[00:02:23] awjr: so, that is ten classes right? [00:07:40] preilly yeah that looks right [00:09:20] preilly: yeah, ten classes looks right, which in and of itself is not that big of a deal, it just feels like we're replacing one possible custom solution with another [00:10:35] preilly: i can go ahead and just do it for the fields that we care about for now, but we'll have to be mindful if we ever want to use other fields [00:12:04] awjr: or just write a method that strips tables and replaces them with divs [00:12:09] awjr: heh heh [00:12:15] so… ugly [00:12:39] i think what it boils down to is HTMLForm offends my sensibilities [00:12:45] awjr: nothing, "]*>|]*>|]*>|]*>|]*>" can't fix [00:12:51] lol [00:13:16] i just died a little inside [00:14:21] awjr: ha ha ha [00:14:34] awjr: you could always make HTMLForm2 [00:14:38] * preilly hides  [00:14:40] lol [00:18:24] awjr: you know I think HTMLForm should support a decorator pattern [00:18:42] preilly: go for it! [00:19:30] awjr: I actually am curious now what preg_replace( '/\<[\/]?(table|tr|td)([^\>]*)\>/i', '', $htmlForm ); would look like [00:19:35] s/I/I'm [00:20:03] s/am// [00:27:31] awjr: can you respond to https://ticket.wikimedia.org/otrs/index.pl?Action=AgentTicketZoom&TicketID=6496828 [00:28:27] is there a nifty interwiki for mobile? [00:29:50] Amgine: what do you mean? [00:29:59] Amgine: interwiki links? [00:30:20] like mwp: for m.wikipedia.org [00:30:32] [[mwp:Main Page]] [00:30:39] Amgine: nope [00:30:46] darn. [00:31:59] preilly: i've already responded to that ticket [00:32:13] tfinc: asked you NOT me [00:32:14] see correspondence #6 and #7 [00:32:17] awjr: ^^ [00:32:44] doh tfinc ^ [00:39:41] haha [00:39:43] ok [00:40:00] awjr: you already responded to their response ? [00:40:06] tfinc yes [00:40:12] see responses 6 and 7 [00:40:35] OTRS UI FTFAIL [00:40:36] yes [00:40:37] mybad [00:40:49] hehehe [00:40:51] no worries [00:45:13] * awjr stabs HTMLForm in the face [00:45:29] this is turning into a real hackjob. [00:45:31] tfinc: ready for it [00:45:46] * tfinc worries a bit when he hears that [00:45:47] tfinc: feast your eyes on this: http://i.imgur.com/l87nA.jpg [00:46:01] tfinc: Help me 2Pac, you're my only hope. [00:46:09] only a matter of time before we saw that [00:46:24] ahahaha [00:47:15] why would you want a protected static function? [00:47:28] useless. [00:47:32] tfinc: I presume it's best not for me to respond to emails necessarily on the mobile-feedback-l? I'll do bug reports but emails to customers (I imagine you could call them that) from "Thehelpfulone" would be awkward I imagine (and I'm not official @wikimedia.org) [00:47:42] awjr: same reason you'd want this alarm clock http://www.modernman.com/wake-up-defusable-bomb-alarm-clock/ [00:47:59] roflmaobbq [00:48:09] preilly where do you come up with this shit [00:48:26] preilly: that's going on the wedding registry [00:50:28] Thehelpfulone: lets hold off on the emails for now [00:50:49] Thehelpfulone: i imagine that changing after we see the kinds of emails that come in [00:51:06] yeah sure, that's what I was thinking [00:51:15] I can just do the bug reports (if you don't beat me to them!) [00:51:26] Thehelpfulone: the race is on! [00:51:37] heh [00:51:56] Thehelpfulone: you should be worried if i ever get a cli to bugzilla [00:53:08] you're a bugzilla admin, don't you already have that? [00:53:20] tfinc: http://www.bugzilla.org/docs/2.16/html/cmdline.html [00:53:22] although bugzilla is probably buggy in itself :P [00:53:28] heh I thought so [00:53:45] i can only imagine the horrors of what that holds [00:54:01] tfinc: are you scared of a little sed -e 's/,/ /g' | wc | awk '{printf $2 "\n"}' [00:55:13] haha .. now if you could just make it so that i never have to leave vim .. i'm in [00:55:33] tfinc: have you looked at: http://www.liquidx.net/pybugz/ [00:56:01] * tfinc has been doing python coding lately and likes this [00:56:14] plus PuBugz .. how could you go wrong with a name like that [00:56:30] and its in pip [00:56:41] oh wait [00:56:42] nm [00:58:12] tfinc: https://answers.launchpad.net/pybugz/+packages [00:58:51] i'd have to spin up a virt to use those [00:59:43] tfinc: have you seen: http://www.theoldmonk.net/gitzilla/ [01:03:38] tfinc: https://github.com/williamh/pybugz [01:07:37] New patchset: awjrichards; "Merged previous contact us form with MaxSem's new special page magic using HTMLForm." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5217 [01:09:42] New patchset: awjrichards; "Merged previous contact us form with MaxSem's new special page magic using HTMLForm." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5217 [01:10:16] preilly ^ [01:11:40] awjr: preilly just stepped out for the day [01:11:42] i'm taking off too [06:30:29] hey YuviPanda [06:30:34] flyingclimber: heya [06:30:46] I've been tailing the reviews [06:31:52] I see one in UK store about crashing :( [06:31:59] checking US now [06:32:04] positive reviews in the US store [06:32:53] Hi, I'm fighting with MobileFrontend extension at http://espiral.org [06:33:18] I'd appreciate if you could check with your mobile devices whether you see a desktop homepage or a mobile one [06:33:54] I only get desktop, unless I force mobile manually - then again I have only a Nokia N9 and a Kindle Fire at hand right now... [06:34:13] and maybe those devices are not recognized as mobiles by the extension [06:36:00] flyingclimber: would you mind checking http://espiral.org in your mobile device, please? [06:36:07] hey qgil_ [06:36:08] fu [06:36:10] sure* [06:36:15] happy to help [06:36:18] thanks! [06:36:43] * flyingclimber loads up the simulator first [06:36:54] ah! simulators! [06:37:06] gosh, it's so late that I had forgotten them ;) [06:37:35] flyingclimber: we still have a lot of work left. I'm unable to replicate the 5.1 scrolling issues [06:38:02] qgil_: i'm seeing the desktop view [06:38:10] YuviPanda: i don't see them on 5.0 either [06:38:12] yep, looks like that [06:39:15] I'm kind of lost, then. The Extension page doesn't really help if the WURFL stuff is not recognized [06:39:21] and I'm not a sysadmin [06:39:27] YuviPanda: do we have anyone who can reliability show 5.x as breaking ? [06:40:07] flyingclimber: no. brion found it intermittently slow but was not able to reproduce it [06:40:13] I'll leave a message in the discussion page. Thank you for your help flyingclimber [06:40:19] qgil_: thanks! [06:41:14] YuviPanda: is it a subset of iOS devices ? [06:41:21] YuviPanda: or is it all of them? [06:41:43] flyingclimber: the reports have been on brion's iPod touch and on the store. [06:44:19] flyingclimber: so no, no pinpointing information [06:44:36] lifeeth: dealt with Indian Customs on phone imports? [06:44:40] bribey or workable? [06:45:02] YuviPanda : If it is through DHL/Fedex/UPS it gets through without an issue [06:45:14] if you are going with post -- highly unlikely that you will get it [06:45:30] yeah, post just lost my sim card and it was coming from inside india :D [06:45:34] YuviPanda : I mean you will have to pay the customs fee though [06:45:37] ah! :) [07:14:47] YuviPanda: that'll make things significantly harder . i bet we can get someone from twitter [07:16:16] we so far haven't found anyone on twitter reporting this. I'm wondering if I should tweet out asking for it [07:16:48] YuviPanda: do it [07:17:02] yup, figuring out correct way to phrase this [07:25:19] YuviPanda: nice tweet [07:25:45] :) [07:26:00] flyingclimber: progress on QA front? I remember reading that awjr is doing another round [07:26:17] yeah, let me pm you [07:26:22] ok [07:39:26] https://en.wikipedia.org/wiki/Wikipedia:Mobile_Extension_Feedback#Saved_pages [07:41:54] OuKB: sweet. [07:42:41] have contacted RIL (Pocket) about their iOS SDK. looking into saved pages [10:12:07] New patchset: Reedy; "Pass RequestContext" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5222 [10:13:36] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) C: 1; - https://gerrit.wikimedia.org/r/5222 [10:16:08] New patchset: Reedy; "Pass RequestContext" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5222 [10:17:06] I don't see the difference [10:17:13] A much longer commit summary [10:18:03] was it really needed? :P [10:18:16] Yup :D [10:18:17] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) C: 1; - https://gerrit.wikimedia.org/r/5222 [10:19:01] New review: Reedy; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5222 [10:19:03] Change merged: Reedy; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5222 [10:21:29] MaxSem: have you tried https://gerrit.wikimedia.org/r/#change,5217 ? I'm getting errors when I try to access it... [10:21:34] New review: Hashar; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5222 [10:21:58] nope [10:22:45] I'm getting Descriptor with no class: Array - which seems to be a problem in getForm caused by use of custom input types [10:24:41] New review: MaxSem; "(no comment)" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5222 [10:40:02] New review: Jdlrobson; "This doesn't work for me - something doesn't seem to be picking up your changes to $typeMappings" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) C: -1; - https://gerrit.wikimedia.org/r/5217 [10:54:07] YuviPanda: fixed the content flash problem [10:54:13] jdlrobson: w00 [10:54:14] woo [10:54:15] i mean [10:54:16] woot [10:54:19] -webkit-transform again [10:54:23] ugh [10:54:31] now looking into the scrolling problem [10:54:33] jdlrobson: removing that translate3D messes things up on iOS [10:54:34] (horizontal) [10:54:38] ok [10:54:41] its only applied to android devices YuviPanda [10:54:51] it now is, I guess :) [10:55:05] https://github.com/yuvipanda/WikipediaMobile/pull/3 [10:56:21] jdlrobson: https://github.com/jdlrobson/WikipediaMobile/commit/f9367652c4f75f46eebdaa6c6e45063b28da1903 [10:56:57] yup [10:57:08] jdlrobson: so, i'm considering getting rid of line 68 and 73 on app.css [10:57:15] and just adding those translates to iOS [10:57:23] instead of setting them in general and removing them in android [10:57:33] we could also do that yes [10:57:43] seems cleaner to me [10:57:43] that might be more readable [10:57:46] yes [10:57:49] yep agreed [10:57:57] let me go do that [10:59:12] testing [12:07:06] YuviPanda: think ive got the horizontal scroll problem [12:07:27] viewport issues or more css issues? [12:07:31] css [12:10:16] jdlrobson: also, my fix (of applying it only to iOS) doesn't seem to be working [12:10:17] investigaitng [12:13:14] pushes.. [12:15:38] at least this one made sense [12:16:43] ok im noticing that my screen flash problem only fixed the initial page load [12:16:54] it doesn't seem to fix the screen flash after searching or clicking page links [12:16:56] grr [12:22:15] it's more css problems... [12:23:27] ugh [12:23:32] not webkit transform [12:28:20] not sure yet still trying to find the troublesome css by divide and conquer [12:29:36] that seems to be the only viable debuggins strategy :| [12:29:53] jdlrobson: is there something about the nature of the app's structure that is causing these kind of 'fix only by divide and conquer' bugs? [12:30:01] they're incredibly irritating [12:30:14] well the thing i'm slowly learning is vendor prefixes are causing problems [12:30:39] and we have badly documented hacks that might be there for a specific device but it is not clear [12:31:36] (the .android .ios classes should help with these) [12:31:45] :| [12:31:49] true [12:36:03] seems to be related to scroller surprise surprise [12:36:52] i'm scared everytime we touch anything scroller because maybe it'll break iscroll [12:37:15] iScroll is just for iphone ? [12:37:24] (ios) [12:39:51] yeah [12:39:52] 4.x [12:42:23] grr so yeh its iscroll css causing problems in ics here [12:42:57] so ios targeting would help here [12:43:07] just trying to work out the rule in particular.. [12:43:19] then shall we merge? [12:43:22] (to master?) [12:44:57] yeah [12:45:16] i've narrowed it down to 1 or 2 lines [12:45:22] * 1 of 2 [12:45:58] -webkit-overflow-scrolling: touch; :/ [12:47:22] :/ [12:47:42] jdlrobson: that's something we can move to iOS only [12:47:46] IIRC not supported in Android [12:48:02] * YuviPanda checks [12:48:20] it's causing the screen flashes though [12:49:08] pushed [12:49:12] pulling [12:49:30] https://github.com/jdlrobson/WikipediaMobile/commit/d4e2605a7ab27da32d242b85bde6a2f9a2af995b [12:49:35] fun times [12:49:39] pulling and merging i hope? [12:49:44] we then need to tackle the homepage [12:49:56] unless there are any outstanding large bugs [12:53:08] btw YuviPanda there seems to be a lot of MobileFrontend bugs on https://bugzilla.wikimedia.org/buglist.cgi?field0-0-0=priority&resolution=---&resolution=LATER&resolution=DUPLICATE&query_format=advanced&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&type0-0-0=notequals&value0-0-0=enhancement&product=Wikimedia%20Mobile&list_id=108591 [12:53:12] jdlrobson: checking. [12:53:19] (listed as apps in the bug) [12:53:37] jdlrobson: you're looking for Product: Wikipedia App [12:53:39] not Wikipedia Mobile [12:53:41] we re-orged it [12:54:47] ahh ok ... so am I using the wrong filter for MobileFrontend? [12:54:51] I tend to use extensions/MobileFrontend [12:54:57] e.g. this url https://bugzilla.wikimedia.org/buglist.cgi?query_format=advanced&field0-0-0=bug_severity&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&type0-0-0=notequals&value0-0-0=enhancement&component=MobileFrontend&resolution=---&resolution=LATER&resolution=DUPLICATE&product=MediaWiki%20extensions&list_id=108593 [12:55:06] jdlrobson: that's where it should be. [12:55:22] IIRC tfinc re-orged it so that 'Wikipedia Mobile' is 'obsolete' [12:55:38] someone needs to sit for a day, triage all the 'Mobile' ones and move them appropriately [12:56:03] ok cool - just checking - i find people are opening bugs under Wikipedia Mobile - which I guess is what I'd do if I wasn't familiar with the codebase and that the mobile site is actually an extension [12:56:13] true, should do that soon [12:56:19] right i was going to pop out for some air [12:56:20] but tfinc is the best person to poke about this [12:56:23] anything you want me to do before that? [12:57:04] jdlrobson: so, android 4.x [12:57:11] article doesn't scroll to top after nav :D [12:57:32] let me pull that in and go check that [12:57:54] jdlrobson: is there anything we could do to prevent these bugs from occuring in the first place? [12:58:00] (weird css ones) [12:58:04] app's architecture, etc? [12:59:51] jdlrobson: https://github.com/jdlrobson/WikipediaMobile/commit/4ba192637355111768cca1f7911cad6abb9f9436 and https://github.com/jdlrobson/WikipediaMobile/commit/d4e2605a7ab27da32d242b85bde6a2f9a2af995b only? [13:00:52] only? [13:01:32] nevermind I can't seem to read [13:01:34] well.. by targeting hacks like this at specific browsers i'd hope these bugs would stop appearing [13:02:36] i'm not even sure how effective -webkit-overflow-scrolling: touch; is but luckily the original commit https://github.com/jdlrobson/WikipediaMobile/commit/f130a3e3c823309d3e5d8cc57d202e67f3631fce did explain it was for ios [13:03:13] we're going to have to be vigilant about pull requests including hacks in future to avoid bugs going into the code base [13:03:36] we'll have to ensure hacks target the right browsers [13:03:48] jdlrobson: right. makes sense. [13:04:18] anyway... air and errands. Be back online in about 30mins-1hr [14:05:26] jdlrobson: I still see the flash of no content until I scroll issue on ICS [14:05:30] even after your fixes [14:05:30] :( [14:05:35] have you turned it off and on again? [14:05:40] good point [14:05:42] let me try that [14:05:46] :) [14:05:50] also, how're gun laws in the UK? [14:05:58] might be easier than this. [14:06:03] http://www.youtube.com/watch?v=nn2FB1P_Mn8 [14:16:55] any luck YuviPanda ? [14:17:06] jdlrobson: nope. [14:17:24] and still flashes after turning it off and on again? [14:17:35] yes [14:17:36] it does [14:17:43] and the window.scrollTo hack does not work [14:17:50] grr wtf is with these emulators [14:18:12] is this still all on the api branch? [14:19:32] jdlrobson: yes [14:19:36] i pushed after picking up your changes [14:19:56] ill verify in my emulator [14:20:16] ok [14:22:56] mm so on my emulator i cant seem to load any page [14:23:43] has anything else changed? [14:25:37] jdlrobson: can't? [14:25:38] no [14:25:38] ... [14:26:01] it might be the wifi connection in my cafe - but nothing ever loads in the emulator [14:26:18] let me check ios [14:26:26] i just see the ajax loader [14:26:28] jdlrobson: ok [14:26:32] ios is fine.. [14:26:33] mm [14:27:05] jdlrobson: try scrolling? [14:27:16] nope [14:27:27] works fine on device :| [14:27:31] jdlrobson: what commit are you on? [14:27:43] are you on a fresh checkout of api-move? [14:28:00] well i was on 0e0ce9b9fab501ba4c44e2d5730dc0b077df4efb and did a git merge [14:28:07] i'll try again with a fresh clean one [14:28:40] merge and cherry-pick are not friends :D [14:29:58] ok brb food [14:31:47] still not loading.. [14:36:34] ERROR!{"readyState":0,"status":0,"statusText":"abort"} at file:///android_asset/www/js/search.js:96 < assuming this is my connection so moving somewhere more reliable [15:03:05] New patchset: MaxSem; "Restore message lost in merges and refactorings" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5228 [15:13:26] New review: Siebrand; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5228 [15:13:29] Change merged: Siebrand; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5228 [15:22:23] so YuviPanda - on a better internet connection i can see content again and bizarrely the white flashes have returned which is most confusing [15:26:17] bizarrely i seem to have fixed it as well [15:47:36] jdlrobson: bizzare css bugs? You don't say [15:47:45] :D [15:47:45] jdlrobson: yes, readyState '0' is a connection bug. [15:47:55] I should poke heatherw again about that icon [15:47:55] does that commit fix ICS for you? [15:48:09] jdlrobson: anything now? [15:48:11] i'm just checking [15:48:15] came back after lunch [15:48:17] I mean dinner [15:49:06] jdlrobson: where? I don't see anything at https://github.com/yuvipanda/WikipediaMobile/pull/3 [15:49:43] right [15:49:45] i see another one [15:51:01] trying [15:52:21] hi YuviPanda [15:52:29] heya heatherw [15:52:43] did you make a screenshot of where that happens? [15:53:01] heatherw: it doesn't happen yet :D [15:53:06] but wait I can explain with a screenshot [15:53:31] this: "the spinner just keeps doing its thing." YuviPanda [15:53:40] yeah, I'm guessing that wasn't very clear [15:53:46] looking for a screenshot to explain, wait a minute [15:53:47] :D [15:54:41] heatherw: https://p.twimg.com/Aqu9gXsCEAArzm8.jpg [15:54:57] heatherw: so, essentially, when the network is down [15:55:06] heatherw: the spinner just keeps spinning until the network is back [15:55:15] heatherw: I want to replace it with some other icon that says 'network is down' [15:55:18] that shot has a few problems ;P [15:55:21] rather than just the generic spinner [15:55:24] right [15:55:29] i was picking for *any* shot :D [15:55:44] heatherw: i could ask you to 'forget about everything but the spinner' :D [15:55:56] already forgotten YuviPanda [15:55:56] or I could make you another shot that doesn't have that [15:56:00] heatherw: :) [15:56:15] heatherw: am I clearer now or still phrasing things badly? [15:56:37] it made sense, i just needed to see the context (size, etc) [15:56:50] thanks! YuviPanda [15:56:55] heatherw: :) [15:57:30] heatherw YuviPanda - something like this :D http://3.bp.blogspot.com/-agtpn2Q43bk/TxzTkYjzD-I/AAAAAAAAFu4/2va8paYaG6w/s1600/2510442.jpg [15:58:07] jdlrobson: http://rlv.zcache.com/no_network_connection_available_flyer-p244075390443863723b2pv5_400.jpg [15:58:08] ? [15:58:08] :D [15:58:17] jdlrobson: lol [15:58:21] :) [15:58:35] jdlrobson: but I think this will be a pleasing one for 'no internet' http://plug-in.bestbuy.ca/t5/image/serverpage/image-id/3632iD197CF6F101B87CF/image-size/original?v=mpbl-1&px=-1 [15:58:36] :D [15:59:23] that should just be on everything [15:59:27] :D [15:59:42] jdlrobson now has no internet [15:59:54] heh [16:00:39] YuviPanda: lets chat in 15min [16:01:05] tfinc: cool [16:01:53] hey tfinc [16:03:34] jdlrobson: unintentional side effects? [16:03:42] the top bar is no longer position:fixed! [16:04:49] how did that happen? [16:05:06] the search bar you mean? [16:05:14] can you send me a screenshot? [16:05:28] jdlrobson: yes [16:05:32] it essentially scrolls with the content [16:05:40] mm [16:05:48] and that was after cherry picking my commit?! [16:06:03] yes, am recertng that to confirm [16:06:35] jdlrobson: yup [16:06:41] reverting fixes it [16:06:42] overflow-y: auto; does that!? [16:06:44] let me get you a screenie [16:07:04] wait a minute.. [16:08:05] i think i know what you mean [16:08:13] i just for some reason thought it was always like that [16:08:16] we scroll the search? [16:08:28] we *don't* scroll the search [16:08:32] it remains on top no matter what [16:08:39] and now with your commit, search scrolls [16:09:03] sorry yeh i know what you mean [16:09:13] i guess i just thought we hid search on a scroll [16:09:19] of course my change would do that [16:09:26] weird it fixes the flashing bug.. [16:10:06] (ie. content more important debate) [16:10:39] jdlrobson: https://minus.com/mLcrPZHeF/ [16:10:55] yep we are on the same wavelength [16:10:59] i'm exploring it now [16:11:13] jdlrobson: yeah, that one. I think we figured we'd eventually autohide them after a timeout, have them tap to show, etc. I don't think we even started on that debate :) [16:11:14] ok [16:11:30] mm [16:12:20] we could just use position fixed for android.. avoiding any use of scrolling divs where possible [16:16:15] ok [16:27:57] YuviPanda: thoughts? https://github.com/jdlrobson/WikipediaMobile/commit/b219e68e5bc1aefb6aa8bdd77ae35011cf01cb0f [16:28:13] it seems to fix the scrolling problem and i suspect it will obsolete scroll hack [16:50:31] jdlrobson: saw your comment about my code changes for contact-us-redesign [16:50:35] what were you doing when you got that error? [16:50:59] just tryin to access the form via the contact link [16:51:23] hmm [16:51:32] http://localhost/w/index.php?title=Special:MobileFeedback&returnto=Main+Page [16:51:43] i'm up to date on mediawiki [16:51:47] anything else i should be up to date on? [16:51:57] not sure - lemme dig into this for a sec [16:52:22] hmm i can't replicate the problem [16:52:30] out of curiosity, what version of PHP are you using? [16:53:46] 5.3.6 [16:54:20] jdlrobson: jdlrobson was on a call with tfinc [16:54:22] checking now [16:54:30] no worries guessed as much [16:54:39] New patchset: MaxSem; "Work in progress: creation of mobile skin. Footer and WML view don't work yet." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/5240 [16:54:55] I find doing 'git fetch jdlrobson' funny no matter how many times I do it. It's like having a dog named git :D [16:55:12] yehhh i find that funny too :) [16:55:43] cue mental image of a large dog dragging jdlrobson along [16:56:07] jdlrobson: also, when the beta site goes live, what's the plan on footers for non js browsers? [16:56:15] also, in general, what's the plan for non-js browsers? [16:56:25] greetings all [16:56:56] i'm working from home today so lets see if phil or preilly can host the call [16:57:01] if not i'll start it up [16:57:31] tfinc: ok [16:58:33] jdlrobson: so, it looks fine, except when I scroll down the search bar again acts a little bit weird [16:58:40] jdlrobson: there are flashes of white above the search bar [16:58:43] in what way is it weird? [16:58:45] mm [16:58:52] as if it tries to scroll down but is 'stuck' there [16:59:01] cant' replicate on mine annoyingly [16:59:03] * jdlrobson googles [16:59:04] so intermittently there's few pixels gap over there as I scroll [16:59:17] linSmith: are phil or patrick in the office ? [16:59:19] jdlrobson: that is very consistent with how 'natural' scrolling works [16:59:32] but except that it doesn't seem to scroll, just the 'bounce' type effects [16:59:52] hey preilly [17:00:01] i'm battling feeling sick so i won't be in the office [17:00:06] tfinc: okay [17:00:09] preilly: can you host the call and dial me in? [17:00:15] tfinc: sure [17:00:20] thanks [17:01:19] jdlrobson: i can't replicate the problem [17:01:26] im going to keep digging tho [17:02:10] tfinc: I'm not sure, it's my wfh day [17:02:23] awjr, you're not on skype [17:02:31] awjr: you on Skype? [17:02:36] linSmith: i just got him. no worries [17:02:40] oh. [17:02:44] opened skype and forgot to sign in [17:02:45] im on now [17:07:42] did i get through or did I just mute myself out? [17:08:27] * YuviPanda pokes brion  [17:09:15] woot [17:10:00] YuviPanda: mute [17:10:04] hey YuviPanda [17:10:08] awjr: let me know if you want to troubleshoot on my pc [17:10:10] damn. [17:10:17] (via skype screen sharing or something) [17:10:34] jdlrobson: im a dumbass [17:10:56] brion: heya! You reported slow scrolling with the app on 5.1 as well, right? Do you remember which article was causing the issue and on which device? [17:10:59] and was it consistent? [17:11:00] jdlrobson: i forgot that i had a monkeyed with the HTMLForm class to use late static bindings to get $typeMappings to work [17:11:20] preilly: damn. I was saying that that's something i'd poke brion to get info on, since I've not been able to isolate it to where and when it was happening [17:11:24] jdlrobson: which unfortunately is not a good solution since late static bindings are new in php 5.3 but MW has to support php >= 5.2 [17:11:44] awjr, so what's the problem with tables? [17:11:58] awjr: was there ever a decision reached on 5.3 for extensions? [17:12:04] jdlrobson: scrollable tables look great on the beta now [17:12:11] yeh i saw your bug [17:12:14] MaxSem: what isn't the problem with tables? [17:12:16] (bug comment) [17:12:16] YuviPanda: don't recall which article but saw it on several, intermittent. on ipod touch [17:12:22] let's see if i can find one [17:12:40] YuviPanda: more or less - basically it was that the php requirement will stay the same until someone changes it. [17:13:01] awjr, by the way - happy birthday!:) [17:13:03] MaxSem: thables are a nightmare to style [17:13:06] thanks MaxSem :) [17:13:27] yeh happy birthday awjr - skype has informed me ;-) [17:13:28] awjr: and it'll be the same for extensions and core? [17:13:38] thanks guys :) [17:14:03] wooh, happy birthday GSoC mentor :D [17:14:04] awjr: ^ [17:14:24] * tfinc hands awjr a cold beverage for his bday [17:14:28] YuviPanda, problem with 5.3 is that WMF might move to hphp that doesn't implement 5.3 fully [17:14:45] YuviPanda: iirc, the vague consensus reached was extensions running on the cluster should match the same compatibility as core [17:14:53] tfinc: thanks! it's 5 o'clock somewhere [17:15:00] and thanks, YuviPanda :) [17:15:11] MaxSem: the last I remember reading was that it was on hold until Facebook made changes to the hphp vm? [17:15:28] yeah [17:15:41] late static bindings is one of the reasons PHP drives me insane [17:15:46] but not specifically to wait for them to make it 5.3-compliant [17:16:01] rather, lack of it pre 5.3 and the general weirdness of how php handles static properties and methods [17:16:15] PHP being weird? You don't say... [17:16:27] though I'll admit it has become much less crappy now than it used to be [17:16:34] YuviPanda +1 [17:17:04] awjr: props to roan for updating https://wikitech.wikimedia.org/view/How_to_deploy_code [17:17:09] preilly: --^ [17:17:10] tfinc aye [17:18:04] tfinc i was very pleased to see that in my inbox this am [17:18:09] tfinc: yeah [17:18:33] awjr: i kept pushing them to do it. thankfully things stabilized to where they felt comfortable writing it [17:18:46] tfinc heh i was harassing roan about it too :p [17:21:23] YuviPanda: figures… now that i'm looking for it i can't repro the slow scrolling ;) [17:23:41] * jdlrobson grabbing some dinner awjr let me know when you've pushed your new contact us changes and ill finish it up [17:23:58] jdlrobson im pushing a fix now [17:24:30] brion: :( [17:25:52] New patchset: awjrichards; "Merged previous contact us form with MaxSem's new special page magic using HTMLForm." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5217 [17:26:03] MaxSem: isn't this https://gerrit.wikimedia.org/r/#change,5240 on the master branch? [17:26:03] jdlrobson ^ [17:26:48] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 0; - https://gerrit.wikimedia.org/r/5217 [17:27:29] preilly, topic in Gerrit and my local branch are called 2012/skin-rewrite [17:27:47] what's the difference between branch and topic? [17:27:56] MaxSem: yeah, but that is not a remote branch [17:28:05] MaxSem: that would be merged directly into master [17:28:35] MaxSem: in gerrit, 'branch' refers to the remote branch whereas topic refers to your local branch [17:28:37] i think [17:28:44] grrr [17:28:48] awjr: you are correct [17:29:01] so how did I manage to commit into remote master? [17:29:09] MaxSem: do you know how to make remote branches? [17:29:38] preilly: rename to Partner Engineer is done http://hire.jobvite.com/Jobvite/Job.aspx?j=oX2hWfwW&c=qSa9VfwQ [17:29:46] preilly: i found the same kind of job both at twitter and facebook [17:29:51] telling me that it made sense [17:30:11] MaxSem: https://gerrit.wikimedia.org/r/#admin,project,mediawiki/extensions/MobileFrontend,branches [17:30:21] tfinc: cool [17:30:43] MaxSem: I just made a remote branch called "skin-rewrite" at 5176335e4468347fba2ea9b3ad4a4f1ee29844a6 [17:33:40] preilly, so what shall I do? [17:33:53] MaxSem: so, now: git pull; git checkout skin-rewrite; and you should see, [17:33:53] "Branch skin-rewrite set up to track remote branch skin-rewrite from origin." [17:33:54] "Switched to a new branch 'skin-rewrite'" [17:34:16] then replace that commit [17:35:01] preilly: whats the timeline for ops taking our puppet packages and installing them in production? [17:36:15] tfinc: there is currently no timeline set [17:36:21] k [17:36:32] tfinc: probably next week when Ryan is back in the office is realistic [17:36:49] k. even though were waiting for praekelt for further work [17:36:53] i'd like us to do a blog post about it [17:36:55] to raise awareness [17:37:02] and to get people testing it [17:37:11] even if its only over jabber/gtalk now [17:38:15] tfinc: okay [17:58:17] awjr: there? [17:58:38] jdlrobson: yeppers [17:59:02] so awjr - a few issues - should be a field storing email not subject which suggests you'll need a 'HTMLTextEmailDiv' which creates an email inside a div. Also I'm not sure why do we need divs wrapping these fields anyway? why not just use email ? [17:59:14] and HTMLTextFieldDiv etc [17:59:38] also post should include the hash fragment [18:00:09] and errors should be above rather than below the thing they are pointing at [18:01:00] jdlrobson: ah i see in the design there's a field for email address. but why? aren't we just writing this data to a wikipage? [18:01:10] what are we supposed to do with the email field? [18:01:22] jdlrobson: which hash fragment are you referring to? [18:01:23] i think the idea was to be able to reply to people not familiar with wikis [18:01:29] subjects are useless any how [18:01:40] hash - http://localhost/w/index.php/Special:MobileFeedback#section_1 [18:01:41] jdlrobson: uh but… where is that email address supposed to go? [18:01:45] the #section_1 bit [18:01:49] otherwise on a post the form is hidden [18:01:55] good question awjr [18:02:15] jdlrobson the specs really need to be updated to what we discussed with tfinc a few weeks back for clarity [18:03:10] jdlrobson as i recall the last conversations we had about this, subject/message/user agent would get written to a page on mediawiki.org for technical problems. [18:03:17] if that's the case, an email address is useless for us. [18:03:34] well the simplest thing would be to drop the email [18:03:37] yes [18:03:37] and subject [18:03:55] the subject, i think, is useful, so we can create a new section on the wiki page with the heading being the subject [18:04:11] and give them a link to the wiki page/watch page so they can see replies? [18:04:33] i'm not sure subject is useful to be honest.. [18:04:44] in bugzilla for instance the subjects tend to confuse me [18:05:24] meh i dont care that much about the subject, but in my mind it would for an easy visual separation of messages on the wiki page as well as be useful for locating specific messages once that page starts filling up with 'em [18:05:25] but that aside why do we need the divs? [18:05:37] jdlrobson i dont know, maybe we don't. [18:05:48] jdlrobson do you want the fields wrapped in anything? [18:05:53] nope [18:06:10] jdlrobson ok - so we should just have the fields get plopped out, wrapped in nothing [18:06:11] the form is enough wrapping [18:06:14] k [18:06:28] just the errors should be wrapped in a div [18:06:35] jdlrobson that should not be a problem. in order to do it, i'll have to extend the HTMLForm stuff a little further [18:06:40] sweet [18:06:55] these are things HTMLForm should be able to handle anywho [18:07:27] i think i might start fixing HTMLForm on my 20% day because it's kind of not-that-useful right now [18:11:50] also awjr > https://gist.github.com/2415518 [18:13:17] YuviPanda: are you using/planning on using zim file format? [18:13:32] planning on, yes [18:14:13] kk. [18:14:35] preilly, "then replace that commit" means abandon+cherrypick+commit into skins-rewrite? [18:17:10] jdlrobson if you're otherwise ok with the changeset, i can merge it and you can add that css [18:17:31] sounds good [18:17:39] we removed the divs? [18:17:53] jdlrobson not yet but i'll do that in the next changeset [18:17:57] sweet [18:17:59] poke me when ready [18:18:11] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5217 [18:18:14] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5217 [18:18:20] jdlrobson ok changeset merged [18:18:31] go ahead and add your css and i'll work on getting rid of the divs [18:19:55] tfinc: hey what's the support email we use for apps? I need one for setting up on blackberry [18:20:02] for the playbook woo [18:20:06] brion: mobile-feedback-l [18:20:21] at wikimedia or at lists.wikimedia? [18:20:34] awjr: merged to master? [18:20:35] mobile-feedback-l@lists.wikimedia.org, [18:20:38] tx [18:20:43] jdlrobson: merged to contact-us-redesign [18:23:10] http://www.newscientist.com/blogs/onepercent/2012/04/bloated-website-code-drains-yo.html [18:23:18] jdlrobson: --^ [18:24:03] awesome comment at the end [18:26:58] interesting [18:27:18] they are right about jquery but going forward i expect us to use that more [18:27:26] but yes the comment at the end has hit the nail on the head [18:28:03] and also we are a lot better than we were now we have minified js and css :-) [18:28:28] exactly. i wanted to pass that on you incase this article gets traction [18:28:40] i doubt that it will [18:29:27] though for casual WP lookups initial page load means a lot [18:30:31] MaxSem: you mean directly from google ? [18:31:08] tfinc, or main page load+inline search+viewing 1 result [18:32:01] i bet if we had sessions tracking we'd see that our js/css is on most peoples phones as we get *a lot* of users who browse multiple times a day [18:32:10] those kinds of articles are so unhelpful- I would love if Apple/Google did standard energy usage ratings on the app store though [18:32:28] though my own use case was different: read WP whole way from work to home [18:32:50] iOS already has energy usage measurement for developers, so the capability is there in a general sense [18:33:19] Rolken: Android has the same [18:33:19] awjr: how does the form action attribute get set? [18:33:25] Rolken: you can see exactly how much each app takes up [18:33:41] jdlrobson it gets set automagically by HTMLForm, i think. but it's also overridable [18:33:45] i think [18:34:28] we need to include the hash...e.g. action="http://localhost/w/index.php/Special:MobileFeedback#mf-feedback-form" [18:34:29] jdlrobson yeah there's a 'setAction' method in HTMLForm [18:34:48] jdlrobson: ok let me get the form displaying in raw html then i'll take care of that [18:35:14] New patchset: Jdlrobson; "styling fixes" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5245 [18:47:43] New patchset: awjrichards; "Add support for selecting form display type. This now supports 'div' and 'table', raw forthcoming." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5246 [18:47:58] jdlrobson ^ [18:48:16] that changeset returns the form in raw HTML (no table/div wrappers) [18:48:30] the way errors are returned may still require some tweaking [18:48:37] im going to see about fixing the action [18:57:45] New patchset: awjrichards; "Set form action to include mf-feedback-form hash" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5247 [18:58:01] jdlrobson: ^ adds better form action [18:58:35] jdlrobson: note that that changeset is not dependent on the previous one i made, so it does not include the changes for raw-html form [18:58:48] that way they can be merged independently if needed [18:59:58] New review: Jdlrobson; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) C: 1; - https://gerrit.wikimedia.org/r/5247 [19:03:07] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5247 [19:03:09] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5247 [19:09:22] jdlrobson: i've got take off in a few minutes for ~1.5 hours or so. but if you can take a look at https://gerrit.wikimedia.org/r/5246 i can keep moving forward with the raw html form presentation and then start working on actual form handling on submit [19:09:37] i'm winding down myself [19:09:58] that patch looks fine in terms of what it does [19:10:09] alright cool [19:10:16] being no php guru i'm not sure about the approach [19:10:43] i think it's about as sane as it can be if we want to keep using HTMLForm [19:11:33] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5245 [19:11:36] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5245 [19:13:25] New patchset: awjrichards; "Add support for selecting form display type. This now supports 'div' and 'table', raw forthcoming." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5246 [19:14:22] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/5246 [19:14:24] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/5246 [19:17:40] tfinc, how should stuff like https://bugzilla.wikimedia.org/show_bug.cgi?id=35969 be categorised? it belongs to mobile site, but not MobileFrontend [19:22:34] playbook release submitted, hoping all goes well. :) our contacts will expedite review [19:23:18] lunchies time i think [19:37:23] jdlrobson: so scrolling on iOS 5 is broken [19:37:45] possibly related to -webkit-overflow-scrolling changes [19:37:47] investigatin [19:38:48] MaxSem: hmm [19:39:12] MaxSem: i'm putting it under content for now [19:39:28] jdlrobson, do we need