is up there now :) [00:11:57] again, if we are talking about betas, that's ok - in fact, it might be better for another reason - we don't trigger the email workflow with errors [00:13:05] jdlrobson: was there a question about keeping things on one page, like expanding a form underneath each item? [00:13:35] awjr: I was thinking a little bit about the email handling [00:14:19] yes philinje just knocking something up [00:14:21] awjr: I think it might make sense to write it to a local database as a queue and have a job that pulls the messages and emails them in bulk [00:15:23] preilly makes sense [00:15:32] please email me any other questions as I have to go offline for a while, or let me know if we should have a Skype call tomorrow [00:16:19] i like the idea of keeping things on one page [00:17:02] preilly i think that's a good idea and potentially even necessary depending on the volume [00:17:18] awjr: yeah [00:17:23] but will definitely take more dev time [00:19:06] tfinc now's probably a good time to talk about scope - esp since phil just left im not sure how much more we'll be able to get hashed out [00:19:13] ok [00:19:24] so heatherw awjr this is what i meant - > http://mobile-geo.wmflabs.org/w/index.php?title=Main_Page&mobileaction=leave_feedback#section_1 [00:19:30] irc/voice/other ? [00:19:40] tfinc: semaphore? [00:19:57] LOLSpeak? [00:20:08] irc or voice is fine with me, jdlrobson do you have a preference? [00:20:31] nope [00:20:38] New patchset: preilly; "Remove temporary hack to allow for testing on localhost" [mediawiki/extensions/ZeroRatedMobileAccess] (master) - https://gerrit.wikimedia.org/r/4147 [00:20:49] lets do Skype then [00:21:07] k [00:24:27] jdlrobson: i am not sure i am seeing what i am supposed to see at your link [00:36:54] tfinc: http://en.wikipedia.org/wiki/Wikipedia:Mobile_Extension_Feedback [00:37:07] http://es.wikipedia.org/wiki/Wikipedia:Mobile_Extension_Feedback [00:52:06] * tfinc looks over the contact us page to see what needs to be modified [00:53:14] heads up - i will be heading to bed soon [00:53:18] jdlrobson: k [00:54:15] jdlrobson: wimp [00:54:21] jdlrobson: it's 1:53am Tuesday (BST) - Time in London, UK [00:54:39] I need to be up for 9am to work some more... what's wimpish about that :P [00:55:27] so tfinc awjr i think this is what you need from me -> http://mobile-geo.wmflabs.org/w/index.php?title=Main_Page&mobileaction=leave_feedback&r=313&useformat=mobile - I'll work with heatherw, lindsey and anyone else interested to deal with styling, copy text and error handling (for which I'll need some server side changes from awjr) [00:55:27] jdlrobson: ha ha [00:55:34] jdlrobson: YuviPanda would make it work [00:55:46] YuviPanda: is like a duck he always seems to be awake [00:55:54] TIL that ducks are always awake [00:56:02] hah [00:56:08] (citation needed) [00:56:40] why is mobile-geo sooo slow ? [00:56:52] also, apparently ducks sleep with only half their brains - other half is awake (http://www.straightdope.com/columns/read/1333/can-one-side-of-a-ducks-brain-sleep-while-the-other-half-is-awake) [00:56:58] i donnnn'tt know tfinc - I thought I setup memcached [00:57:04] asked preilly to look at it [00:57:14] preilly: what can we do to make it faster? [00:57:22] tfinc: MAGIC [00:57:29] yep and YuviPanda ducks trust no one - always one eye open [00:57:32] ah [00:57:40] tfinc: I'll look at it again [00:57:41] i know its a virt so i'm not expecting to benchmark anything but its slow as hell as is [00:57:41] ducks have lunch and dinner, I guess? :D [00:57:42] jdlrobson: https://gist.github.com/5cfb8084bf35633a96d1 [00:57:44] is what i've [00:57:55] jdlrobson: it essentially converts sections into a tree [00:58:04] jdlrobson: how will "Please enter an email address." look on a mobile device? [00:58:06] sweet YuviPanda [00:58:11] jdlrobson: my plan now is to use a template to turn this Page object into html [00:58:19] beautiful preilly with unicorns and rainbows [00:58:21] like, an actual js template :) [00:58:41] unicorns! [00:58:55] 'Would you be kind enough to bequeath us with your electronic mailing address, sir?' [00:59:13] jdlrobson: think it'll be usable for MFE at any point? [00:59:14] preilly: i didn't know "This homepage needs to be configured. Read more here" was going out .. nice [00:59:26] when can we have English (RP) as a recognised wikipedia language? [00:59:49] jdlrobson: https://ang.wikipedia.org/wiki/H%C4%93afods%C4%ABde [01:00:07] too old, perhaps :) [01:01:04] so YuviPanda the main use case I have for that code at the moment is my html5 history experiments [01:01:08] trying to create a mobile page for http://ms.wikipedia.org/wiki/Laman_Utama is no fun [01:01:23] jdlrobson: ah [01:02:14] eek tfinc what happened to the create page link that was there before :-( [01:02:33] jdlrobson: what do you mean ? [01:02:49] well when I go to that page I see the configure homepage message [01:03:34] jdlrobson: which makes sense as ms wiki doesn't have a configured mobile main page [01:03:43] but that should only show on the home page [01:04:00] or is that the homepage? [01:04:04] * jdlrobson confused [01:04:08] that is the homepage [01:04:33] ok phew i thought i was losing it:-) [01:04:39] jdlrobson: go to bed [01:04:43] lol [01:05:01] Laman Utama === Home Page [01:05:16] in Malay [01:06:14] yep I've got that now preilly. my brain was just a bit slow to catch up and I misread tfinc's 'is no fun' and YuviPanda's 'ah' [01:06:28] got it [01:06:36] also, I got scared a bit too [01:06:37] see i didn't even get the sarcasm there [01:07:30] ha ha heh heh [01:07:38] GO TO BED <- jdlrobson [01:08:17] okk... [01:08:20] SLEEP(100) [01:08:25] that's usually the line for me [01:12:24] heatherw: ok … lets chat [01:23:02] I thought that jdlrobson was a javascript guy so I wonder why he didn't use setTimeout [01:23:41] preilly: setTimeout doesn't block [01:51:03] New patchset: preilly; "Stylize file and remove unused code" [mediawiki/extensions/ZeroRatedMobileAccess] (master) - https://gerrit.wikimedia.org/r/4148 [01:51:14] http://ms.m.wikipedia.org/ exists now [01:51:36] * tfinc debates about cleaning up the styling [01:51:45] eh .. we can do it later [02:05:25] New review: Demon; "Needs rebasing anyway to make this merge, so might as well fix mf-feedback.js while we're at it." [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: -2; - https://gerrit.wikimedia.org/r/3963 [02:07:21] New review: Demon; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/4133 [02:07:24] Change merged: Demon; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4133 [02:09:05] okay, a nice small chunk of API work done [02:09:08] good night everyone [02:11:02] New review: Demon; "One minor bit of trailing w/s still. Is also going to need rebasing once the parent is fixed." [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: -1; - https://gerrit.wikimedia.org/r/3964 [02:32:31] open /Applications/Xcode.app/Contents/Developer/Platforms/iPhoneSimulator.platform/Developer/Applications/iPhone\ Simulator.app/ [03:04:47] can someone test the speed of http://mobile-geo.wmflabs.org/w/index.php/Special:Random [11:20:18] New patchset: Jdlrobson; "separate select menu into sections, using mailto links for article feedback" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/4156 [12:01:38] New patchset: Jdlrobson; "removed minified javascript" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4158 [13:12:14] [WikipediaMobile] jdlrobson pushed 4 new commits to master: http://git.io/byp05g [13:12:14] [WikipediaMobile/master] Added migration code for PhoneGap database issues - YuviPanda [13:12:14] [WikipediaMobile/master] Updated CHANGELOG and bumped version number - YuviPanda [13:12:14] [WikipediaMobile/master] Added comments clarifiying data migration - YuviPanda [13:12:37] Project WikipediaMobile - Nightly builds build #268: SUCCESS in 9.9 sec: https://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/268/ [13:12:38] * yuvipanda: Added migration code for PhoneGap database issues [13:12:38] * yuvipanda: Updated CHANGELOG and bumped version number [13:12:39] * yuvipanda: Added comments clarifiying data migration [13:51:13] jdlrobson: heya [13:54:44] hey YuviPanda [13:54:47] just commenting on 192 [13:56:19] done [13:57:02] so YuviPanda https://github.com/wikimedia/WikipediaMobile/pull/193 looks fine apart from one comment and the fact I just trust you on java related stuff [13:57:27] jdlrobson: yes, will clarify that [13:57:34] also what is the state of https://github.com/wikimedia/WikipediaMobile/pull/173 [13:57:45] i'm wary it's quite an old pull request [13:57:52] jdlrobson: I did make a comment above (fixing languages to confirm to wikipedia's language codes) [13:57:53] and there has been no feedback [13:58:02] jdlrobson: hmm, let me poke kelson on it and see if he wants to take a look [13:58:20] i'm weary of pulling support for qt in and letting it rot [13:58:22] ahh yes I see [13:58:32] is it possible we can link to an article that explains wikipedia language codes [13:58:34] like how we have winphone bits but they don't work [13:58:40] jdlrobson: yeah, will clarify [13:58:45] great [13:59:04] brb just got to knock up some lunch [13:59:13] k [14:12:57] back btw [14:24:25] hey guys, any API feedback? [14:35:25] MaxSem: is 'prop' disregarded with section=all [14:35:27] or am I missing something [14:38:02] prop=sections means whether to return information about not requested sections [14:38:46] with sections=all you're requesting information about all sections [14:38:50] MaxSem: hmm, essentially prop=normalizedtitle doesn't seem to do anything [14:39:43] it is returned only if you provided a non-canonical title [14:40:24] ah! [14:40:24] i thought it handled redirects, etc too [14:40:44] MaxSem: also, mobileview doesn't seem to do anything for Main_Page [14:41:14] should it? [14:41:56] now there's two of you [14:42:31] i'm real [14:42:36] the other one is the network ghost [14:42:52] it is returned only if you provided a non-canonical title [14:42:53] [20:09:59] ah! [14:42:53] [20:10:15] i thought it handled redirects, etc too [14:42:54] [20:10:35] MaxSem: also, mobileview doesn't seem to do anything for Main_Page [14:43:03] [18:41:16] should it? [14:43:13] good question. [15:08:23] [18:40:26] i thought it handled redirects, etc too [15:08:43] MaxSem: that was in reference to 'normalizedtitle' [15:09:15] it actually handles redirects now, unless you've provided redirect=no [15:09:39] MaxSem: hmm, so if it was a redirect, i'd get a normalizedtitle back? [15:09:55] MaxSem: wouldn't it be consistent to always send back a title if prop includes normalizedtitle [15:09:56] ? [15:10:00] you'll get "redirectedfrom" [15:10:52] I get "redirected": "Chennai", [15:10:53] I thought it would be a waste of bandwidth. a small waste but stil.. [15:11:23] right, "redirected" [15:11:42] MaxSem: hmm, in what cases would normalizedtitle actually turn up? [15:11:46] can you give me an example? [15:12:27] api.php?action=mobileview&format=json&page=foo [15:12:28] vs. [15:12:31] api.php?action=mobileview&format=json&page=Foo [15:13:02] also Unicode normalisation, etc [15:13:18] aah [15:13:49] damn, I'm even more determined that we should deploy new APIs on labs first [15:20:16] New patchset: Jdlrobson; "increase size of wikipedia logo in footer" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4171 [15:20:17] New patchset: Jdlrobson; "increase font size of links at bottom of footer and color of more/less" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4172 [15:20:38] MaxSem: I'm reworking the code a bit to make pointing to random domains easier. [15:20:49] sehr gut [15:24:31] [WikipediaMobile] jdlrobson pushed 2 new commits to master: http://git.io/n5GRSw [15:24:31] [WikipediaMobile/master] Added clarification on language fix - YuviPanda [15:24:31] [WikipediaMobile/master] Merge pull request #193 from yuvipanda/bug35650 - Jon Robson [15:24:44] Project WikipediaMobile - Nightly builds build #269: SUCCESS in 6.7 sec: https://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/269/ [15:24:44] * yuvipanda: Make default language for search be the system's default locale [15:24:45] * yuvipanda: Added clarification on language fix [15:41:12] jdlrobson: mobilefrontend.init does a bunch of stuff I don't seem to need [15:41:23] * jdlrobson looks [15:41:26] like the desktop view and language stuff [15:41:37] i'm thinking i should just use individual toggles? [15:42:18] https://github.com/wikimedia/WikipediaMobile/blob/master/assets/www/js/mobilefrontend.js [15:42:31] you're not using the default MobileFrontend code you're using that stub ^ [15:43:14] jdlrobson: aaah, right. I changed it to default in a commit :D [15:43:16] will change back [15:44:29] jdlrobson: thoughts on making *all* sections expandable? [15:44:32] rather than just h2s? [15:44:59] it looks ugly [15:45:09] check my mail about the use of history api [15:45:42] (subject HTML history interface / api use for MobileFrontend) [15:46:07] i think the useful line was "It enables sub section expansion for which the existing design doesn't really seem to work" [15:48:39] jdlrobson: hmm [15:49:56] do you see what i mean? [15:51:11] jdlrobson: yes, it was quite ugly on my phone [15:51:18] so i like the idea [15:51:25] but i don't think it's worth it right now [15:51:39] jdlrobson: 'threading' on mobile phones itself is quite something [15:51:39] to be fair most sections tend to be small enough anyway [15:53:11] YuviPanda: this bug is for the app right not MobileFrontend ? https://bugzilla.wikimedia.org/show_bug.cgi?id=35616 [15:53:31] jdlrobson: it's actually for both [15:53:48] i find the discussion very confusing [15:53:56] but as phil said, mobile web allows a way to switch to desktop view and see history [15:53:58] while the app doesn't [15:54:29] true [15:54:34] ok [15:54:42] so it's not so critical [15:55:23] jdlrobson: depends on whom you ask :) [15:56:14] sure ok i understand this now anyway [16:03:52] jdlrobson: also, #search in app.css conflicts with #search in beta_common.css [16:03:57] should i just change order [16:04:02] or do we want to do something else? [16:04:24] hmm, even changing order might not fix this [16:04:36] it has properties that aren't redefined in app.css, so we will have issues [16:04:44] jdlrobson: how do we deal with such conflicts? [16:04:48] make the css more specific? [16:23:36] jdlrobson: ^ [16:35:29] hmm, looks like i need to do nesting myself anywy [16:35:53] New review: Reedy; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/4158 [16:35:55] Change merged: Reedy; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4158 [16:36:58] dinner [16:37:01] or something resembling it [16:37:02] brb [16:54:56] YuviPanda: "This version has been approved and is being held by the developer" :D [16:55:03] 3.1 is apple approved [16:56:18] SWEET [16:56:46] i'll release it once i get into the office [16:57:06] k, commuting in [17:13:49] flyingclimber: preilly are you guys about? [17:14:50] and/or MaxSem and jdlrobson? [17:14:56] * MaxSem  [17:15:19] wassup? [17:16:05] Just CC'd you on a bug, see emails [17:16:30] on it [17:19:27] Cheers [17:22:47] tf:) [17:22:51] tfdamn [17:23:38] jdlrobson: thoughts on CS overrides? [17:24:31] you mean CSS !important ? [17:24:51] jdlrobson: well, more like conflicts between app's UI elements and beta ui elements [17:24:55] jdlrobson: in this case, #search [17:25:18] jdlrobson: beta_common.css has a left property for it, while app.css doesn't. [17:25:30] jdlrobson: other properties cancel out (overriden by app.css), but left isn't [17:25:36] jdlrobson: ideally, i'd want them to have different IDs [17:25:42] or the CSS rules to be more specific [17:25:46] so this problem doesn't happen [17:25:58] can't you do left: auto ? [17:26:03] * MaxSem runs blame [17:26:10] jdlrobson: I *can*, but I think I shouldn't. [17:26:15] can't we make the selectors more specific? [17:26:21] .mf-container #search [17:26:25] for example? [17:26:32] #mf-container #search even [17:26:33] well ideally we should have the same css file for both app and MFE :) [17:26:35] s/even/rather/ [17:26:39] same look and feel etc [17:26:45] jdlrobson: ah [17:27:03] I agree that the web beta looks sweeter than the app :) [17:27:12] but i guess we are some time off that [17:27:19] (we need new nav at least I guess) [17:28:05] I think we discussed separating the css [17:28:12] Reedy, will you be able to sync it once I commit a fix into Git? [17:28:26] Sure [17:28:31] so section related css goes into a separate css file and search into another [17:28:51] MaxSem: you could attach a patch to the bug, I can fix the site and then we can commit it etc.. [17:28:57] jdlrobson: right. [17:29:04] jdlrobson: when do you think that 'separation' can happen? [17:29:11] as soon as we have ResourceLoader [17:29:29] ah [17:29:37] annd… how soon is that? :D [17:29:46] https://bugzilla.wikimedia.org/show_bug.cgi?id=34325 [17:30:00] is it beta_ only that needs this separation or both? [17:30:12] jdlrobson: i'm currently using beta_common [17:30:17] k [17:30:42] i'm hoping I can get ResourceLoader for css in by end of week, then hopefully things can move quickly (you may need to poke me again to remind me) [17:31:00] jdlrobson: so, right now I simply hack around whatever conflicts I find? [17:31:09] * YuviPanda doesn't like this :( [17:31:13] ergg yeh that sucks [17:32:01] send me a patch of the changes you'd like and I'll see what i can do [17:32:08] is it just search? [17:32:25] jdlrobson: currently just search [17:32:27] (you can assume MFE will have a class mfe on body [17:32:43] jdlrobson: so [17:32:55] #search to .mfe #search should do the trick [17:32:57] for me [17:33:00] yeh that sort of thing [17:33:03] jdlrobson: do you want a full patch? [17:33:03] ok [17:33:20] if you send me a patch with all the bad definitions i'll make it happen tomorrow [17:33:30] I've got to run now but will leave my laptop on and come back later for anything urgent [17:33:35] jdlrobson: cool [17:35:18] greetings all [17:35:20] YuviPanda: you around? [17:35:24] tfinc: yes [17:35:26] saw the approval [17:35:28] \o/ [17:35:30] :D [17:35:37] just make sure that people *know* they will lose saved pages [17:35:56] i'm adding this to the release notes: "NOTE: This is a brand new codebase and will delete your old history and bookmarks. If you do not want this to happen please do not upgrade. We apologize for any inconvenience." [17:36:15] good enough for me [17:37:00] YuviPanda: k. i'm going to look over the copy and release this [17:37:07] \o/ [17:37:12] YuviPanda: lets schedule the blog post for the afternoon [17:37:19] coool [17:37:20] after we get a review or two :D [17:37:47] granted i should check the communications calendar before i saw were going to test today [17:37:55] also while it's on my mind - could we make MobileFrontend have special urls for the beta - it makes it easier to test e.g. http://en.m.wikipedia.org/wiki/beta/San_Francisco or even http://en.m.wikipedia.org/wiki/San_Francisco?usebeta=yes [17:38:01] comment: I believe there's UCOSP standups today, though I am not certain if it's in 20 minutes or 3:20 minutes. [17:38:42] oh wow .. the blog is busy today [17:38:43] tfinc: yes [17:38:44] YuviPanda: https://office.wikimedia.org/wiki/Communications/Calendar [17:38:56] s/today/this week/ [17:39:09] but its far too light on tech [17:39:13] it's busy this week [17:39:13] yeah [17:39:15] we can fix this [17:39:17] i see 0 tech [17:39:30] :) [17:42:19] YuviPanda: go ahead and it to the calendar [17:42:54] today is the third. [17:42:55] right [17:43:19] * YuviPanda looks up the title [17:43:27] Reedy, patch attached [17:43:57] YuviPanda: http://etherpad.wikimedia.org/Wikipedia-iOS [17:44:03] lets update the description [17:44:48] i could just use the same copy as android [17:45:34] i like how the iOS app has a line for developers [17:46:29] tfinc: updated calendar [17:46:32] thanks [17:47:50] i hate that we can't change the name [17:48:18] YuviPanda: what do you think about the dev line in the description ? [17:48:35] the github line? [17:48:41] that whole line in general [17:48:44] we don't have it here https://github.com/wikimedia/WikipediaMobile [17:49:51] for the love of ... "Good but Great app but it needs some improvements like: 1- An option to save articles to read them offline." .. do people just not read the features ?!? [17:50:35] :D [17:50:41] people don't read, news at 11 [17:51:02] people still watch news @ 11 ? [17:51:20] not sure, it's been a few years since I've watched any TV [17:51:30] i haven't watched broadcast tv in years [17:51:36] MaxSem: In Soviet Russia the news watches you at 11 [17:51:50] it's amazing how people are able to put up with Ads [17:51:54] and crappy ones at that [17:52:28] YuviPanda, my brother sells "Turn on your brain, throw your TV away" T-shirts [17:52:40] I bought one from him [17:53:13] :D [17:53:18] it assumes you have a TV [17:53:48] anyone who has a computer can watch IPTV [17:54:55] YuviPanda: android app listing has been updated with dev line [17:55:06] woohoo [17:55:13] it'll take a couple of minutes for the market to pick it up [17:55:21] MaxSem: is "Bug 35673 - XSS vulnerability in mobile search" fixed now? [17:55:43] plugged on live site, going to push a git fix [17:55:44] YuviPanda: any last changes to the text for iOS ? [17:55:52] i'm going to send it to amit as well [17:55:54] tfinc: checking one last time [17:57:26] okay, so just switching to Sanitizer::encodeAttribute is the fix? [17:57:31] MaxSem Reedy [17:57:34] ^^ [17:58:01] preilly, you have to take into acoount what will you use escaped text for [17:58:26] New patchset: MaxSem; "Revert and redo r114191 aka 3f791ab2 - introduces XSS" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4191 [17:59:09] MaxSem, Reedy: Is Sanitizer::encodeAttribute multibyte character safe? [17:59:25] escapeJsString() is for escaping data to be included in JS, it's not suitable for HTML [17:59:29] should be [17:59:58] Reedy, can you unhide the bug now? [18:00:10] ya [18:00:21] tfinc: a few tweaks [18:00:36] i like them [18:00:47] Amit will look at within the hour and then we'll release it [18:04:14] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/4191 [18:04:16] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4191 [18:04:22] tfinc: did you say you were writing email notes about the Android Beta W (for tweaks) or did i misunderstand? [18:04:43] i did and am… will send it today [18:04:52] awesome, thanks! [18:05:15] MaxSem: so, if fixes are for bugzilla bugs doesn't it make sense to name the local branch and topic after the bug number instead of "topic 2012/search-xss-fix" for example? [18:05:36] forgot about that [18:07:09] MaxSem: no worries [18:20:55] YuviPanda: you made it into the signpost http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2012-04-02/Technology_report :) [18:21:55] woohoo [18:26:13] jdlrobson: how do you think we should handle things that are wider than our viewport? [18:26:17] images, templates, etc? [18:27:02] throw them away! [18:27:16] and never listen to me [18:27:36] heatherw: find out who put them there and send them a sternly worded email? [18:27:39] (or talk page message) [18:27:40] ? [18:27:44] [18:27:48] YuviPanda: thats a question i ask all the front end dev candidates :) [18:27:52] or at least most [18:27:55] :D [18:27:59] yes, spend the rest of your days doing that [18:28:07] its a fun problem [18:28:15] currently we handle it by not handling it [18:28:17] as we have lots of amazing content that rolls right of the screen [18:28:23] yeah [18:28:26] if it is wider than your phone's screen, you don't need to look at it [18:28:29] get a better phone, duh [18:28:53] YuviPanda: amits edits are in [18:28:59] * YuviPanda looks [18:29:32] tfinc: 'fork us' is inaccurate [18:29:51] they can't fork us, they can only fork the repo [18:29:59] (unless we're all git repos living in the matrix) [18:30:01] YuviPanda: so fix it [18:30:15] (in which case I demand access to revert) [18:30:16] tfinc: done [18:30:24] Amit was editing that section and took out the word 'fork' [18:30:26] i added it back [18:30:33] and explained why it was necessary [18:31:45] WOW [18:31:47] "The following is not recommended for use in this field: Google. Your app may be rejected if you use this term." [18:32:12] now i want to see what else the'll reject [18:34:50] :D [18:34:55] woah [18:35:35] tfinc: they have Google apps in there by default, and they will block apps with that word on it? [18:35:35] o_O [18:35:44] tfinc: I hope Apple eventually dies in a submarine full of electric eels [18:35:51] its beyond bizarre to me [18:36:36] what did I miss/ [18:36:40] ? [18:37:12] nothing in the 3 seconds you were gone [18:37:13] :) [18:37:16] ah [18:37:17] :D [18:37:34] unless someone has a blocker i'm going to release 3.1 to the app store [18:37:38] Jarry1250, hi. I haven't seen you application in Melange - you should post it ASAP [18:37:55] Oh, crikey, thanks for reminding me MaxSem [18:37:55] \o/ [18:38:05] I'll do that now. [18:38:07] * tfinc waits for anyone to raise a blocker [18:38:25] philinje: you mentioned lingering blocker… did you follow up with YuviPanda ? [18:38:29] blockers* [18:39:26] ? [18:40:26] YuviPand_: Have you seen the latest comments on iOS RC1? [18:41:47] philinje: ? [18:42:05] haven't [18:42:06] looking [18:42:43] philinje: I see nothing new [18:43:00] and nothing blocking that we're not tackling in 1.2 [18:43:06] hmm, let me file a few more bugs [18:43:10] ok [18:43:15] most are enhancements [18:57:40] ok .. it doesn't sound like we have blockers then. i'm going to release the binary [18:57:47] \o/ [18:57:53] s/binary/version :D [18:58:21] to the tune of 'release the kraken'? [18:58:25] yes! [18:58:49] it has been released! [18:58:54] woot [18:59:06] this is a pretty big step in unifying our code bases [18:59:08] nice job yuvi [18:59:10] :D [19:00:40] as soon as its up lets tweet [19:01:33] * YuviPanda opens itunes [19:03:30] jdlrobson: Jon, just wondering if something happened to the CSS of the first paragraph of Featured Article, possibly in general [19:04:04] this is a persistent bug that has been fixed before [19:07:03] MaxSem, awjr, jdlrobson: what do you guys think of this change: http://pastebin.mozilla.org/1552392 [19:07:45] whee [19:08:20] I propose to call it SkinMobile [19:08:36] BatMobile [19:13:41] MaxSem: okay, how about: http://pastebin.mozilla.org/1552402 [19:14:40] awjr: yo [19:14:50] tfinc yo [19:15:06] how's it going? [19:15:17] http://upload.wikimedia.org/wikipedia/commons/thumb/0/07/Us_yoyo_national_1a.jpg/220px-Us_yoyo_national_1a.jpg [19:15:28] awjr: i'm debating wether we should mail mobile-tech about mobile into core or go directly to wikitech [19:15:31] preilly, skinname and stylename seem to be initialized earlier in core skins [19:15:47] tfinc: wikitech-l i think is the most sensible [19:16:14] are you happy enough with it to go out to wikitech ? [19:16:34] tfinc one sec lemme refresh my memory [19:16:45] MaxSem: okay [19:17:00] tfinc: do you want my opinion? [19:17:06] nah [19:17:07] :D [19:17:08] sure [19:17:39] tfinc: depends - should we add some reqs/methodology/test plan/etc, or should we just start looking for feedback and shape the rest as we go? [19:17:48] * preilly — adding tfinc to auto ignore list [19:18:22] i think we should have a solid goal and rough plan before we engage in a broader discussion [19:18:48] tfinc i think it's safe to say we have that [19:19:17] World domination. [19:19:36] ok … i'll mail out post food [19:19:39] time to grab lunch [19:20:56] jdlrobson: Jon, are you there? [19:21:24] New patchset: preilly; "Remove dependency on output buffering and replace with rudimentary skin" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4198 [19:21:46] MaxSem, awjr, jdlrobson ^^ [19:22:08] * MaxSem looks [19:22:16] wowzers [19:22:23] preilly is there much difference in performance? [19:24:48] awjr: I'm measuring it now [19:25:25] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) C: -1; - https://gerrit.wikimedia.org/r/4198 [19:27:10] there should be performance improvements once we manipulate DOM only for content, i.e. make beta and zero features string-based [19:27:28] New patchset: preilly; "Remove dependency on output buffering and replace with rudimentary skin" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4198 [19:28:58] MaxSem: please review this again ^^ [19:34:28] MaxSem: you there? [19:34:55] * ajerith waits for Canonical to build his debs. [19:35:17] ajerith: gets excited about debs if they're for me [19:35:48] YuviPanda: 3.1 is live in the store [19:35:55] \o/ [19:35:59] * YuviPanda heads to itunes [19:36:05] preilly, I've looked and tested, but would like awjr's opinion too [19:36:19] MaxSem, preilly: i'll take a look in a few minutes [19:36:21] preilly: They're for you. [19:36:31] awjr: review https://gerrit.wikimedia.org/r/4198 [19:36:39] And you can start being excited once we know whether they'll actually work. [19:36:42] ajerith: cool — that's awesome [19:37:03] ajerith: okay, scaling back excitement level to 2 [19:37:18] None of the people I know who know how to build debs out of Python apps are available, although one of them gave me a script he used. [19:37:37] So I've been trying to figure it out on my own. [19:37:41] ajerith: okay, cool — I hope that it works [19:37:53] Turns out I need to package four deps as well as vumi. [19:37:54] ajerith: I've got faith in you [19:37:58] YuviPanda: http://itunes.apple.com/us/app/wikipedia-mobile/id324715238?ls=1&mt=8 [19:38:06] And presumably vumi-wikipedia as well. [19:38:06] tfinc_: wh00t [19:38:08] yes updating link in blog post [19:38:11] :D [19:38:13] ajerith: yeah, that too [19:38:13] preilly: That makes one of us. :-P [19:38:25] ajerith: :) [19:38:35] preilly: https://launchpad.net/~jerith/+archive/vumitest-daa090d24139/+builds?build_state=pending [19:38:49] * preilly *clicking* [19:38:54] The crappy PPA name is because you can't get rid of a package version once it's built. [19:39:15] tfinc: what should I sign as? [19:39:26] So once I know how to do this thing, I'll blow the PPA away and start again with a fresh one. [19:39:55] YuviPanda: the blog post? [19:39:58] yeah [19:40:14] same thing as http://wikimediafoundation.org/wiki/Staff_and_contractors [19:40:18] ah [19:40:19] Software Developer - Mobile (contractor) [19:40:20] right [19:40:38] ok [19:40:38] done [19:41:08] * tfinc wonders if we should crowd source patricks photo through draw something [19:41:16] * tfinc bets jorm would like that [19:41:48] * preilly thinks hmm [19:43:23] i've a few of them... [19:43:24] :D [19:44:48] YuviPanda: now that its up go ahead and tweet about it [19:45:01] yup [19:45:28] after you publish the blog post send a mail to mobile-l@ referencing to catch anyone who might not be on the blog [19:45:43] so [19:45:46] before my change: [19:45:49] Overall Summary [19:45:50] Total Incl. Wall Time (microsec): 171,250 microsecs [19:45:51] Total Incl. CPU (microsecs): 140,000 microsecs [19:45:53] Total Incl. MemUse (bytes): 13,865,632 bytes [19:45:54] Total Incl. PeakMemUse (bytes): 13,906,848 bytes [19:45:54] Number of Function Calls: 43,993 [19:45:58] and after my change: [19:46:08] Overall Summary [19:46:09] Total Incl. Wall Time (microsec): 121,621 microsecs [19:46:10] Total Incl. CPU (microsecs): 100,000 microsecs [19:46:11] Total Incl. MemUse (bytes): 14,539,856 bytes [19:46:12] Total Incl. PeakMemUse (bytes): 14,556,832 bytes [19:46:13] Number of Function Calls: 20,980 [19:47:10] MaxSem, awjr, jdlrobson, tfinc ^^ [19:47:50] preilly that is great [19:47:55] looking at changes now [19:48:08] w00t [19:48:46] why this change resulted in such serious speedup? [19:49:08] 43,993 function calls vs 20,980 function calls. [19:49:45] yeah, but why? [19:53:02] 521 � � ob_start( array( $this, 'DOMParse' ) ); <- line removed [19:53:27] no, that's not all of it. [19:54:31] this function is called in another place [19:55:08] yah, but that seems to be a substantive removal of a lot of work. [19:57:02] MaxSem: we can also rework our getMsg method now too [19:57:31] by total elimination! [19:57:34] New review: awjrichards; "Overall this looks good - there is, however, trailing whitespace in MobileFrontend.body.php. Also, I..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/4198 [19:57:37] MaxSem: since we are no longer in the, "Output buffering handler" we can now use wfMsg and the like [19:58:07] preilly this should open the possibility of using resource loader too, right? [19:59:24] no, RL is restricted by mobile devices' capability [19:59:31] :( [19:59:45] if we really wanted, we could have hacked it even with ob_start() [20:00:00] we can use RL for css now [20:00:10] without the need for vector hacks [20:04:39] New patchset: preilly; "Remove dependency on output buffering and replace with rudimentary skin" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4198 [20:05:11] awjr, MaxSem ^^ [20:05:23] * awjr takes a look [20:06:17] New patchset: preilly; "Remove dependency on output buffering and replace with rudimentary skin" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4198 [20:06:29] awjr: bad Deprecated: Call-time pass-by-reference has been deprecated in /var/www/w/extensions/MobileFrontend/MobileFrontend.body.php on line 134 [20:06:39] what I don't like about gated trunk is that reviewers can nitpick the committer to detah about things that used to be ok'd in SVN times [20:07:36] MaxSem: yeah [20:07:41] also, look now: [20:07:44] Overall Summary [20:07:44] Total Incl. Wall Time (microsec): 80,969 microsecs [20:07:46] Total Incl. CPU (microsecs): 70,000 microsecs [20:07:47] Total Incl. MemUse (bytes): 10,527,384 bytes [20:07:48] Total Incl. PeakMemUse (bytes): 10,562,488 bytes [20:07:49] Number of Function Calls: 20,981 [20:07:59] !!! [20:08:01] awesome [20:09:02] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/4198 [20:09:04] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4198 [20:09:14] booyakasha [20:09:25] awjr: sweet! [20:09:37] preilly that is a great chaneg [20:09:38] MaxSem, awjr: thanks, for reviewing this [20:09:39] *change [20:10:00] awjr: we should fix up getMsg method now too [20:10:06] preilly np [20:10:10] since we are no longer in the, "Output buffering handler" we can now use wfMsg and the like [20:10:20] preilly yeah that should also reduce overall times [20:10:26] yeap [20:10:47] im amazed switching to using $this in the skin dropped the times as much as they did too [20:11:07] awjr: yeah [20:11:14] but like i said, there would otherwise have been a bunch of duplicate work going on [20:16:08] okay, what's next? I can remove the vestiges of cached messages from MobileFromatter [20:19:20] MaxSem: yes [20:20:24] awjr, useformat=mobile doesn't seem to stick for me [20:21:55] MaxSem: do you have the right JS loaded? [20:25:59] preilly, last week, once I clicked on mobile view it continued to work until I clicked on desktop view [20:26:07] now, it doesn't stick [20:26:25] MaxSem: is the cookie being set correctly? [20:27:37] preilly, mf_useformat:mobile [20:28:15] MaxSem: that cookie is set? [20:28:21] yes [20:28:58] hey YuviPanda [20:28:58] MaxSem: it might have to do with shouldDisplayMobileView [20:29:44] YuviPanda I fixed the scroll problem when you save a page... when you have time could you test it on your device? thanks :) [20:30:12] MaxSem: I think we need to call $this->checkUseFormatCookie(); before it on line 130 [20:31:06] MaxSem: can you try putting, "$this->checkUseFormatCookie();" right before, "if ( !$this->shouldDisplayMobileView() ) {…" on line 130 [20:31:18] MaxSem: and see if that fixes your issue [20:31:25] okay [20:32:33] MaxSem, awjr: also, on another note once we fix the getMsg method we can totally remove the, "public static $messageKeys = array(… " array as it won't be needed [20:36:14] Astagi: heya [20:36:15] will do :) [20:36:16] saw it [20:36:58] thanks YuviPanda :) [20:37:22] tfinc: ping [20:37:35] YuviPanda: can't right now .. talking with phil [20:37:38] okay [20:37:40] YuviPanda: he is talking-to phil ATM [20:37:50] * preilly — damn, too slow [20:37:50] preilly: ok [20:37:58] it's just to see if your device is an exception, but I think that my phone is the exception :P [20:38:21] Astagi: our app just got released on iTunes, so checking tha tup [20:38:23] will cehck in a moment [20:38:48] great :) [20:39:04] ok np YuviPanda :) [20:41:58] MaxSem: did that fix your issue? [20:43:39] YuviPanda: i'm seeing bad tweets about the update [20:43:48] yes, that's what i'm pinging you about. [20:43:53] seems to have issues with iOS 5.1 [20:44:00] whats the issue ? [20:44:47] preilly, I'm in the middle of committing a different change, will investigate after it [20:44:51] tfinc: most of the UI seems broken [20:45:05] tfinc: i suggest pulling the app down until we fix this :| [20:45:19] hmm [20:45:31] or maybe give me a 2 hour timebox. [20:45:33] YuviPanda: is this only in 5.1? [20:45:35] yes [20:45:41] can we exclude it for 5.1 devices? [20:46:20] preilly: our testing showed it works on 5.0, and i just tested on my simulator (works). We had intermittent crashing reports from 5.1, but not 'none of the UI works' [20:48:09] YuviPanda: i have no control over which devices get it [20:48:14] nor can i roll it back [20:48:16] the store sucks [20:48:22] compared to the android market [20:48:25] we have to fix this [20:48:33] yes, on it [20:49:08] apparently i need to download a full new version of XCode before I can even debug this :| [20:50:10] New patchset: awjrichards; "Set 'contact us' title in the special page, add corresponding i18n message" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3964 [20:50:23] YuviPanda: yes, and you'll have to do it through the mac app store [20:50:55] gah [20:50:59] 'not enough disk space to install this' [20:51:52] YuviPanda: who else can help you with this ? [20:52:13] New patchset: awjrichards; "Updating feedback form id to 'mf-feedback-from' for clarity" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3963 [20:52:30] i'm guessing brion, assuming he already has the 5.1 bits in place [20:52:48] brion: any chance we could get your help on this? its urgent :( [20:53:08] i cleared up space, downloading [20:53:15] am thankfully on someone's much faster internet [20:53:16] what can i do? [20:53:37] brion: seriously what can't you do — let's be real :) [20:53:41] :D [20:53:46] teehee [20:53:48] brion: the app is crashing and burning on iOS 5.1 [20:53:51] YuviPanda: you have brions attention .. shoot [20:53:54] looks like a fatal js error to me [20:53:58] ok i can check that after my meeting with erik [20:54:02] Hrm. brion can do anything? [20:54:11] ajerith: indeed [20:54:13] brion: i think just opening up the app and looking at the log should tell us what's wrong [20:54:24] * ajerith hands his package-a-twisted-plugin-for-ubuntu problem over. :-P [20:54:51] ajerith: he doesn't even need a mobile device he just can use an abacus [20:55:14] brion: do you know when that would be? [20:55:39] YuviPanda, prolly ~30mins, maybe less maybe more [20:55:44] but i'll look for 5 mins first ;) [20:55:49] brion: okay :) [20:55:55] i'm downloading the new SDK [20:55:55] where does it die? [20:56:07] brion: well, essentially most of the UI doesn't work. [20:56:10] actions don't do anything [20:56:13] plus the spinner stays there [20:56:26] ok so to check: did y'all run 'make remotes' before building? [20:56:29] brion: i'm thinking there's a js error happening someplace on startup [20:56:35] brion: this is from the app store version [20:56:41] brion: plus toggling works, so there was make rmeotes done [20:56:52] in simulator, no errors visible [20:56:54] what can I do to help you help yourself? [20:57:12] brion: 5.1? [20:57:19] brion: does the spinner keep spinning? [20:57:22] yeah [20:57:25] no [20:57:29] drphil: :| [20:57:31] err [20:57:33] brion: :| [20:57:35] leemme try on ipad3 [20:57:46] brion: navigate to a page, see if the back button is now enabled? [20:57:58] Jarry1250_: hello. [20:58:25] ok i can confirm the breaky behavior w/ app store version on ipad3. lemme see if i can get it to show logs [20:58:55] YuviPanda: i'm about to step into two hours of meetings. keep me udpated [20:58:59] drphil: tfinc ok [20:58:59] updated even [20:59:03] brion: ok [20:59:07] err [20:59:12] i don't know why i keep including drphil :) [20:59:18] ROFL [20:59:46] YuviPanda: Hello :) [20:59:51] Jarry1250: heya! [20:59:58] our iOS wikipedia app just got released [21:00:02] and looks like it is broken :( [21:00:14] Jarry1250: is it possible for you to download from the app store and verify brokenness/non-brokenness? [21:00:28] YuviPanda: http://pastebin.com/3E0CKhie not very detailed in there. i'll check more after meeting [21:00:36] ok [21:00:39] Could probably manage that at some point in the next hour, yes. What qualifies as "broken" in this context? [21:00:45] brokenness verified but not sure why :) [21:01:01] New patchset: MaxSem; "Remove support for cached messages now that ob_start() is gone" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4202 [21:02:51] YuviPanda: we found a way to remove it from the store but it'll take upwards of 24hrs to do [21:03:00] tfinc: sigh. [21:03:10] Jarry1250: there is *no* way to not push it out to just 5.1? [21:03:35] err [21:03:36] tfinc: ^ [21:03:47] i don't not see an option to do that [21:04:59] but were looking [21:07:44] apple is also capping downloads at 160 KB/s [21:07:45] sucks [21:07:51] New review: awjrichards; "1) The change in MobileFrontend.body.php should actually be made in SpecialMobileFrontend.php. The f..." [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: -2; - https://gerrit.wikimedia.org/r/4156 [21:10:23] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: -1; - https://gerrit.wikimedia.org/r/4146 [21:13:04] hmm, might have identified the issue [21:13:06] (potentially) [21:16:10] YuviPanda: what is it? [21:16:30] YuviPanda: Just downloading now (at less than 160KB/s) [21:17:02] preilly: apparently, iOS 5.1 changed path of where databases and caches are stored, and that is causing problems. [21:17:14] preilly: haven't verified yet, am waiting for it to finish downloading before i can [21:18:24] Jarry1250: what version of iOS are you on? [21:19:02] Yuvi: only 5.0.1 -- so I guess it's not going to be broken for me? [21:19:52] Jarry1250: shouldn't be [21:20:19] Indeed, it seems non-broken. [21:20:39] I'd offer to update to 5.1 but it would take about 6.734 hours to download (approximately) [21:26:51] YuviPanda: so at various times i see these in log: [21:26:53] Apr 3 14:26:23 unknown sandboxd[5811] : Wikipedia-iOS(5807) deny file-write-create /private/var/mobile/Applications/E7472FB7-C941-45DD-BAB8-C9E58570EEF6 [21:28:02] i think the database stuff may be failing in some way [21:28:14] brion_: possibly https://issues.apache.org/jira/browse/CB-330 [21:28:49] though that looks like another (but unrelated) bug [21:28:57] (and just as bad) [21:29:03] hrm [21:29:22] yeah that looks separate; in that case it should "work" but not necessarily preserve data in some cases [21:29:40] yes [21:30:30] so i can search, go to pages, etc, but the spinner doesn't stop. 'save' doesn't work. 'saved pages' list never comes up [21:30:50] yes [21:30:58] brion_: no log messages? [21:31:15] just the ones i sent [21:31:31] New patchset: MaxSem; "Unbreak sticky cookies after change 4198" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4209 [21:31:42] preilly, ^^ [21:31:48] ipad and iphone 5.1 simulator doesn't have the problem. [21:32:13] sigh. [21:32:18] looks like printf debugging again [21:32:23] :( [21:32:26] * YuviPanda waits for it to install [21:34:27] If everything worked perfectly, coding would be a boring job. [21:34:30] awjr: New patchset: MaxSem; "Unbreak sticky cookies after change 4198" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/4209 [21:34:59] If it were easy, we wouldn't get paid the big bucks to do it. [21:35:16] You get paid big bucks? [21:35:34] oh interesting [21:35:44] YuviPanda: i deleted the app and reinstalled it -- now it works [21:35:47] * preilly — thinks what are big bucks? [21:35:53] Amgine, compared to burger flippers;) [21:35:55] brion_: wah. [21:36:34] maybe something in the upgrade from the old app didn't set up the data dirs right [21:36:36] but…. no idea how [21:37:14] About 2.5" by 5" or thereabouts. [21:37:35] tfinc: the app is off the app store [21:37:53] * ajerith measures some currency. [21:38:20] tfinc: do we know if the problems are only on upgrades, or also on new installs? [21:40:00] YuviPanda: good .. i'm glad that moved fast [21:41:00] brion_: its happening on upgrades at the very least [21:41:50] I can verify that this issue does not exist on the simulator. [21:41:56] ugh, upgrade seems to have wiped out my profiles [21:43:05] YuviPanda: does iOS block search on the main page loading ? [21:43:06] well in the simulator you won't be upgrading from the old app, so that might be why it's not happening there -- if it doesn't happen on clean installs, it won't happen there [21:43:10] no [21:43:33] brion_: well, now that the app isn't on the store, we can't actually say. [21:43:43] poop [21:43:44] brion_: can you install into your device 'clean' and see if that works? [21:44:10] YuviPanda: i did -- deleted the app and reinstalled it from App Store [21:44:23] (in purchased tab since it wasn't showing up in searches anymore) [21:44:36] brion_: and it works properly? [21:45:02] yep [21:45:08] hmmm [21:45:16] brion_: so there are *at least* two issues [21:46:06] one being the upgrade issue and the other being the persistence issue [21:46:57] http://leuksman.com/misc/ipad3-freshinstall.png <- saved pages working ok after fresh reinstall [21:47:40] New review: awjrichards; "Why should getUseFormat() always check cookies? It is intended to be a getter for $this->useFormat -..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/4209 [21:55:26] I need to redo my certificates, etc? [21:55:26] :| [21:57:01] YuviPanda: Certificate Request Approved [21:59:57] preilly: thanks. [22:00:34] preilly: http://id.wikipedia.org/w/index.php?title=Templat:HU/BoxContent&action=history&uselang=en [22:01:04] New review: MaxSem; "Because checkUseFormatCookie() already calls setUseFormat() to change current state of this variable..." [mediawiki/extensions/MobileFrontend] (master) C: 0; - https://gerrit.wikimedia.org/r/4209 [22:03:45] kul: http://id.wikipedia.org/w/index.php?title=Templat%3AHU%2FBoxContent&diff=5375371&oldid=5088670 [22:04:49] preilly: i think my certificate needs to be added to the provisioning profile. I don't see it there [22:08:35] tfinc: ^ [22:18:39] YuviPanda: k [22:18:46] so what do you need? [22:18:59] tfinc: https://developer.apple.com/ios/my/provision/view.action?provDisplayId=27DX88WDH4 [22:19:02] doesn't show my certificate [22:19:14] You do not have access to the page you requested. [22:19:18] hmm [22:19:20] you need to log in [22:19:28] i am logged in [22:19:31] to https://developer.apple.com/ios/ [22:19:38] and to to 'provisioning' [22:19:43] and click on the provisioning certificate [22:19:45] thats where i was [22:19:52] hmm [22:19:56] i don't see my certificate there [22:20:06] that needs to be added [22:20:18] its called iOS Team Provisioning Profile: * [22:20:23] and your on the list [22:20:24] we setup a new certificate just now... [22:20:38] my device is on the list, but my certificate isn't. [22:20:45] so XCode refuses to deploy [22:20:46] hmm [22:21:00] that cert is managed by Xcode [22:21:03] i don't control it [22:21:17] hmm [22:21:24] i could make a dew dev profile [22:21:48] dev* [22:22:18] I HATE that apple doesn't let you delete and/or archive app id's [22:22:28] we have four that are really similar (my bad) that i can't get rid of [22:22:51] "You are not permitted to view this team's certificates. If you need additional access, please consult one of your Team Admins." [22:23:03] if I follow http://stackoverflow.com/questions/999313/iphone-app-signing-a-valid-signing-identity-matching-this-profile-could-not-be [22:23:09] this seems to be a not-just-me problem [22:23:12] ok .. i just created a new profile called 'Dev' [22:23:17] with your cert and everyones device [22:23:19] can you see it? [22:23:52] tfinc: yup, that works [22:24:06] woot! [22:26:03] tfinc: okay, this works perfectly [22:26:04] no bugs [22:26:13] :| [22:30:25] YuviPanda: glad to help [22:30:42] especially with stupid bureaucracy [22:31:02] tfinc: this bug seems to manifest only when upgrading from the previous version [22:31:12] am building the old app to see if I can repro it [22:31:17] clean installing works as expected [22:32:43] k [22:32:53] we can ask apple to expedite the fix when its ready [22:33:17] ok [23:05:49] New review: Jdlrobson; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) C: 1; - https://gerrit.wikimedia.org/r/3963 [23:07:34] Change abandoned: Jdlrobson; "No longer needed - select is removed in I8b17ebc916e5a22f73c4a9526617f826388dc6bb" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/4146 [23:16:36] YuviPanda: hows is our fix looking ? [23:16:49] tfinc: so, when I simulate an upgrade, it straight out crashes [23:17:13] tfinc: am following the logs, and getting some help at #phonegap as well. [23:17:15] do we need to tell people to unintall and install ? [23:17:21] apparently this is a known issue that bit a lot of people. [23:17:26] tfinc: that's a solution. [23:17:30] if people uninstall and reinstall it will work [23:17:42] so, if you want, we can put the app back up and tell people 'if it crashes, reinstall' [23:43:07] tfinc: he doesn't really have any actionable ideas that could help us [23:43:08] tfinc: I'm going to ping another person [23:43:39] jdlrobson you around? i was trying to amend https://gerrit.wikimedia.org/r/4156 but wound up fixing the issues in another commit. if you're amenable to the changes, we should abandon 4156 [23:44:15] New review: Reedy; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3965 [23:45:49] * preilly — reminds awjr that it's 12:45am Wednesday (BST) - in London, UK [23:46:10] preilly: heh thanks, i know - he had just made some commit like 20 minutes ago [23:46:28] awjr: I was just joking ;) [23:46:39] :p [23:47:21] awjr: also, I was just talking to him in a pm [23:47:32] awjr: which somehow made that comment funnier to me [23:47:42] heh [23:55:55] tfinc: from conversation on #phonegap, we're not the only ones with stability + other issues on iOS 5.1 [23:56:08] still trying to repro the 'UI weridness' [23:56:12] k