[00:01:00] so my emulator is 4.0.3 and doesn't have the choppiness [00:01:03] trying chrome beta just for fun [00:01:18] brion it seems to work fine for me in chrome beta and dolphin hd [00:01:26] i get non-mobile view in chrome o_O [00:02:04] brion do you have &mobileaction=beta [00:02:40] ok forcing it with &useformat=mobile also.... in chrome beta it looks a LOT smoother [00:02:58] feels a bit choppy to me, but acceptably so -- no horrible visual flickering [00:03:20] yeah, like, a big sluggish but no flicker [00:03:27] definitely not "iphone smooth" [00:03:38] if i had a nickel.. [00:03:55] :) [00:05:41] so.. browser sniffing? :S [00:05:56] oh actually im seeing the flicker in dolphin hd now too [00:06:00] WTF, libxml does not close self-closing tags at all when saving HTML [00:07:40] jdlrobson what about just not animating the toggle? [00:07:55] or does that just look horrible? [00:08:41] jdlrobson: hi, did the comments about the TOC make sense? [00:08:55] philinje: you tell me... did I implement them correctly ? :) [00:09:16] that's what I mean awjr1 - we could prevent animations on toggle for this particular phone [00:09:18] an isee [00:09:37] MaxSem: LIBXML_NOEMPTYTAG [00:09:38] the original idea was a full-screen sliding in from the right [00:09:52] jdlrobson probably the easiest thing to do for now in this case [00:09:55] MaxSem: see http://cz2.php.net/manual/en/libxml.constants.php [00:10:06] MaxSem: Expand empty tags (e.g.
to

[00:10:10] preilly, I'm attempting to switch from saveXML to saveHTML [00:10:25] MaxSem: This option is currently just available in the DOMDocument::save and DOMDocument::saveXML functions. [00:10:29] but the way you have it is sort of ok [00:10:52] philinje: that's what I have done [00:10:55] no [00:10:55] ? [00:11:03] jonrobson.me.uk/wikipedia/experiments/nav/index.html [00:11:33] full-screen meaning the whole screen? [00:12:10] New patchset: MaxSem; "Bug 36018 - Empty XHTML tag br in mobile view" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5869 [00:12:52] jdlrobson: Lindsey took a look and prefers the whole screen approach [00:13:04] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5867 [00:13:06] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5867 [00:13:15] jdlrobson: also, she will send the 4 variations in rough form based on this design [00:13:25] philinje: are you saying that you want the table of contents to hide the search as well... I'm really confused now? [00:13:31] yes [00:13:37] but how would you close it then? [00:13:48] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5868 [00:13:50] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5868 [00:13:56] needs to be a back, hang on, let me send the comp [00:14:11] https://upload.wikimedia.org/wikipedia/mediawiki/6/65/7-article-actions.png [00:14:44] 2nd one down on the left [00:15:13] jdlrobson: does that make sense? [00:15:16] MaxSem: so, you are defaulting to saveHTML();? [00:16:02] preilly, yes, because XML content in a HTML5 skin created some problems [00:16:13] couple other things: the icons in the main menu, and the transition speed seems slow [00:16:27] MaxSem: just the empty tags [00:16:28] ? [00:17:22] Yes. Much more. I guess I'm just used to UI's where I can click on the same icon again to undo an action. Not such a fan of these sort of overlays [00:17:45] jdlrobson: also, let's replace the W button with a new one Lindsey will send which doesn't have the little arrow inside [00:17:45] awjr: hoping you can test something in a bit [00:18:16] philinje: will have to wait tomorrow now though - I'm keen to fix a bug in the beta then sleep :) [00:18:36] ok, just want to send this out tomorrow [00:18:48] preilly, yes. I noticed they were mangling beta opt in dialogs [00:19:16] MaxSem: so, you saw
or
or

[00:19:43] was:

- treated as a double newline [00:19:48] now:
[00:19:58] MaxSem: which is NOT xhtml [00:20:24] preilly, we don't output XHTML anymore [00:20:35] In HTML the
tag has no end tag. [00:20:36] In XHTML the
tag must be properly closed, like this:
. [00:24:06] jdlrobson philinje I think the problem is that those wireframes were created with the iOS app in mind and what jon is working on is the mobile web browser version - we'll need to consider the differences in interactions and find out what makes most sense for the platform we are using [00:24:55] MaxSem: http://www.w3.org/TR/html5/syntax.html#void-elements [00:25:20] MaxSem: so, I guess I'm fine with in outputting just a
[00:25:26] preilly: MaxSem : can you guys throw in an update for http://www.mediawiki.org/wiki/Mobile_support_in_MediaWiki_core [00:28:22] agreed linSmith [00:29:07] linSmith: Can you include the second screen in the rough mock-ups, meaning the step showing the actual TOC? [00:29:16] tfinc: updated http://www.mediawiki.org/wiki/Mobile_support_in_MediaWiki_core [00:29:20] thanks [00:30:55] tfinc: np [00:31:34] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5788 [00:31:37] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5788 [00:31:56] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5789 [00:31:58] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5789 [00:32:26] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5797 [00:32:29] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5797 [00:32:45] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5802 [00:33:16] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5869 [00:33:18] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5869 [00:33:47] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5801 [00:33:49] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5802 [00:33:51] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5801 [00:34:10] MaxSem: are you happy with https://gerrit.wikimedia.org/r/#change,5361 ? [00:35:09] linSmith: please focus on the web browser versions for now, would be great to get those to him tonight if possible [00:35:13] preilly, I can be proud of myself:P ask awjr if I'm wrong about this;) [00:35:24] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5361 [00:35:26] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5361 [00:35:48] tfinc: http://bit.ly/IaGqaX - do you see slow toggling on this page? [00:35:54] testing [00:36:20] i think if you do we should just turn off css transitions for the time being [00:36:45] jdlrobson: Lindsey will send the 2 web versions early tomorrow, which is late tomorrow for you [00:36:56] ok philinje great [00:37:25] while you're waiting, the icons in the main menu and the transition speed would be great to work on [00:37:28] the only change I need to do is to make the table of contents full screen? [00:37:58] preilly, thanks [00:37:58] let's leave that for now, until Lindsey sends the comps [00:38:31] MaxSem: no no thank you! [00:38:47] * tfinc is still waiting for the page to load [00:38:50] ok .. done [00:39:01] MaxSem: I'm getting some weirdness in my footer [00:39:04] MaxSem:
  • test
  • [00:39:34] MaxSem:
  • <li class="noprint">test</li>
  • [00:39:39] tfinc: did any of those emails look like known issues/open bugs? i can reply to them [00:39:46] jdlrobson: tiny bit better. it feels like we collapse in one motion rather then two [00:39:56] RD: i haven't taken a deep dive yet [00:40:00] but i'll let you know [00:40:08] OK. Sounds good. [00:40:08] s/collapse/expand [00:40:10] Thanks :) [00:40:12] tfinc: is it good enough or should I just scrap animations altogether? [00:40:47] hmm [00:40:52] do you have a version up without animations ? [00:41:02] I can do it very easily.. [00:41:04] 1 minute [00:41:05] i'd like to test them side by side [00:41:06] MaxSem: http://preilly.m.mediawiki.org/index.php?title=MediaWiki:Copyright&action=edit [00:41:06] thanks [00:41:14] MaxSem: is
  • test
  • [00:41:48] tfinc: refresh now [00:41:52] should have no animations [00:42:35] you may need to cache bust [00:42:44] jdlrobson: looks the same to me. should i flush my cache? [00:42:46] k [00:43:08] actually.. wait a minute [00:43:12] something strange happening here [00:44:12] ahh sorry didn't realise awjr had set up the new instance already [00:44:43] ok please retry now tfinc [00:44:49] this is what you should have seen the first time [00:44:53] no difference after cache clear [00:45:36] let me try with your latest chance [00:45:39] change* [00:45:56] argh awjr i'm confused with what you've done to mobile-geo :S [00:46:19] i've just messed with your branch as well and made things even more confused [00:46:23] jdlrobson: i'm still seeing the animation but it feels better now [00:47:10] tfinc: yeh.. so I've messed up the dev version somewhat but if the animation works nicer and you would consider the animation deployable that means I have fixed it [00:47:11] i'm happy with this version [00:47:13] ok cool [00:47:18] :) [00:47:25] one more win before you crash for the night [00:47:26] i've now completely broken that instance [00:47:39] preilly, what copyright-related options do you have in LocalSettings? [00:47:39] good work jdlrobson you can now go to bed feeling accomplished [00:48:17] also philinje: I'm still a little bit confused about what you want from the mockup since we keep changing minds so please make sure it's extremely clear in your e-mail about what needs to be done... also the beta comes first so please ensure you prioritise what I should focus on as tomorrow is my last day and I don't want to disappoint you/mess with your plans [00:48:33] (last day of the week) [00:48:39] which is actually today :-) [00:48:41] yes, got it [00:48:46] tfinc: just going to push that now [00:49:22] conversations are happening all the time, i will send an email before you wake up [00:49:23] great philinje :) [00:49:34] preilly, it's almost morning in my parts, will continue tomorrow [00:49:39] also images help a lot :) [00:51:58] New patchset: Jdlrobson; "disable animations on buggy ios 4.0.2" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5873 [00:52:50] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5873 [00:52:52] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5873 [00:53:05] jdlrobson: do we really want all these /* fix for bug 36190 */ comments all over the place? [00:53:22] jdlrobson: isn't just linking to the commit from the bug good enough? [00:53:29] disappears in minification so it's useful for working out why they are there [00:53:37] sometimes I look at code and think why is this here [00:53:41] and sometimes git review is no help with that [00:53:52] jdlrobson: okay, that makes total sense [00:54:02] sweet [00:55:08] MaxSem: $wgRightsPage = "MediaWiki:Copyright"; [00:55:28] MaxSem: $wgRightsUrl = ""; [00:55:29] $wgRightsText = ""; [00:55:30] $wgRightsIcon = ""; [00:55:43] awjr: please check https://bugzilla.wikimedia.org/show_bug.cgi?id=36190 as I'm off to bed. I'll look again tomorrow with Yuvi's help if it's still a problem (hope Yuvi has that phone) [00:56:39] jdlrobson: thanks, for all your hard work today — it's awesome having you working with us [00:57:39] thanks preilly looking forward to getting out there so i can do more :-) [00:57:50] night all! [08:01:33] hallo YuviPanda [08:01:38] hello aharoni [08:01:50] there were no nightly builds of the app lately? [08:02:22] hmm, we fixed them a week or so back? [08:02:29] right, they seem to have broken :D [08:02:42] aharoni: will poke hashar and see what's up [08:02:46] i'm looking at http://dumps.wikimedia.org/android/ [08:02:54] let's do something simple. [08:03:05] aharoni: oh, that is not the nightly [08:03:09] nightlies at https://integration.mediawiki.org/WikipediaMobile/nightly/?C=M;O=D [08:03:18] that is just where we put our betas and RCs (dumps.wikimedia.org) [08:03:23] oh [08:03:36] 19-apr is too old :) [08:03:43] i was in Kazakhstan a few days ago [08:04:16] in a Turkic Wikimedia Conference. Turkic languages, like Turkish, Kazakh, Uzbek, Sakha etc. [08:04:28] so, with Sakha there's a funny bug. [08:04:42] :D [08:04:46] can you try a little something with a recent version? on some device or emulator? [08:04:52] aharoni: sure [08:04:56] i've latest version on my device [08:04:57] are you familiar with the "More Locale" app trick? [08:05:10] Gerard wrote about it: http://ultimategerardm.blogspot.com/2012/02/trick-android-to-use-your-language.html [08:05:26] aware of it, but haven't actually used it. [08:05:48] do you have a couple of minutes to try it? [08:06:02] yes! [08:06:11] i'm downloading said app [08:08:26] aharoni: which locale would you want me to try things with? [08:08:33] language code sah [08:09:52] aharoni: yup [08:10:03] ok, now look at the icon for the wikimedia app., [08:10:23] in what language does the icon title appear to be? [08:10:36] devanagri? [08:10:39] ha! [08:10:42] hi [08:10:43] :P [08:10:48] bad joke, nevermind :) [08:10:51] but yes, devanagri [08:11:04] language sah is Sakha. [08:11:18] it's spoken in Russia and written in the Cyrillic alphabet. [08:11:33] your phone, as well as mine, things that it's not "sah", but "sa" - Sanskrit. [08:11:35] so, I see no strings for us from sah. I see 'sa' and it is devenagri [08:11:36] yeah [08:11:45] really? [08:11:48] open the app. [08:12:02] check out in which language do you see the search box and the menus. [08:12:36] yeah, so in *our* app it works [08:12:39] in android it doesn't [08:12:46] aharoni: app name is localized separately [08:12:56] aharoni: since it has to go through java [08:12:59] do you see strings in the Cyrillic script in the app itself? [08:13:14] yes [08:13:14] i do [08:13:19] ok, so that's the bug. [08:13:31] it's probably the first app in history that was translated to Sakha. [08:13:40] interestingly, trying to add 'sah' to android's resource file tells me it is not valid [08:13:53] aharoni: :D That feels good :) [08:14:14] so, is it an Android OS bug? [08:14:48] aharoni: possibly. I'm investigating. [08:14:56] i thought that it may be a bug in the Android OS or in the More Locale app. [08:14:59] aharoni: do file a bug on bugzilla and assign to me? [08:15:05] OK [08:15:13] aharoni: well, it is reporting correct locale to us (sah) since we see it in the app [08:15:22] when i saw that the icon is devanagari and the app is in Sakha, i thought that it may be a mix of a bunch of things... [08:16:01] hmm. where do you see the locales? [08:16:04] aharoni: it is essentially that android handles app name under icon, while we manage app. [08:16:05] https://github.com/wikimedia/WikipediaMobile/tree/master/res ? [08:16:14] aharoni: for the app's name alone, yes. [08:16:21] it is a bit out of date - you'll see no sah entry [08:16:47] oh no, that's just the title. [08:17:05] yes [08:17:07] maybe adding an entry there will fix it? [08:17:13] yes, that's what I'm trying right now [08:17:34] except for the fact that android complains that values-sah is not valid :) [08:17:36] trying to figure it out [08:22:02] https://bugzilla.wikimedia.org/show_bug.cgi?id=36264 [08:23:58] aharoni: thanks [08:26:02] aharoni: question - do you know of another 'popular' language that has a three letter code? [08:26:06] *only a [08:26:14] (for testing) [08:26:18] ehh... let's see... [08:28:38] nothing really popular... [08:28:45] maybe lad/la, bar/ba [08:29:03] ceb/ce [08:29:07] hmm [08:29:15] sakha has only a 3 letter code [08:29:37] oh, you mean a language that has both 3-letter and 2-letter? [08:29:46] no, a language with *only* a 3 letter code [08:29:52] sakha has no 2 letter code [08:30:00] yes, sah is supposed to have only 3-letter. [08:30:09] do you know of any other such languages? [08:30:21] http://noc.wikimedia.org/conf/langlist [08:30:31] or better: [08:30:50] in MediaWiki, languages/Names.php [08:31:17] usually, if it's 3 letters in Names.php, it only has a 3-letter code. [08:31:44] awesome [08:31:45] looking [08:32:21] according to the standard, all languages with 2-letter codes have also 3-letter codes, for example English is 'en' and 'eng'. But the suggestion is to use the shortest code. [08:33:06] (I don't like this suggestion, but that's what it is...) [08:33:13] :) [08:33:15] * YuviPanda tries bho [08:33:35] uh oh [08:33:50] aharoni: names.php says bho is bhojpuri [08:33:57] http://en.wikipedia.org/wiki/Bhojpuri_language says it is also bh? [08:34:37] that's one broken-ish case... [08:34:57] by mistake, Wikimedia uses 'bh' instead of 'bho'. [08:35:17] it should be 'bho', without getting into details. better try something else. [08:35:44] :D [08:35:45] ok [08:36:22] * YuviPanda tries kbd [08:36:25] only 3 letter code [08:37:09] i LOVE kbd! [08:37:34] right, so that errors out as well :| [08:38:13] aharoni: your love for languages is actually quite contagious :) [08:38:35] aharoni: so, this *might* be a bug in Android's build system. It doesn't seem to accept three letter language codes [08:38:58] anything with a three letter code gets rejected [08:39:01] ooga-chaka. [08:39:14] let me update the bug, then trawl android bugs to see if this has been reported [08:39:22] else we can report it and watch it get (mostly) ignored :) [08:39:40] * YuviPanda sighs about android bug responsiveness [08:39:42] that's the main reason i don't like the separation between 2-letter codes and 3-letter codes. [08:40:04] why couldn't everyone have 3 letters and have fun? Would've lasted us until First Contact [08:40:46] languages with 2-letter codes are usually the more popular ones. this is an example of how less popular languages get neglected. [08:40:53] i'd deprecate the 2-letter codes. [08:43:05] sigh. [08:45:24] aharoni: I'm happy to see that while Tamil renders horribly broken in rest of Android, it renders just fine on our app :) [08:45:49] an accidental advantage of being on webkit :) [08:46:38] and you probably have the fonts, too. [08:47:01] so you have a font file; the Webkit browser renders it correctly; and the rest of the OS renders it incorrectly? [08:48:09] IIRC it comes included with 4.0 version of Android [08:48:22] (the font file) [08:48:31] yes, webkit renders it correctly, rest of OS renders it incorrectly [09:08:34] New review: Jdlrobson; "Was there a reason why $wgMFMinifyJS was introduced?" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5361 [10:06:00] New review: MaxSem; "Because ExtMobileFrontend doesn't do frontend stuff anymore and awjr disliked the special-casing wit..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5361 [10:07:07] aharoni, any other localisation concerns about MF? [10:07:56] MaxSem: not currently. [10:08:06] i saw some nice fixes in the last few days - thank you. [10:17:17] Hey YuviPanda just looking at the CC-By-SA license bug (35616) then my last remaining bug for beta is v [10:17:17] https://bugzilla.wikimedia.org/show_bug.cgi?id=35513 [10:18:09] jdlrobson: right. I shall now go make patches that make the apps use the beta version [10:18:18] i'm going to assume it is pointless to test master, since it doesn't depend on it anymore [10:18:29] agreed [10:18:44] and it should become obvious if master has a problem :) [10:18:54] jdlrobson: also, there are 4 unclosed pull reqs from me :) [10:19:12] yep well when i've done this bug https://bugzilla.wikimedia.org/show_bug.cgi?id=35513 - i'm going to switch to app mode [10:19:19] I have lots of e-mails with stars on in my gmail from you [10:19:23] :D [10:19:28] that need to be unstarred :) [10:19:35] in this circumstance I don't like stars :) [10:20:11] :D I'm still at 'select everything, archive everything' mode of handling email :D [10:21:26] jdlrobson: is there a way to directly enable beta without asking the opt in question? [10:21:40] append mobileaction=beta [10:21:41] i remember that with mobileaction you could do opt_in_cookie and it'll just set cookie... [10:21:43] ah [10:21:43] right [10:21:45] to url [10:22:19] jdlrobson: does that 'toggle' it or is it 'per page'? [10:23:54] per page [10:24:03] so if you append that to all requests.. you get beta [10:24:16] what server are you pointing the app at? [10:24:19] cool [10:24:27] currently production. i've beta 1.1.1 [10:24:31] will change to custom host now [10:24:38] jdlrobson: where do you want me to point it to? [10:24:50] well localhost would be ideal if that's possible [10:24:56] I can do that [10:24:59] otherwise we could use the very slow mobile-geo.wmflabs.org/w/index.php/Main_Page [10:25:21] i just want to make sure we test against a dev version rather than current production [10:25:28] as minor tweaks have been made but might be damaging [10:25:40] jdlrobson: so essentially i'm rewriting all urls before they go out. so am regex-replacing the host. [10:25:47] far from ideal, but it should work for our purposes [10:29:48] jdlrobson: poke me when you're available to go into app mode :) [10:29:53] k will do [10:52:01] New patchset: Jdlrobson; "correct bad commit 70ccfa053d1576b6a97764b448c5c98be0a3cf7c" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5879 [10:53:51] YuviPanda: did we decide what text to use for https://github.com/wikimedia/WikipediaMobile/pull/213 [10:53:57] or is that pull now out of date [10:54:11] jdlrobson: nope, it has text and text + images. We're still bikeshedding [10:54:23] i'll close then [10:54:27] until we've decided :) [10:55:19] jdlrobson: i'd suggest leaving it open so that we actually *do* decide [10:55:48] ok [10:56:57] so YuviPanda I'm guessing I can't fix the footer text on the mobile site either [10:57:07] because of bikeshedding or.. ? [10:57:11] yeh [10:57:17] I'd rather just have one commit saying this is what we do [10:57:24] otherwise we'll be changing the production version [10:57:40] right. so holding on to that seems fine for now, though we should push that out soon. [10:58:43] ok let me go into app mode [10:58:54] okay, let me get you a patch that'll hit localhost from the app [10:59:04] though i'll note that you need to give it an IP, rather than 'localhost' [10:59:08] since localhost on the phone is the phone itself [10:59:59] i'm going to merge some stuff first.. [11:00:05] cool [11:02:47] i'm just going to test them on ics and osx if that's ok [11:02:58] ok [11:03:18] jdlrobson: i'd prefer it if you tested them on 2.x and iOS, since i do device testing on iCS anyway [11:03:36] ok np [11:03:47] i can only run one android emulator at a time you see [11:03:58] yeah, so gingerbread! [11:04:00] 2.3.3 ok? [11:04:14] also any preference on ios version? [11:04:28] jdlrobson: good question. [11:04:42] 4.x and 5.x have enough differences... [11:04:50] so pick one at random [11:05:35] android emulator is sooo slow [11:06:19] woww toggling is glitchy on iphone 4.2 [11:06:26] can you check your master and see if it's the same for you? [11:06:46] we might need to update to latest sections.css [11:06:56] jdlrobson: one of the pull requests does that [11:06:59] (it's flickering like crazy) [11:07:01] ok cool :) [11:07:06] jdlrobson: is this the 'two buttons' issue or something else? [11:07:08] you are always one step ahead :) [11:07:08] flickering? [11:07:25] it's very noticeable on the emulator [11:07:42] so load up iphone 4.2 and check it out [11:08:22] i've only 4.3, switching branches now [11:09:09] it's flickery [11:09:22] plus expand + collapse -> extra spacing under the headers [11:09:45] jdlrobson: also, http://hastebin.com/sajawawixe.js is the diff on top of v1.1 branch that'll let you point it anywhere [11:09:55] jdlrobson: i'm looking into the flickr. [11:09:58] *flicker [11:10:09] update to latest css sections.css [11:10:12] i believe that will fix it [11:10:17] had similar problem in mobile site [11:11:10] oh [11:12:13] jdlrobson: wonderful. still flickers and scrolling is now messed up again [11:12:16] pages half cut off [11:13:12] scrolling is messed up? [11:13:21] jdlrobson: yes, looks like iscroll is not being reset. [11:13:32] it doesn't seem to update itself to the new document height [11:13:35] + there is flicker [11:13:36] it scrolls on my master.. [11:14:20] after expansion, are you able to scroll down to the bototm? [11:15:00] yes [11:15:10] jdlrobson: latest MF CSS? [11:15:37] ah no.. haven't tried latest [11:15:52] i'll look into that in a bit [11:15:56] jdlrobson: so, adding a 500ms timeout to the iScroll reset fixes it [11:16:08] lets hold of doing that if we can :) [11:16:16] yup [11:17:11] [WikipediaMobile] jdlrobson pushed 3 new commits to master: http://git.io/wcl1aQ [11:17:11] [WikipediaMobile/master] Don't reset scroll when not needed - YuviPanda [11:17:11] [WikipediaMobile/master] Removed unneccessary .empty() calls - YuviPanda [11:17:11] [WikipediaMobile/master] Merge remote-tracking branch 'yuvipanda/scroll-reset-on-overlay-fix' into tmp - Jon Robson [11:18:31] mm doesn't look like my rebase worked [11:18:56] :) [11:39:56] jdlrobson: i'll make a commit removing that class. [11:39:59] makes sense [11:40:14] New review: Jdlrobson; "I'm not happy with the change to toggle.js The introduced code should be put in a separate javascrip..." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5771 [11:42:20] jdlrobson: force pushed, also updated to MF as of today than yesterday. [11:42:29] thanks [11:47:53] [WikipediaMobile] jdlrobson pushed 3 new commits to master: http://git.io/8gcf9Q [11:47:53] [WikipediaMobile/master] Updated to latest version of MobileFrontend - YuviPanda [11:47:53] [WikipediaMobile/master] Remove extra button in section headers - YuviPanda [11:47:53] [WikipediaMobile/master] when toggling add openSection class to button - Jon Robson [11:48:21] right YuviPanda will look at fastclick in a bit [11:48:26] ok [11:48:26] have some errands to run [11:48:30] jdlrobson: ok [11:48:32] note my last commit I added ^^ [11:49:05] * YuviPanda pulls to check [12:35:07] YuviPanda: do you think you could amend the commit for https://github.com/yuvipanda/WikipediaMobile/commit/3f3d5ec9d848e1e20f46aa58ceec7cb74fcd0f3e#commitcomment-1260992 [12:48:43] jdlrobson: sure. [12:48:48] sorry, stepped out for some coke [12:48:53] thanks YuviPanda [12:49:37] jdlrobson: i'm sure there's a guideline for commit messages for mw. I think we should start adhering to it so I don't keep doing these things [12:49:41] have you tested the older app against the new site or do you need me to do that? [12:49:55] I just like long commit messages... they are helpful [12:50:07] jdlrobson: i've a patch that tests the older app against the new site. I see that footers are missing [12:50:38] jdlrobson: otherwise things *mostly* seem to work, though i'll do a bit more heavy duty testing in a bit. Thoughts on the footer? [12:50:48] i think I even sent you a link to that patch... [12:50:55] mm I missed that [12:51:06] right [12:51:06] http://hastebin.com/sajawawixe.js [12:51:08] jdlrobson: ^ [12:51:11] on branch v1.1 [12:51:14] ah yes [12:51:43] why would the footer be missing... i thought we didn't have a footer in the app? [12:51:52] jdlrobson: we did :) [12:51:58] mm [12:52:03] jdlrobson: whatever was the footer's contents were being displayed [12:52:09] well if it goes is that necessarily a bad thing? since it's an old app I mean [12:53:04] I personally wouldn't sweat it. [12:56:46] right I need to step out in a bit so will have limited internet [12:56:52] jdlrobson: ok. [12:57:00] hmm, I should've actually made this two commits instead of one. sigh [12:57:01] but you should see a merge soon for the fast click once you've fixed that commit [13:02:19] right, I seem to have messed up the conflict resolution [13:02:20] * YuviPanda tries again [13:56:54] [WikipediaMobile] jdlrobson pushed 3 new commits to master: http://git.io/9QoWDg [13:56:54] [WikipediaMobile/master] Fastclick for all menu items - YuviPanda [13:56:54] [WikipediaMobile/master] Replace doFocusHack with more descriptive setupFastClick - YuviPanda [13:56:54] [WikipediaMobile/master] Make search focus properly - YuviPanda [14:04:27] [WikipediaMobile] yuvipanda pushed 1 new commit to master: http://git.io/boWY0w [14:04:27] [WikipediaMobile/master] suggested amendments to STYLE_GUIDELINES - Jon Robson [14:05:32] New patchset: Jdlrobson; "make beta the default" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5885 [15:37:56] preilly, hi [15:38:12] regarding that problem you reported yesterday... [15:38:41] how do i repro it besides addingthat HTML to Mediawiki:Copyright [16:04:47] MaxSem: i'lll be ready to chat in about 10min [16:06:16] okie [16:14:20] ok .. i'm ready [16:30:12] jdlrobson, do we need a blogpost to announce the switch to beta? [16:30:28] I expect so MaxSem [16:30:33] I can work on something over the weekend [16:37:13] jdlrobson: so, teseting on iOS seems fine as well. [16:37:14] *testing [16:37:18] sweet [16:37:22] checking older version of android now [16:38:06] jdlrobson: note that our latest update to MF seems to have broken scrolling on iOS 4.x. Using a setTimeout seems to fix it, but you wanted to investigate further (just a reminder) [16:38:21] argh i forgot about that [16:38:34] i'm away tomorrow as well but should be able to grab some time saturday [16:39:10] shouldnt the feedback on beta be collected at one place instead of seperate wikis ? [16:40:20] jdlrobson: we need to keep a counter of how many times scrolling breaks :( [16:40:55] yeh... even better yuvi.in/scrolling-status.html > Y days since last time scrolling broke [16:43:44] for all we know it coulr actually be much more broken, considering we don't have a device [16:43:45] true that [16:43:59] jdlrobson: Y days since scrolling broke and someone noticed? :) [16:44:58] :) [16:45:05] New patchset: Jdlrobson; "bug 36261 drop -webkit-overflow-scrolling in favour of jump to top of page" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5889 [16:45:35] is the scrolling problem present on ios4 emulator? [16:45:38] or just real device? [16:48:49] jdlrobson: the 'cut off'? [16:48:53] jdlrobson: on the sim [16:49:49] the 'cut off' ? [16:51:14] jdlrobson: well, you expand a section and iScroll doesn't detect the new height [16:51:21] ok let me see [16:51:36] * jdlrobson has a feeling he might no the problem [16:51:44] it might be because we're not using toggle.js :) [16:55:31] jdlrobson: if we were using toggle.js, we'd not be calling iscroll at all :P [16:55:41] ? [16:56:28] :) nevermind. [16:56:44] we'll switch to toggle.js when toggle.js does on-demand DOM loading, no? [16:58:03] YuviPanda: found another toggle.js bug [16:58:11] so sections are being cut of [16:59:05] we need this commit https://gerrit.wikimedia.org/r/#change,5801 [16:59:11] the one we've currently got pinned is broken [16:59:29] * YuviPanda looks [16:59:42] actually scrap that I think I forgot to do git submodule update [16:59:51] :) [16:59:52] do you mind if I commit directly to master for this minor bug [17:00:27] actually git submodule update didn't work [17:00:30] problem still there [17:00:38] jdlrobson: so unrelated? [17:02:28] no.. we need to update the commit we use for the submodule [17:02:35] i'll do that now [17:02:40] ok [17:05:19] how do I change the submodule commit? [17:05:22] never done this before? [17:05:29] jdlrobson: cd to assets/www/MobileFrontend [17:05:36] jdlrobson: pull/update to whatever commit you want [17:05:44] get back out of that folder [17:05:49] and you can do git commit -a [17:06:33] ahh wait a minute I can't do it anyway as it hasn't been merged yet https://gerrit.wikimedia.org/r/#change,5879 [17:07:00] augh, right [17:08:51] [WikipediaMobile] jdlrobson pushed 1 new commit to master: http://git.io/NPWDHw [17:08:51] [WikipediaMobile/master] add togglefix class - Jon Robson [17:09:02] so we just need to repin when that's been merged [17:09:39] this fixes it? [17:09:41] * YuviPanda pulls to try [17:12:30] jdlrobson: can you move that code to a new js file [17:12:46] jdlrobson: so, seems fine with beta on v1.0.x as well [17:12:50] jdlrobson: I only put it in toggle because I was thinking of disable/enable as a toggle like event [17:13:09] ok preilly np [17:13:28] YuviPanda: that doesn't fix scroll problem it fixes a possibly unrelated problem [17:13:35] (once preilly has merged https://gerrit.wikimedia.org/r/#change,5879) [17:15:43] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5879 [17:15:45] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5879 [17:16:20] * jdlrobson salutes preilly  [17:16:42] jdlrobson: couple of small changes [17:17:20] jdlrobson: that does fix it :) want to commit it or should I? [17:17:56] jdlrobson: nevermind, pushed [17:18:01] philinje: just fixed up your demos [17:18:03] [WikipediaMobile] yuvipanda pushed 1 new commit to master: http://git.io/k0WMrQ [17:18:03] [WikipediaMobile/master] Updated to newer version of MF that fixes scroll issues on iOS 4 - YuviPanda [17:18:05] gotta shoot off soon [17:18:12] good work YuviPanda [17:18:30] * jdlrobson checks scroll issue [17:19:47] * jdlrobson hopes that we will have a decision on https://bugzilla.wikimedia.org/show_bug.cgi?id=34292 by saturday so he can fix the last remaining blocker [17:20:18] jdlrobson: will send an email [17:20:20] YuviPanda: scrolling seems fixed for me now [17:20:29] jdlrobson: yup [17:20:30] in 4.3 \o/ [17:20:34] :D [17:20:39] on the simulator, that is :) [17:20:41] glad that was simple-ish [17:20:43] true [17:20:55] will we see awjr today? [17:20:57] right YuviPanda anything else you need before I leave app-mindset [17:21:19] * YuviPanda looks [17:21:36] jdlrobson: looks good :) [17:21:37] for now [17:21:45] sweet [17:21:59] I'm away tomorrow BUT saturday if we have reached consensus for the footer I will merge in your changes [17:22:07] jdlrobson: i'll comment on and close that bug about testing app with beta versions [17:22:08] along with the equivalent fix for the mobile site [17:22:09] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5885 [17:22:11] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5885 [17:22:13] thanks YuviPanda [17:22:33] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5889 [17:22:36] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5889 [17:24:27] greetings all [17:24:28] right i'm off to watch some poetry [17:24:34] jdlrobson: seeya [17:24:37] hi tfinc bye tfinc :) [17:24:39] jdlrobson: just sent the email [17:24:41] jdlrobson: thanks for being awesome [17:24:51] ^ +1 [17:25:07] jdlrobson: are you able to check in later? [17:25:10] cheers philinje tfinc - let me know if anything needs fixing up saturday - really don't mind - i'll be happy to have a quiet day in coding :) [17:25:13] YuviPanda: how is the beta shaping up? [17:25:26] pretty good, actually. [17:25:27] YuviPanda: i'm eager to finish this work as our next projects are beating down our doors [17:25:31] let me bring up the blockers [17:25:33] tfinc: WLM? [17:25:36] yes [17:25:43] i'd say unlikely phil but possibly if needed... but if so it will only be for 30mins/1hr around midnight bst [17:25:44] and your time with us will be over before we know it [17:25:47] not to mention berlin [17:25:48] woohoo [17:26:03] indeed, i've less than two months. it's scary. sortof [17:26:06] jdlrobson: midnight would be most helpful, thanks for being so flexible [17:26:36] actually philinje i have an hour train journey I can also be available on.. but skype not so good on mobile internet :) [17:26:55] there are two smallish changes that would be great - in the email [17:26:56] i'll ping you if that happens [17:27:09] smallish changes are good for trains so shouldn't be a problem :-) [17:27:22] fingers crossed - thanks so much [17:27:26] right... must go! but speak to you all later [17:27:32] have fun [17:27:38] philinje: tfinc so, finishing bikeshedding on https://bugzilla.wikimedia.org/show_bug.cgi?id=34292 would be good. I'll finish https://bugzilla.wikimedia.org/show_bug.cgi?id=35398 before today, which leaves us with two blockers before a beta. [17:28:00] YuviPanda: adding some comments after talking to Erik on 34292 [17:28:08] YuviPanda: what is the tracking bug for the release? [17:28:18] tfinc: milestone, not bugs [17:28:20] right [17:28:33] philinje: also, https://bugzilla.wikimedia.org/show_bug.cgi?id=36017 [17:28:40] for the beta we're just going to blank the page, but deciding on that would be nice [17:28:43] * tfinc dislikes milestones having now used them [17:28:47] +1 [17:29:10] if bugzilla had a better way of surfacing them then they would be fine [17:29:21] conceptually their correct but in practice they are inefficient [17:29:23] as they are, I think they're not really very usable. [17:29:27] yup [17:31:03] YuviPanda: i have another idea instead of the blank page [17:32:36] philinje: i see your comments on https://bugzilla.wikimedia.org/show_bug.cgi?id=34292 … can you put in the final text [17:32:43] ok [17:34:43] MaxSem: what's up with the footer now? [17:35:17] YuviPanda: you now have a final for https://bugzilla.wikimedia.org/show_bug.cgi?id=34292 [17:35:50] changing [17:37:34] preilly, what do I need to do to repro it besides adding that HTML to Mediawiki:Copyrights ? [17:38:11] MaxSem: that is it I believe [17:38:54] nope, I don't see anything. something in LocalSettings? [17:40:06] YuviPanda: hows does https://bugzilla.wikimedia.org/show_bug.cgi?id=35513 look for the beta design launch on monday? [17:40:27] tfinc: oh, me and jon tested it today. [17:40:28] works fine [17:40:42] tested v1.0.x, 1.1.x and 3.1.2 (iOS) [17:43:02] updated bug [17:43:11] YuviPanda: how can i easily test the the new designs in the app ? [17:43:20] tfinc: i've a patch I can give you. [17:43:25] yes please [17:43:27] one second [17:43:34] YuviPanda: lets resolve the bug then [17:43:38] we can re-open if shit explodes [17:43:57] tfinc: http://hastebin.com/sajawawixe.js [17:44:09] that patch will apply on v1.1.x branches and v3.1.x-iOS branches [17:44:28] you might have to uninstall/reinstall to see changes [17:48:16] awjr, did you commit your HTMLForm tweaks? [17:49:02] MaxSem: not to core, i will probably be working on that tomorrow [17:49:14] YuviPanda: thanks [17:49:51] philinje: http://bug-attachment.wikimedia.org/attachment.cgi?id=10475 [17:49:56] does not fit in one line [17:50:26] but it's fine i think [17:51:40] looks great [17:51:49] YuviPanda: whats the best branch to build off of .. master ? [17:51:58] maybe no period at the end of the first line [17:52:09] awjr, I've got an idea: make the forms it generates work for both desktop and mobile. the main difference is that on desktop labels are to the left from their fields while on mobile they're above. if we make them positionable with CSS, we could stop rewriting Special:Userlogin, for example [17:52:11] tfinc: you want to test current state? [17:52:13] tfinc: then master [17:52:21] philinje: makes sense [17:52:27] k [17:52:34] MaxSem: what html did you put? [17:53:08] preilly,
  • test
  • [17:53:11] cordova-1.6.1.js .. fancy. i'm going to have to start using the new name [17:53:19] MaxSem: I've got
  • test
  • as well [17:53:21] MaxSem: good idea [17:53:26] tfinc: we've not updated the app. just winphone :) [17:53:35] MaxSem: and what is the output that you see? [17:54:14] MaxSem: I see, "
  • <li class="noprint">test</li>
  • ". [17:54:14] preilly, I see nothing copyright-related at all [17:54:32] MaxSem: are you in the beta or classic view? [17:54:38] awjr, but that will require forms built with divs [17:54:42] preilly, beta [17:54:49] YuviPanda: i real like hastbin's highlight colors [17:54:49] MaxSem: try it in classic view [17:55:00] MaxSem: i am making it form output configurable [17:55:08] tfinc: plus it autodetects language. a bit messy (thought mine was js when it was diff) but really good :) [17:55:11] so you can get forms back in divs, tables, just raw elements, etc [17:55:14] tfinc: it's the solarized theme. You can get it for vim too :) [17:55:15] they just need to add a color for functions called [17:57:42] anybody know if jon will be back on today? [17:59:29] awjr: mail him [17:59:33] philinje: --^ [17:59:37] k [18:00:03] philinje: awjr was wondering if jons was going to drop in again today [18:00:37] Jon said he could do a little work on the train, probably soon, and then not until midnight if something irgent [18:00:40] urgent [18:00:51] kk [18:00:58] bug 36190 is still not fixed :( [18:01:04] at least not for ICS 4.0.2 [18:02:55] awjr: do you know how to make iTerm the default shell on OSX? Spolight continues to use terminal as its default launcher and Finder gives me a "The operation can’t be completed." if i try to change the "Change all" for scripts [18:03:15] s/shell/terminal app [18:03:48] tfinc try this:http://www.tech-recipes.com/rx/2573/os_x_how_to_change_default_application_to_open_file_type/ [18:04:38] awjr: thats what i meant by in the latter part of what i said [18:04:47] ah i see [18:04:50] then no :( [18:04:52] it makes lion unhappy [18:04:58] but yes. that would be the proper way [18:05:00] tragedy [18:06:33] YuviPanda: lolz. i'll have to make one patch to your patch s/localhost/en.wiki/ :D [18:07:00] ah yes, that patch was for jon who wanted to test on his localhost, + instructions to not make it localhost but an IP :) [18:07:03] tfinc: let me give you another one [18:07:14] I'm rebuilding now [18:08:16] tfinc: http://hastebin.com/sanuqoviru.coffee [18:08:23] though for some reason it thinks it is coffeescript now :D [18:12:44] i go the main page to load but not much else [18:13:55] tfinc: ? [18:14:03] make remotes? [18:14:09] YuviPanda: i can search just fine but no article ever shows up [18:14:17] let me stash this change and see if something else is broken [18:14:21] tfinc: Terminal.app also provides the following Services for other applications: New Terminal at Folder, New Terminal Tab at Folder, Open man page in Terminal, Search man pages in Terminal [18:14:22] The association with Terminal.app is hard-coded. You need to create your own Services e.g. using Automator and/or AppleScript to replace these. [18:14:25] tfinc: OS / version? [18:14:40] YuviPanda: Android Emu 2.3 [18:14:47] hmm [18:14:50] app? [18:14:52] 1.1.1? [18:15:19] YuviPanda: yes. i'm also not seeing any menu interface text [18:15:22] just the keys [18:15:22] tfinc: what branch are you building off of? [18:15:28] master [18:15:32] looks like a js error. Did the patch apply cleanly? [18:15:42] tfinc: did you do a git submodule update --init? [18:15:55] YuviPanda: i just made the change directly [18:15:56] no [18:15:58] i didn't init [18:16:15] doing that now [18:16:28] ahh .. MF depedancy [18:16:30] tfinc: wait, that patch was for testing older versions of the app with the new beta [18:16:40] so that patch never entered the picture here. NEvermind [18:17:47] building post sub module init/pull [18:18:17] ok [18:18:38] philinje: just realized there's no mockup for what's supposed to happen when a user clicks on the 'General' section of the feedback form [18:19:36] right now, 'technical problem' will display the form and 'article feedback' displays the list of various feedback links. what is supposed to be displayed under 'general'? [18:20:04] ok, hang on [18:22:57] awjr: General - goes to info-XX@wikimedia.org (where XX refers to language code) [18:23:12] so no filtering required, pretty straightforward [18:24:12] philinje: so just clicking on the 'general' section header will just open up an email to send, rather than expand the section like we do for technical and article feedback? [18:24:28] dealing with emulator issues [18:24:59] http://www.html5rocks.com/en/mobile/responsivedesign/ [18:26:24] awjr: question: is the way Technical Problem supposed to work via form or mailto? [18:26:43] meaning in how you plan to implement it [18:26:50] philinje, it's a form [18:27:03] that posts to a wiki page [18:27:42] ok, then General should just be a mailto, since it is purely email [18:28:24] maybe we need to open something under General and show a link that says, Click here to send an email? [18:28:26] philinje would you mind swinging by my desk so i can show you what im talking about? [18:29:18] tfinc: I like the recommendation of using https://developers.google.com/closure/compiler/ [18:30:14] screw the emulator . i'm just going to test on device [18:30:28] tfinc: jon should write a blog post regarding http://www.html5rocks.com/en/mobile/responsivedesign/ [18:30:31] preilly: agreed. that would be a nice perf boost [18:30:42] preilly: with real world use cases [18:30:48] tfinc: yeah [18:33:44] YuviPanda|afk: articles expand but they *dont* collapse on an HTC incredible running Android 2.2 [18:34:14] tfinc: other things work? [18:35:09] YuviPanda: i'm also a bit confused. i built off of master (no patch) and i'm picking up the new expand/collapse arrow. is that because your styling them independently ? [18:35:29] the patch was never for master [18:35:36] tfinc: what exactly are you trying to test? [18:35:55] tfinc: 1. app's current state 2. Older apps' state when we make beta of MF default? [18:36:06] #2 [18:36:13] so, master is the wrong branch fo rhtat [18:36:15] *for that [18:36:21] what branch should i use then [18:36:34] v1.1.1 for android market release [18:36:37] k [18:36:38] v3.1.2-iOS for iOS market release [18:36:45] http://hastebin.com/sanuqoviru.coffee [18:36:47] is the patch [18:36:49] that works on both of them [18:37:24] New patchset: MaxSem; "Don't escape copyright message, it's already in HTML" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5902 [18:37:29] preilly, ^^^ [18:37:30] v1.1.1-Android [18:37:34] right [18:37:35] sorry. [18:38:06] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5902 [18:38:08] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5902 [18:41:07] MaxSem: done [18:44:06] YuviPanda: i'm seeing some issues on both Android 2.3.6 and Android 2.3.4. 1) the spinner doesn't go away 2) articles only show unexpanded [18:44:15] this is on a fresh install [18:44:26] tfinc: 1. did the patch apply cleanly 2. did you do make? [18:48:01] tfinc: ^ [18:51:56] one sec [18:53:52] YuviPanda: Android-v1.1.1 has no sub modules right? [18:54:01] tfinc: it has a makefile [18:54:32] android v1.1.1 has a makefile, android v1.0 has nothing, iOS v3.1.2 has nothing (iOS v3.1.1 and before have a makefile) [19:05:55] YuviPanda: ok i've rebuilt. made the patch manually as git am was not working. log cat shows that were accessing the beta but the app shows the old style show/hide. is that expected ? [19:06:08] tfinc: yup [19:06:23] great. old app stays exactly the same. new version picks up all the awesome [19:06:31] yup, that's the idea :) [19:06:43] all is well … nothing to see here [19:06:58] it does have one minor change that both me and jdlrobson are completely okay with [19:07:06] and one that you haven't spotted either [19:07:07] YuviPanda: can you add the those points about builds onto our build page … if there their then i just need to read it next time [19:07:15] ORLY? [19:07:21] YA RLY [19:07:34] do tell [19:07:49] tfinc: so, current README is up to date with current instructions. READMEs in appropriate branches should have appropriate instructions for them. I can, however, make a wiki page [19:07:54] tfinc: no footer. [19:08:28] tfinc: if you uninstall app, clear patch and re-install, you'll find that there is a sortof uglyish footer turning up. Doesn't turn up with beta [19:09:40] YuviPanda: i see. i wonder how legal will take tht [19:09:50] lunch time upstairs [19:13:11] [WikipediaMobile] yuvipanda pushed 1 new commit to master: http://git.io/QwzsJw [19:13:11] [WikipediaMobile/master] Revert "add togglefix class" - YuviPanda [19:19:07] New patchset: L10n-bot; "Localisation updates from http://translatewiki.net." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5921 [19:21:50] YuviPanda: did you see preillys comments about https://developers.google.com/closure/compiler/ ? [19:22:01] where? [19:22:25] on the response to the new scientist doc? I don't recall seeing it there... [19:22:28] might've missed it? [19:23:13] he mentioned it earlier as a way of speeding up our js [19:24:32] tfinc: so, we tried YUI compresser last time and it didn't really give us much of a speedup. [19:24:48] will try closure compiler once the beta is out, I suppose. [19:24:56] yeah, not a priority now [19:25:01] just putting it out there for discussion [19:26:29] definitely doable. [19:26:49] problem with performance is we should measuer and see where we're being hit [19:26:52] and then take steps [19:26:59] rather than blindly fix things one by one. [19:27:09] we will not be doing any premature optimization [19:27:12] thats a disaster [19:27:15] profile [19:27:15] indeed [19:31:33] tfinc, preilly: I have flights booked, going to be in SF 13-20 May. [19:31:47] jerith: fun! [19:31:54] The 14th is full, but I'm free the rest of the rest of the week. [19:32:02] come camp out [19:32:09] I'll definitely come visit. [19:32:38] If we talk about SMS/USSD at all, I can even claim a work day. ;-) [19:32:49] we can make this happen [19:34:17] More seriously, we can probably do some work on that stuff and get some of the remaining issues clobbered. [19:35:06] (And I need to buy MaxSem a beer (or equivalent) for his work on the vumi app.) [19:35:22] jerith: MaxSem is remote. he's not in SF [19:35:41] your closer to him then we are [19:35:45] tfinc: Aww, that's unfortunate. [19:36:42] It explains why he's active earlier in the day than the rest of you, though. [19:37:40] jerith: yes, he's in Russia [19:38:02] he's also active later as well. [19:38:09] he's active all the time [19:38:13] Ah. That would need a different visa altogether. [19:39:11] I suppose I'll also have to upgrade the beer to vodka. :-) [19:45:43] w00t, toggling code is broken now [19:51:01] YuviPanda: on master/ [19:51:02] ? [19:52:12] tfinc: yes, they are. New CSS we pulled in from MobileFrontend seems to have broken it. looking. [19:54:07] right. have it fixed. [19:56:58] k [20:23:43] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5921 [20:23:46] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5921 [20:42:44] tfinc: help brion [21:06:13] tfinc: http://blog.wikimedia.org/2012/04/26/analyzing-mobile-browser-energy-consumption/ is live now [21:06:29] cool [21:37:10] New patchset: awjrichards; "* Makes user-facing text rely on i18n messages rather than hardcoded value * provides configurable links for the various links that appear on the contact us form" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [21:46:07] New review: Siebrand; "Should those headers use Title Case?" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: 0; - https://gerrit.wikimedia.org/r/5931 [22:04:55] New patchset: MaxSem; "Got rid of remaining static variables" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5934 [22:09:40] New patchset: MaxSem; "Reorganized extension setup file, separating initialization code from configuration" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5935 [22:17:24] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5934 [22:17:27] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5934 [22:17:58] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5935 [22:18:00] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5935 [22:23:41] New patchset: MaxSem; "Move skins to separate files" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5936 [22:24:32] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5936 [22:24:35] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5936 [22:36:11] is there a MW way to do something like wfMsg() but to pull back a message in a language other than what the user is using? (eg i have uselang=es but i want to force an en message to be displayed) [22:50:12] New patchset: MaxSem; "Made noindexing disableable with a configuration var" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5937 [22:50:49] awjr, wfMessage('foo')->inLanguage('sq')->text() [22:51:03] or inContentLanguage() [22:52:26] cool thanks MaxSem - Reedy also suggested wfMsgExt() [22:52:52] Sam's so conservative... [22:52:57] lol [22:53:55] MaxSem: $oldLang = $wgLang; $wgLang='en'; wfMsg( 'foobar' ); $wgLang=$oldLang; [22:54:45] * MaxSem returns to his den to shock Reedy with a solution involving 50 lines of MessageCache manipulation [22:54:59] Hmm [22:55:13] Read MessagesEn.php into memory, run a regex over the string [23:23:43] New patchset: awjrichards; "* Makes user-facing text rely on i18n messages rather than hardcoded value * provides configurable links for the various links that appear on the contact us form * exposes a hook for alternative configuration handling of the aforementioned feedback form l" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:24:27] preilly if you've got a second i could use your opinion on something a little funk i did in ^ [23:24:36] s/funk/funky [23:25:33] New review: Brion VIBBER; "Comment on the toggle sounds backwards..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/5937 [23:26:30] ha! pastie.org is offline due to getting ddos'd :( [23:27:01] * tfinc responds to the OTRS queue [23:28:00] * MaxSem gets his bag of nitpicks [23:28:58] Thanks Tomasz. [23:30:07] New patchset: awjrichards; "* Makes user-facing text rely on i18n messages rather than hardcoded value * provides configurable links for the various links that appear on the contact us form * exposes a hook for alternative configuration handling of the aforementioned feedback form l" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:33:52] New review: awjrichards; "I'm particularly curious about people's thoughts on specials/MobileFeedback.php line 31. I put in a ..." [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 0 C: 0; - https://gerrit.wikimedia.org/r/5931 [23:34:44] preilly, MaxSem ^ [23:35:13] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:38:55] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) C: 0; - https://gerrit.wikimedia.org/r/5937 [23:41:07] New patchset: MaxSem; "Fix comment from 8074a228045cc787a9a2927a70ca5f3c0c109288 / gerrit 5937" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5938 [23:41:30] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:41:46] MaxSem: "This uses implicit conversion from Message to string." - is that really a bad thing? [23:43:20] New patchset: awjrichards; "* Makes user-facing text rely on i18n messages rather than hardcoded value * provides configurable links for the various links that appear on the contact us form * exposes a hook for alternative configuration handling of the aforementioned feedback form l" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:43:32] awjr, the problem with it is that you have to remeber in what mode does that implicit conversion work. in this particular case, it's 'parse', which is ok, but in other cases a different mode may be needed, so this habit is bad [23:43:50] MaxSem i see what you mean. [23:46:48] wait, MaxSem wfMsg() returns an actual string, not a message object [23:49:11] awjr, you're using $this->msg() [23:51:18] MaxSem: good point :p [23:55:55] New patchset: awjrichards; "* Makes user-facing text rely on i18n messages rather than hardcoded value * provides configurable links for the various links that appear on the contact us form * exposes a hook for alternative configuration handling of the aforementioned feedback form l" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5931 [23:56:01] MaxSem ^ [23:56:54] I'm particularly curious about people's thoughts on specials/MobileFeedback.php line 31 (in https://gerrit.wikimedia.org/r/#change,5931). I put in a hook to allow the opportunity for someone to configure $wgMFFeedbackLinks in some kind of complicated, potentially expensive way ONLY when getFeedbackHtml() runs. I'm not sure this is the best approach, but I wanted a simple way to be able to do more variable/fancy configurations like what we'll ne [23:57:00] preilly, MaxSem ^ [23:57:33] awjr, text() is not HTML-escaped [23:57:55] as well as wfMsg() [23:59:29] I still recommend you a more generalised approach to feedback topics as they're currently WMF-centric [23:59:42] * MaxSem looks at the clock [23:59:48] gotta sleep! [23:59:54] g'night