[00:00:51] yeah, figured it out [00:01:00] what about the new content API? [00:02:50] philinje: http://www.mediawiki.org/wiki/Wikimedia_Apps/status#2012-05-02 [00:03:02] the beta *just* missed the report by 1 day [00:03:21] it'll be there next month [00:14:13] is getting tired of smashing http://www.mediawiki.org/wiki/Wikimedia_engineering_report/2012/April to get the UCOSP post to line up [00:14:18] Amgine: did you throw in any special formatting ? [00:14:33] i'm running out of ideas on whats breaking it [00:27:40] preilly: MaxSem : can you guys mark which further tasks are done https://www.mediawiki.org/wiki/Mobile_support_in_MediaWiki_core [00:27:41] thanks [00:29:44] is "Offer device-specific skins" what we currently have with HTML/WML separation, or it means finer-grained separation? [00:30:25] finer grained [00:30:30] tablet, mobile, desktop, etc [00:32:45] New patchset: awjrichards; "Abstracted HTML out of 'mobile-frontend-terms-use' i18n message" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6462 [00:33:41] New review: awjrichards; "Addresses bug 36109 (https://bugzilla.wikimedia.org/show_bug.cgi?id=36109)" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6462 [00:40:24] New review: MaxSem; "The main problem with external wikilinks is that they're marked with class="external nofollow", thus..." [mediawiki/extensions/MobileFrontend] (master) C: -1; - https://gerrit.wikimedia.org/r/6462 [00:40:42] awjr, -1 due to overescaping [00:41:25] oh right, i forgot to take out the htmlspecialchars when i switched to trying wikitext [00:41:31] stupid whitspace. [00:42:09] see my full comment though [00:42:27] yeah, i had a hunch that might get -1'd [00:43:01] i feel like there's really just not an elegant way to handle external links with i18n messages [00:43:37] my initial approach was to wrap the anchor text in $1 and $2 and assigning $1 to the and $2 to but it's ugly. [00:44:15] MaxSem what do you think would be the right approach? [00:44:28] New review: MaxSem; "Since we have plans of merging the whole thing into core, what's the point of maintaining 1.19 compa..." [mediawiki/extensions/MobileFrontend] (master) C: 0; - https://gerrit.wikimedia.org/r/6453 [00:45:12] awjr, do something with Parser and message handling [00:45:48] the problem is widespread: quite a lot of extensions have to use HTMl mesages due to messy external link styles [01:01:49] Ryan_Lane: climbing ? [01:02:13] I'm really tired. I think I need to skip this week [01:02:21] I've been running every morning for the past week [01:04:09] hmm .. with our recent app pushes were running out of geonames queries fast [01:04:23] i bought 5million of them a month ago and were already 88% of the way through it [01:07:03] wow [01:11:46] tfinc: wtf are geonames queries? [01:11:58] give me the geolocation for this location? [01:12:07] gps coordinate -> article mappings [01:12:24] "corner of your mom's house and my libido" [01:12:42] ah [01:12:47] we pay for that? [01:13:02] we will until this project finishes https://www.mediawiki.org/wiki/GeoData_Storage_%26_API [01:13:36] if i find the time i'll likely negotiate a deal with geonames [01:13:47] but thats not useful to us long term [01:13:55] ah [01:13:56] heh [01:15:55] tfinc: no [01:16:06] [01:16:24] holy crap. [01:24:34] Where is the ': Wiktionary Mobile App' text coming from? [01:24:59] Amgine: http://www.mediawiki.org/w/index.php?title=UCOSP_Spring_2012&action=edit [01:29:22] [01:32:55] !blame guillom [01:33:13] I'll get back to this after the hockey game is over. [01:33:22] Amgine: who's playing? [02:50:45] tfinc: Washington Capitals vs New York Rangers, tied 1-1. Because it's playoffs they *have* to play to a win. [02:50:56] So, second overtime period. [02:51:52] nice [03:01:54] tfinc: could this because the dates are completely wack? [03:02:22] Amgine: tinker and see. the -monthly pull it into the report [03:02:28] but feel free to experiment [03:03:32] kk. 5th period (of 3) is apparently starting. And there's a cold drink there. But I'll be back to tinker. [03:14:45] ok .. its time for me to leave the office [03:34:30] damn. *still* tied. [04:20:40] and NYC manages to get a goal. [04:30:17] tfinc: fixed. [09:58:06] New patchset: Jdlrobson; "clean up section margin css and remove negative margins" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6476 [09:58:06] New patchset: Jdlrobson; "fix search on android 2.2" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6477 [10:07:55] New patchset: Jdlrobson; "remove border-radius references" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6478 [10:07:56] New patchset: Jdlrobson; "make sure clear search button is not visible outside full screen search" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6479 [11:36:54] New patchset: Jdlrobson; "remove cols and size attributes from old contact us form (bug 36390)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6490 [13:34:55] Hmm, no Jon [13:35:03] * yuvipanda pokes MaxSem about ImagePage refactor [13:50:58] [WikipediaMobile] yuvipanda pushed 1 new commit to master: http://git.io/gWMoYw [13:50:58] [WikipediaMobile/master] Remove stray global assignment (was used for debugging) - YuviPanda [14:14:32] * yuvipanda pokes jdlrobson  [14:14:37] yuvipanda: ! [14:14:42] just moving into app mode :) [14:14:44] ah [14:14:49] sweet :) [14:15:06] jdlrobson: Two pull requests, one of which seems to be a bug in MF [14:15:14] (unless some people turn up in irc - https://twitter.com/#!/WikimediaMobile/status/198051036131368961) [14:15:28] yeah, just saw that [14:15:29] RTL ? [14:15:38] jdlrobson: no, the reference reveal [14:15:50] jdlrobson: references.init essentially ignored 'firstRun' [14:16:02] ohh yeh [14:16:08] good catch I'll fix that now [14:16:12] btw rtl might be a problem [14:16:18] mmmhmm? [14:16:22] as ResourceLoader takes care of that [14:16:35] oh, how? [14:16:41] ResourceLoader magic [14:16:46] it flips values [14:16:52] e.g. left -> right padding-left -> padding-right [14:17:03] the only thing I didn't figure out yet was how the RTLness of the section expanders were handled [14:17:06] wah [14:17:08] damn [14:17:12] that is indeed quite deep magic :| [14:17:15] ResourceLoader magic :) [14:17:22] so for the app I'm not sure how to handle that [14:17:30] we can either maintain our own css which is a pain [14:18:27] or we can run our css through RL and have two copies? [14:18:32] yep [14:18:37] but then we have to choose the correct one [14:18:43] jdlrobson: the fixes in that pull req mostly don't have to do with CSS though. They're just me not setting dir properly [14:18:45] e.g. on loading app add stylesheets dynamically [14:19:01] sure.. it just reminded me that there will be a problem [14:19:11] Right. [14:19:51] jdlrobson: also, expansion troubles - it's slower in the app since we do DOM injection on touch as well [14:20:09] so it seems to first expand and then the DOM fills out [14:20:10] yeh I think we should turn off animations there [14:20:16] or rather it expands as the DOM fills out [14:20:27] which makes things odd [14:20:28] +1 [14:20:32] * jdlrobson gets css for yuvipanda  [14:20:53] actually I'll fix it in core [14:21:01] add noanimations to body tag [14:21:19] cool [14:21:33] would that mean no animations for references either? [14:21:36] noAnimations to be exact [14:21:47] I can turn those off too if you want.. [14:21:56] or we could have the class noTransitions which would turn off just sections [14:21:57] jdlrobson: also, reference reveal still has a bug of sorts in Android 2.x that I'm unsure what to make of. [14:22:18] which bug is that? [14:22:23] jdlrobson: touchstart seems to fire only after a bit when you're scrolling, so it the reference scrolls up a bit before disappearing [14:22:34] ahh that one. mm [14:22:46] it's either that, or they implement scrollig by scrolling a large-ish image than by actually scrolling [14:22:59] can you try something for me? [14:23:00] I think it is actually the latter. [14:23:01] sure [14:26:05] 1s yuvipanda [14:26:36] ok http://pastebin.com/UxC7MBkV [14:26:39] try that references code [14:26:50] and when you call references.init [14:27:05] * yuvipanda looks [14:27:12] pass as a 3rd parameter { animation: 'none', animationSpeed: 0 } [14:27:23] * MaxSem 's back [14:28:57] ok [14:29:05] * yuvipanda goes to obtain his 2.x [14:29:06] device [14:30:44] \me been scratching his head with sitename-footer message which is buggy in en.wikiprojects alone [14:31:11] jdlrobson: is the beta footer using any special graphics for to display as "WikipediA" in footer ? [14:32:11] yes it's an image at the moment.. :-/ > https://bugzilla.wikimedia.org/show_bug.cgi?id=35994 [14:32:46] argh! [14:32:58] any ideas on how to fix that? [14:33:25] jdlrobson: i thought Mobile-frontend-footer-sitename is used on footer [14:33:43] jdlrobson: on non english languages it works fine the message is {{SITENAME}} [14:34:16] jdlrobson: that works fine :) [14:36:06] srikanthlogic: it does use {{SITENAME}} [14:36:11] but icon by default is wikipedia [14:36:18] so alt text will be correct but not logo [14:36:32] yuvipanda: great - I'll submit that then [14:36:41] jdlrobson: should we be just not using any logo and keep it text ? [14:36:43] do you want to refactor your pull requests based on that code? [14:36:50] you can also drop animationSpeed from options [14:37:04] e.g. { animation: 'none' } [14:38:26] srikanthlogic: we do for non English wikipedia pages - the design team felt the watermark was necessary though [14:38:50] as text it looks a bit ugly [14:38:54] jdlrobson: but it fails for non wikipedia projeccts [14:39:14] srikanthlogic: it's the same problem as the logo in the top left [14:39:31] I think we just need to allow people to update the logo in their LocalSettings.php [14:39:49] at the moment you can override stylesheets/images/logo-copyright-en.png [14:41:42] hmm, or there must be option to have text / image and give image details if needed [14:42:15] jdlrobson: top left logo is fine, but footer TM is buggy [14:43:22] you mean its buggy customisations aside? in what way? [14:45:02] jdlrobson: I'll refactor that pull req when this goes into master [14:45:08] k [14:45:10] jdlrobson: also noToggleTransitions? [14:45:15] as a class on html? [14:45:16] yeh one sec [14:45:22] we'll try something soon [14:45:29] I'll send you a patch and you can try it out [14:45:50] sure [14:46:08] i'll be off for a while now (need to get on a bus to get on another bus) [14:46:12] but will be online from the bus [14:47:21] k yuvipanda [14:47:23] you still there? [14:47:24] http://pastebin.com/ASZzWQFM [14:47:33] apply that patch to sections.css [14:47:37] ok [14:47:40] and see if that makes them less glitchy :) [14:55:03] will report back in a bit jdlrobson [14:55:04] * yuvipanda goes off [14:55:07] thanks yuvipanda [14:59:47] New patchset: Jdlrobson; "pay attention to firstRun value passed in by user" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6523 [14:59:48] New patchset: Jdlrobson; "allow initialisation of references with options" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6524 [14:59:49] New patchset: Jdlrobson; "allow turning off transitions" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6525 [15:35:22] [WikipediaMobile] jdlrobson pushed 5 new commits to master: http://git.io/wcabQQ [15:35:22] [WikipediaMobile/master] Separate RTLness for UI language and content language - YuviPanda [15:35:22] [WikipediaMobile/master] Sort languages in settings in Unicode order - YuviPanda [15:35:22] [WikipediaMobile/master] Sort langlinks by unicode order & set proper direction - YuviPanda [16:14:43] jdlrobson: hi [16:15:15] hey philinje [16:15:37] how's it going? [16:15:57] just wondering if you had any questions before the meeting [16:20:13] nope [16:23:52] jdlrobson: prototype looks good [16:30:55] jdlrobson: starting [16:56:41] hey jdlrobson [16:56:50] do you know when the firstRun fix will land in master? [16:57:03] when preilly or someone else merges it [16:57:07] so hopefully soon [16:57:07] right [17:01:32] jdlrobson: adding that diff to sections.css definitely looks smoother on my Nexus S. No longer have access to the 2.x device though [17:01:52] jdlrobson: though can we make the class name a bit more descriptive? noTransitions sounds too broad.. [17:01:55] . [17:05:40] mornin [17:06:01] or good evening, in some of your cases [17:06:06] hello awjr [17:06:17] hows it going? [17:06:17] * yuvipanda is working out of a public place again, after quite a while [17:06:49] awjr: pretty awesome, except for the old lady sitting next to me trying to talk to me in a language I do not understand :) [17:06:53] Change abandoned: awjrichards; "(no reason)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6462 [17:06:55] lolz [17:06:57] lol [17:07:19] I do find working in large open spaces better than working out of a tinier room. [17:07:32] we also seem to have broken scrolling in 4.x *again* [17:07:34] sigh [17:07:37] tfinc: update on the iPhone? [17:08:01] yuvipanda: in the app or MF? android or ios? [17:08:08] app [17:23:15] \o/ fixed toggling on blackberry 9600 [17:24:18] \o/ added support for ie9 [17:26:03] you're an animal [17:28:47] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6490 [17:31:32] because it's considered primitive to use those devices? [17:32:19] hehehe [17:32:36] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6490 [17:33:22] "ew wikipedia mobile changed. my eyes" < do we reckon that's a positive tweet? [17:33:47] jdlrobson: am i reading right that this should disable transitions automagically for android 4.0.2? https://gerrit.wikimedia.org/r/#change,6525,patchset=1 [17:33:53] hahaha [17:34:10] im hoping awjr [17:34:19] testing now [17:34:20] are you still getting transition problems? [17:34:30] if so print me your user agent [17:34:58] one sec [17:36:02] jdlrobson: fixed the scrolling issues :) https://github.com/wikimedia/WikipediaMobile/pull/228 [17:36:10] gooo yuvipanda \o. [17:36:20] o/ [17:36:21] whoops I seem to have lost an arm [17:36:30] well, I considered that a hi5 [17:36:32] so good enough [17:36:37] jdlrobson: still some weirdness with the transitions :( [17:36:41] it was in the wrong direction though [17:36:48] mm there should be no transitions awjr at all. this is all very odd [17:36:52] sigh. [17:37:04] jdlrobson d you know how i can quickly pull up my ua? [17:37:06] can you try something for me? [17:37:13] alert(navigator.userAgent) [17:37:15] yes [17:37:16] add that to top of application.js [17:37:19] k [17:38:07] tfinc: we need to figure out how much we're actually going to support Qt and update https://github.com/wikimedia/WikipediaMobile/pull/173 [17:38:17] hmm [17:38:40] yuvipanda: you mean in terms of code review ? [17:38:57] well, in terms of actually supporting it [17:38:58] jdlrobson: Mozilla/5.0 (Linux; U; Android 4.0.2; en-us; Galaxy NExus Build/ICL53F) AppleWebKit/534.30 (KHTML, like Gecko) Version/4.0 Mobile Safari/534.30 [17:39:01] I don't want to merge it in and then forget about it [17:39:04] also awjr please apply this patch http://pastebin.com/1b7b7d7J [17:39:08] and tell me if problems go away [17:39:16] thanks awjr [17:39:16] We had a Win7 folder for a long time that didn't work at all [17:39:35] tfinc: so I'd rather not merge it in if we're not going to actively support it somewhere [17:39:47] beta skin search is completely broken on S60 [17:39:49] though I'm unsure if there's a market for them, besides the solo Meego phone [17:39:58] * srikanthlogic is filing bugs [17:40:07] yuvipanda: how about we keep it on a experimental branch [17:40:16] thanks srikanthlogic [17:40:18] tfinc: that branch will need to be kept up-to-date with master. [17:40:24] if we wanted to actively support this as a production release we would need more engineers [17:40:24] I'm very short on S60 devices [17:40:26] or it'll rot [17:40:30] you have one? [17:40:35] tfinc: true. [17:40:37] jdlrobson: yes E51 [17:40:44] srikanthlogic: are you around tomorrow? [17:40:53] we can debug it together [17:40:55] New review: awjrichards; "This looks ok, but does not appear to have the intended effect on ICS 4.0.2." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/6525 [17:41:06] yuvipanda overall though .. i have to say this is a good problem to have [17:41:09] jdlrobson: same time tomorrow may be ? [17:41:16] heh Indeed. [17:41:27] boo too we have too much volunteer contribs :D [17:41:28] haha [17:41:36] but yes … we have to have an action plan for this [17:41:43] earlier the better - i'll be around from 8am-6pm UK time [17:41:58] honestly .. i think we should be able to use 20% time for the reviews [17:42:12] jdlrobson: will poke you as soon as am around [17:42:21] excellent! [17:42:28] tfinc: so, reviews aren't really the problem. It's ongoing support that is [17:42:38] right [17:42:48] tfinc: if I merge in that pull req, I actually need to actively test on that as well, or it becomes broken and rots [17:42:49] yuvipanda: especially with you going back to school [17:43:05] let me think about it [17:43:21] tfinc: think about it and respond at https://github.com/wikimedia/WikipediaMobile/pull/173 [17:43:22] ? [17:43:22] yuvipanda: put in a comment though [17:43:30] its been open for 2 months [17:43:47] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/6523 [17:44:03] IIRC we responded to the pull req when it was 'announced' on twitter [17:44:04] but will do [17:44:18] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: -1; - https://gerrit.wikimedia.org/r/6525 [17:44:39] tfinc: updates on the iPhone? I'm pretty sure you wouldn't want to pull on duty as iOS 4.x tester again :) [17:44:56] no i would not … /me goes to ebay [17:45:14] heh [17:45:24] alright, It's time for me to locate my bus. BRB [17:46:09] awjr: did that patch do anything? [17:46:20] i wanna finish for day :) [17:47:23] jdlrobson: what is in master appears better than the no transitions patch [17:47:35] really? mm [17:47:38] ok [17:47:39] one more test [17:47:47] yeah, actually - in master, im not seeing any flicker, stutter, etc. [17:48:15] New patchset: Jdlrobson; "add js support for ie9" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6538 [17:48:15] New patchset: Jdlrobson; "respect supports_javascript property of device" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6539 [17:48:16] New patchset: Jdlrobson; "distinguish between blackberry < v5 and >= 5 + enable javascript on >=5" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6540 [17:48:17] New patchset: Jdlrobson; "allow initialisation of references with options" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6524 [17:48:17] New patchset: Jdlrobson; "allow turning off transitions" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6525 [17:48:18] New patchset: Jdlrobson; "pay attention to firstRun value passed in by user" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6523 [17:48:49] http://pastebin.com/b6JBLdBG < awjr [17:48:59] no flicker at all? [17:49:06] * jdlrobson confused [17:50:19] grr i wish it was easy to look under the hood in the native android browser [17:50:27] i'd like to confirm that im getting the right resources [17:50:45] awjr: I'm pretty sure you can plug your device into your laptop [17:50:48] but yes, jdlrobson, that is correct - right now i've got master at ddd4e09dd3248801a56612ef6a3666ea9ec8a76e and i see NO clicker [17:51:02] jdlrobson afaik i can debug with chrome beta, but not with android native browser [17:51:03] then using adb connect to it and view logs [17:51:11] oo that might be [17:51:11] (console.log statements that is) [17:51:38] jdlrobson i think you're right [17:51:41] http://developer.android.com/guide/webapps/debugging.html [17:52:03] I've connected my android to my laptop for the purposes of moving apps to sd card > http://www.bongizmo.com/blog/moving-all-android-apps-to-sdcard-apps2sd-froyo/ [17:52:22] jdlrobson: http://twitpic.com/9guezm/ peak, bug on the way [17:52:34] eek srikanthlogic [17:52:44] looks look it's going to be a lot of work :) [17:52:57] we'll just reduce support of it to toggling (hopefully) [17:53:06] :) if thats not just, the suggestions never come [17:53:15] yeh I can imagine [17:53:34] anyway we'll sort that all out tomorrow [17:53:48] if i move cursor outside textbox, I get to see what I(T9) typed [17:54:01] jdlrobson: sure, works somewhat smoothly on Opera Mobile [17:54:16] Opera Mini is similar to stock browser [17:56:52] ok cool - keep the bugs coming and talk to you tomorrow :-) [17:56:54] see you later guys! [17:57:49] jdlrobson: before you go: i think i was looking at cached JS/css [17:57:54] im going to try again with your latest patch [17:58:08] sweet that's music to my ears ;-) [18:02:01] yuvipanda: phone purchased [18:03:35] awesome! [18:03:35] * yuvipanda pokes MaxSem about ImagePage refactor [18:04:36] jdlrobson: nope, still seeing weirdness with the patch [18:04:38] :( [18:04:46] well we'll debug it some other time [18:04:53] yeah [18:04:58] are you coming to Berlin? [18:05:05] nope [18:05:09] yuvipanda, doing ершы ыегаа кшпре тщц [18:05:12] won't make it this year [18:05:16] ok worse case scenario it can wait till June to be fixed ;-) [18:05:21] lol [18:05:21] *this stuff right now [18:05:31] or jdlrobson wait for 4.0.4 to replace 4.0.2 :p [18:05:32] that's a better lead time then enterprise software anyhow [18:05:43] MaxSem: அப்படினா ? [18:05:52] ah [18:05:57] accidental language switch, oky [18:05:58] *okay [18:06:10] right cya later [18:06:23] preilly: yuvipanda surprisingly wrote that without spelling mistake :D [18:06:45] and what it means? [18:06:55] MaxSem: ^ that's what it means [18:06:59] :P [18:07:17] jdlrobson i take back what i said about things working in master - im still seeing flicker there. i think it might just not be a totally consistent problem [18:07:18] anyway [18:07:18] srikanthlogic: I was unsure about the last character. Still am :P [18:07:20] talk to you later! [18:07:43] yuvipanda: you passed this time [18:07:48] * MaxSem троллолирует [18:11:52] New review: awjrichards; "Still no change on ICS 4.0.2 :( Perhaps we just need to wait for ICS 4.0.2 to go off somewhere and d..." [mediawiki/extensions/MobileFrontend] (master); V: 1 C: -1; - https://gerrit.wikimedia.org/r/6525 [18:12:43] https://twitter.com/#!/JsonCulverhouse/status/197917815813054464 :D [18:24:17] yuvipanda, do you need to see the "This is a file from the Wikimedia Commons. Information from its description page there is shown below." message as well? [18:26:29] yuvipand1, yuvipand2? [18:26:40] New review: awjrichards; "While we're planning to merge this into core, we haven't done it yet. We've talked a lot about maint..." [mediawiki/extensions/MobileFrontend] (master); V: 1 C: -2; - https://gerrit.wikimedia.org/r/6453 [18:27:39] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 0; - https://gerrit.wikimedia.org/r/6523 [18:32:06] MaxSem: looking, one second [18:32:12] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6539 [18:33:36] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/6538 [18:33:50] MaxSem: Assuming that that is different for on-wiki images vs images from commons, I'd suppose having it there is a good idea [18:34:06] MaxSem: In general, as close to mobile veiw as possible [18:34:14] would be nice :) [18:34:14] awjr, the feature I'm currently doing for yuvipanda requires core changes anyway ;) [18:34:46] MaxSem eh? [18:34:53] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/6540 [18:35:05] awjr, yeh:P [18:35:22] MaxSem: what are you doing for yuvi? [18:35:31] mobile view for image pages [18:36:06] MaxSem you're making the changes in MF and it will be dependent on core changes/ [18:36:35] i dont think that that is necessarily a problem, but MF will need to be able to degrade gracefully to maintain support for 1.19 [18:36:46] same thing with the change you checked in [18:37:03] it has been a long-time practice in WMF development that bleeding-edge extensions are guaranteed to work only with bleeding-edge MW [18:37:41] that doesn't mean it's good practice - we talked yesterday about making sure to maintain compatibility with the stable tarball [18:37:47] if someone needs an extension for an older version of MW, they get it from appropriate branch [18:38:27] MaxSem why not just wrap bleeding edge features in a check and make sure they can fallback to maintain support for 1.19? [18:38:55] we don't really offer any sane way to make it clear what version of MW or ext will work with [18:39:05] version checks everywhere would be mess [18:39:27] awjr, we do: take a look at ExtensionDistributor [18:39:28] it wouldn't need to be everywhere, just where you have 1.20 only feature [18:39:28] s [18:39:40] just like we do for our beta features. [18:41:55] yeah…. extension distributor doesn't actually work, i presume due to the switch to git. and regardless, we've talked about keeping our version of master stable and forkable [18:43:11] MaxSem tfinc asked us yesterday to make sure we maintain compatibility with 1.19 - maybe i'm wrong, but having had to deal with compatibility issues in MW before I dont think it's a huge deal to use checks and fallbacks. [18:44:22] in this particular case fallback means copyright terms not satisfied, for example [18:44:41] everything has its own price [18:44:49] oh really? what do you mean? can you give me some more context? [18:45:34] bleh [18:45:38] fine [18:46:08] copypasta wins [18:48:04] Project WikipediaMobile - Nightly builds build #351: SUCCESS in 23 sec: https://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/351/ [18:48:05] * yuvipanda: Added footer showing license and history link [18:48:05] * yuvipanda: Added contact us link to settings page [18:48:06] * yuvipanda: Hide footer during initial load [18:48:06] * yuvipanda: Footer text update after discussion on bug 35513 [18:48:07] * s.mazeland: Localisation updates from http://translatewiki.net. [18:48:07] * jdlrobson: Revert MobileFrontend to an earlier version with cleaner toggle.js [18:48:08] * jdlrobson: Removed some dead MobileFrontend.init() calls [18:48:08] * jdlrobson: Use MF toggle code for toggling, instead of separate code [18:48:09] * jdlrobson: Uses MF's reference reveal code in the app [18:48:09] * jdlrobson: Removed unused deserialize method from page.js [18:48:10] * jdlrobson: Updated MobileFrontend to latest. Doesn't need patching anymore [18:48:10] * yuvipanda: Make initial page load be empty [18:48:11] * yuvipanda: Handle File URLs in English Wikipedia [18:48:11] * jdlrobson: Move from sitematrix.json to wikis.json [18:48:12] * yuvipanda: Fix race condition causing the non-localized menus in Android [18:48:12] * yuvipanda: Hide keyboard when user starts scrolling through search results [18:48:13] * jdlrobson: Fix iOS saved pages code to work with Page objects [18:49:10] :D [18:49:14] wasn't Jenkins supposed to make one build per change? [18:49:21] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6476 [18:49:24] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6476 [18:49:41] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6477 [18:49:44] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6477 [18:50:04] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6478 [18:50:06] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/6524 [18:50:07] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6478 [18:50:07] MaxSem: it's broken [18:50:22] MaxSem: so I can trigger manual builds, but automatic push-based builds don't work [18:50:31] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6479 [18:50:34] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6479 [18:51:04] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6523 [18:51:06] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6523 [18:51:32] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6538 [18:51:58] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6540 [18:52:29] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6539 [18:53:07] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6524 [18:53:10] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6524 [18:53:43] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6525 [18:53:46] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6538 [18:53:47] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6540 [18:53:48] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6539 [18:53:50] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6525 [18:54:18] Change abandoned: preilly; "(no reason)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6453 [19:07:40] gnite everyone [19:07:53] tfinc: don't forget to mail out the device [19:07:57] speaking of core integration - preilly and awjr, is there anything in your parts of code that could take more advantage of Request and context in general being always available? [19:08:07] yuvipand1: i'll have to get it in the mail first [19:08:15] probably [19:08:23] I keep forgetting transporters haven't been invented yet [19:08:30] yuvipand1: yes [19:08:38] yuvipand1: the star trek panel was awesome [19:08:41] nice vid [19:08:51] Got that from jorm [19:09:05] It was hilarious, esp. WilW's comments about Fanfic [19:09:46] But they in general seem to be not fully aware of how much of an Impact they've had on some people... [19:09:49] oh well [19:10:12] I should be sleeping [19:10:29] gnite :) [19:12:24] tfinc: btw, fixed that issue yesterday. [19:16:15] MobileFrontend/javascripts/filepage.js:52: ERROR - Parse error. Internet Explorer has a non-standard intepretation of trailing commas. Arrays will have the wrong length and objects will not parse at all. [19:18:19] preilly, I love this browser more and more every second! [19:18:50] [19:21:22] Amgine_: thanks! [19:21:31] np [19:21:47] New York won in triple over-time. [20:27:59] awjr: whats the beta feedback page that were using on mw.org ? [20:28:03] i can't seem to find it [20:28:19] http://www.mediawiki.org/wiki/MobileFrontendFeedback [20:28:23] tfinc ^ [20:37:22] CamelCase :( [20:38:18] i had a client once who called it 'camelHumped' rather than 'camelCased' and on multiple occasions accused me of being a camelHumper [20:39:31] mmm what if https://www.mediawiki.org/wiki/Special:Block/Mobilefrontend XD [20:40:06] that would be lame. [20:40:42] maintenance scripts that updated MW namespace aeons ago were impossible to block [20:41:44] New patchset: MaxSem; "Output shared file descriptions on action=mobileview" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6554 [20:42:06] awjr: do you think we should change it based on MaxSem suggestion? [20:42:55] preilly you mean put the feedback page in the MW namespace? [20:43:56] awjr: what he suggested [20:44:17] i guess im not sure what exactly he was suggesting [20:44:43] MaxSem what exactly were you suggesting? [20:46:03] preilly, I didn't suggest it, I merely recalled the times of bots with unstoppable kung-fu:) [20:46:33] awjr, how about using sentence case? [20:47:16] MaxSem: im ambivalent on the casing, personally [20:47:29] I'm not sure about namespace, as main NS's purpose on mw.o is pretty vague [20:47:35] it would probably make sense to put the page into a non-main NS tho [20:47:38] yeah [20:47:51] maybe, Extension:MobileFrontend/feedback? [20:47:59] i like that idea [20:48:34] http://www.mediawiki.org/wiki/Extension:MobileFrontend/feedback [20:50:03] yeah - what do you think preilly? [20:51:15] awjr: I agree with MaxSem seems like a swell idea [20:51:36] awjr: I created the page [20:51:46] awjr: do we have to worry about permissions [20:52:06] awjr: for the Mobilefrontend user [20:52:13] preilly: i don't think so, so long as the page already exists [20:54:01] preilly we can push the config to test.wikipedia.org and try it from there [20:55:21] echo "!isGood"; [21:10:04] tfinc: phil said he wanted to talk about something but wouldn't tell me what. do you know what this might be? [21:15:06] philinje: jorm is looking for you [21:19:52] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/6554 [21:19:54] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6554 [21:19:55] hey preilly, do you know which commit was responsible for changing the search UI, apparently it was this Monday 7pm PST but not sure what timezone these changes are logged in git [21:20:47] drdee_: it was pushed on Monday around 6 pm PDT [21:21:23] preilly: cool, do you have the url to the commit by any chance? [21:22:34] drdee_: https://gerrit.wikimedia.org/r/#patch,unified,6216,1,extensions/MobileFrontend [21:23:35] drdee_: it was updated in the repo before it was actually pushed live [21:25:44] drdee_: do you have a JS file created for device detection [21:26:02] drdee_: I mean devicemap [21:26:38] I can't seem to access http://devicemap-vm.apache.org/logcapture/ [21:29:00] preilly: let me check, 1 sec [21:29:25] preilly: https://svn.apache.org/repos/asf/incubator/devicemap/trunk/prototypes/logcapture/ [21:31:58] drdee_: okay great thanks [21:32:06] welcome [21:45:00] preilly: during times like this … could we show something better then the 503 page? [21:45:48] redirect to encyclopedia dramatica? [21:46:56] binasher: ROFL [21:46:56] that would be hilarious [21:46:56] it'd probably kill them though [21:46:56] hmmmmmmmm…. [21:47:25] just one little iframe... [21:49:18] btw, Russia's largest social network, VK, recently used JS to make its visitors DDoS a captcha recognition service. in response, they put up a redirect to xvideos [21:49:39] hah! [21:50:33] so preilly binasher … what could we show instead of an ugly 503 next time? [21:50:54] anything you like [21:50:57] can we put in a custom page for the 503 ? [21:51:01] * tfinc ponders [21:51:02] have brandon design something? [21:51:10] love to. [21:51:12] how could we combine fail whale and a unicorn ? [21:51:18] with possible rainbows [21:51:19] i was just about to go there! [21:51:26] we could throw up the sopa blackout screen. [21:51:27] this can only end well [21:51:35] scare the fuck out of people. [21:51:53] but now with cispa [21:52:01] (and unicorns) [21:52:15] can we have a unicorn stabbing cispa ? [21:52:22] wrapped in a rainbow [21:53:02] tfinc: are you going to be interviewed by the antropologist as well? :) [21:53:20] one can only hope not [21:53:36] :D [21:53:56] drdee_: are the zero filters back up and collecting data ? [21:54:17] yes, 2 have been collecting since day 1 [21:54:35] the other 3 have been launched this week (tuesday iIRC) as Oxygen is now up and running [21:54:43] Oxygen is our brand new third filter box [21:54:45] woot [21:54:46] thanks [21:55:09] deploying of any additional filters should be a breeze [21:56:54] great [21:57:01] actually, do you just need an html file? [21:59:26] jorm: likely [21:59:37] binasher would know better [22:04:19] we'd need to embed the html into a varnish config but just an html file would be good [22:04:36] we could potentially use it on the main site too if it came that way [22:19:15] [WikipediaMobile] siebrand pushed 1 new commit to master: http://git.io/DX-JNQ [22:19:15] [WikipediaMobile/master] Localisation updates from http://translatewiki.net. - Siebrand Mazeland [23:15:27] New patchset: awjrichards; "Adds support for new method WebRequest::removeQueryValue rather than relying on homegrown removeQueryStringParameter, but falls back to original implementation if the method in WebRequest does not exist." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6569 [23:20:33] tfinc: got feedback from Kelly and Philippe [23:20:40] philinje: great [23:20:45] i'm wrapping up a phone screen [23:22:00] awjr: 1270 » » » $this->coreRemoveQueryValueExists = (extra whitespace) [23:23:48] fml [23:25:15] New patchset: awjrichards; "Adds support for new method WebRequest::removeQueryValue rather than relying on homegrown removeQueryStringParameter, but falls back to original implementation if the method in WebRequest does not exist." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/6569 [23:27:49] awjr: 1269 » » if ( !isset( $this->coreRemoveQueryValueExists ) ) { [23:28:18] awjr: coreRemoveQueryValueExists should probably have a default value [23:28:39] awjr: then you can avoid isset altogether [23:28:59] i guess it could get set in the constructor [23:29:40] preilly what about adding a setter for it and having it get set in the constructor? [23:29:50] awjr: protected $coreRemoveQueryValueExists = null; [23:30:13] awjr: you could do that as well [23:30:42] preilly isn't  protected $coreRemoveQueryValueExists = null; essentially the same as it is now? [23:31:39] awjr: I was trying say you should have a default value [23:31:41] how about making it a static in-function variable? it doesn't need to be a per-object var [23:32:31] MaxSem that is true - it crossed my mind but i didn't really have any better reason to do it the way that i did [23:32:37] actually if we are going to start doing more capability testing we should make the whole thing more robust [23:32:53] as we're moving stuff into core we're definitely going to need to do that [23:33:36] what about providing a class for capability testing with static methods? [23:33:38] some sort of method_exists wrapper [23:34:18] if we build a separate class specifically for these kinds of checks, we can easily keep it all organized in one place [23:34:39] awjr: that makes a look more sense to me [23:34:46] s/look/lot [23:35:02] MaxSem what do you think? [23:35:58] awjr, you mean preserve locally the stuff moved to core for some time? [23:36:46] MaxSem: we'll have to do that regardless, but i'm suggesting that we make a class with static methods specifically to check for new functionality in core [23:39:41] awjr, http://dpaste.com/742319/ [23:39:55] 1 extra line per function [23:40:53] also, kept in 1 place instead of capabilities class and calling class [23:40:53] MaxSem if we take that idea but move those checks into their own class, i think we could more easily keep track of what all the possible checks are [23:41:11] that is a good reason [23:41:29] that way we can find those things at-a-glance [23:41:36] you may grep for a comment instead of class name as well;) [23:41:44] lol also true [23:45:13] preilly, do we still need search_bar in DeviceDetection? [23:46:40] awjr: what about $oldVersion = version_compare( $wgVersion, '1.18', '<=' ); vs. an is_callable call [23:47:15] MaxSem: nope [23:47:22] hmm, there are other unused capabilities, too [23:47:32] * OuKB bites MaxSem [23:47:39] preilly: what about bleeding edge changes tho? [23:47:57] * preilly sees a Sockpuppet [23:48:29] awjr: well then is_callable is fine [23:48:47] awjr: hi [23:49:16] philinje: hello [23:49:41] preilly: we could do both and just stick to is_callable for super new stuff but rely on your suggestion for everything else [23:50:49] awjr: that would be more performant [23:50:58] i like it [23:58:42] awjr: i posted some comments about the beta on the talk page of the Contact form project page