[00:33:57] jdlrobson: http://www.meetup.com/jsmeetup/events/68718662/ [00:33:59] also http://www.meetup.com/PhoneGap-SF/events/65366872/ [00:41:48] this jumpy screen is really bugging me now [15:47:23] hey shiny jdlrobson-shinym [15:47:37] hello! [15:48:04] at the office already? [15:50:01] jdlrobson-shinym: plus it looks like I was an idiot and did not push to the wikimedia repo yesterday. You'd need to rebase :( [15:50:19] oh dear [15:50:19] no not at the office yet [15:50:31] still hunting for houses [15:51:09] for some reason I read that as 'haunting for houses' [15:54:16] it's not going well :( [15:54:39] jdlrobson-shinym: :( [15:54:54] I think I found one but now I'm Facebook stalking her and not so sure [15:55:04] (her being the room mate) [15:56:07] jdlrobson-shinym: oooh, not moving in by yourself? [15:56:14] noo I want to meet people [15:56:36] aaaand I just internally realized how much of a taboo there is against having a room-mate of the other gender that it had not even crossed my mind :P [15:59:17] right i should head to the office [16:08:28] whee https://toolserver.org/~erfgoed/api/api.php?action=search&srname~+cas* [16:08:49] yuvipanda, please test^ [16:09:19] MaxSem: sweet! will do [16:09:47] in a different WLM related meeting at the moment, will do when done [16:11:03] eh, https://toolserver.org/~erfgoed/api/api.php?action=search&srname=~+cas* [16:26:49] hey preilly! was https://bugzilla.wikimedia.org/show_bug.cgi?id=37761 on iOS 6? [16:49:17] yuvipanda: can MobileSafari be run on android? [16:49:24] jdlrobson-shinym: nope [16:49:28] Mozilla/5.0 (Linux; U; Android 3.2; zh-cn; GT-P6800 Build/HTJ85B) UC AppleWebKit/534.31 (KHTML, like Gecko) Mobile Safari/534.31 [16:49:36] that seems like a weird user agent no? [16:49:51] I thought Mobile Safari was only in the user agent for iphone? [16:49:54] to... say the least [16:49:55] yes [16:50:02] but user agents, you never know [16:50:32] jdlrobson-shinym: apparently quite common?! [16:50:50] jdlrobson-shinym: http://stackoverflow.com/questions/5341637/how-do-detect-android-tablets-in-general-useragent [16:50:56] jdlrobson-shinym: http://www.gtrifonov.com/2011/04/15/google-android-user-agent-strings-2/ [16:51:00] ignore former [16:52:36] jdlrobson-shinym: so it says mobile safari, but doesn't really mean it [16:52:45] YEH [16:52:52] sorry didn't mean capitals there [16:53:25] OMG CAPITALS! [16:53:26] https://xkcd.com/499/ [17:06:21] yo yuvipanda [17:06:37] do you remember the problem where the layout shifted down the screen on navigating between screens? [17:06:46] I'm seeing that in WLM now and trying to work out what causes it [17:06:57] jdlrobson-shinym: by one pixel? [17:07:05] yeh something like that [17:07:13] did you ever work out the cause? [17:07:56] jdlrobson-shinym: I think they were rendering glitches caused by 'fractional' pixels, but nothing definitive [17:07:59] seems to be input focus related [17:08:16] or soft keyboard [17:08:26] okay, we might be talking about different things? [17:08:53] maybe [17:09:09] jdlrobson-shinym: on the app sometimes everything shifted down by at most one pixel. Are you referring to the same thing? [17:09:51] jdlrobson-shinym: also, I'm unsure but I think you'd need to rebase the patch? I got rid of #show-map, for example - but your diff still has it [17:10:01] mm [17:10:20] rebase to yuvipanda or rebase to master? [17:10:24] jdlrobson-shinym: master? [17:10:25] as I've been working of yuvipanda master [17:10:52] jdlrobson-shinym: let's swtich to wikimedia master :) [17:11:12] jdlrobson-shinym: in fact I think we should switch to gerrit - and start sending this for i18n [17:12:32] jdlrobson-shinym: also, I'm wondering if we should be just binding variables inside anonymous functions to keep state vs using module-level variables [17:12:50] jdlrobson-shinym: example being https://github.com/jdlrobson/WLMMobile/commit/f677999bb56d35391bde4a50ea08e41715eada9d [17:13:26] you think we should be doing what above does or something else? [17:13:52] jdlrobson-shinym: I think we should minimize module level variables [17:14:16] so bind things locally as much as possible [17:14:19] so concretely you think we shouldn't be doing what that commit does? [17:14:24] yes [17:14:24] yuvipanda: I'm working on your WLM requests today [17:14:36] linSmith: Thanks! :) [17:15:19] why yuvipanda [17:15:53] jdlrobson-shinym: because it is easier to see the flow of data? [17:17:28] really? [17:18:27] on a slightly different subject - the bug I've been experiencing of the pixels at the top of the page is due to display: box [17:18:33] is there any reason we use that over block? [17:19:52] jdlrobson-shinym: display: box on which one? [17:20:00] jdlrobson-shinym: on the static 'header'? [17:20:01] the .actiobar [17:20:03] yeh [17:20:08] jdlrobson-shinym: yes, it saves a considerable amount of headache in layout [17:20:14] how so? [17:20:15] and is much cleaner (IMO) than float based layouts [17:20:44] i particularly like setting flex: 1 [17:21:01] plus flexbox was meant for layout, not floats, so :) [17:21:23] I'm growing wearing of using anything with a -webkit- prefix in phonegap [17:22:02] jdlrobson-shinym: well, http://caniuse.com/#search=flexbox [17:22:20] firefox has supported it forever, ie is starting to, etc [17:22:38] (notwithstanding the fact that the standard was revamped quite a bit recently) [17:23:07] sure [17:23:23] I'd just argue that if we don't have to use it - we shouldn't use it - it might bite us later on [17:23:41] (i.e. if spec changes or it's unstable) [17:24:47] jdlrobson-shinym: true, but spec changes would still be backwards-compat - more 'improvements' rather than complete changes, seeing how widespread this one has been implemented [17:24:57] jdlrobson-shinym: i'd agree to that in principle though - for example, indexdb :) [17:25:06] *indexeddb [17:25:15] anyway.. let me see if I can specifically work out why display box causes the layout to jump [17:25:24] i'm curious now :) [17:25:32] heh! [17:25:47] greetings all [17:25:50] jdlrobson-shinym: are you sure that it is not because of position: fixed? [17:26:12] hello tfinc [17:26:21] tfinc: https://bugzilla.wikimedia.org/show_bug.cgi?id=37761 -> was the iPhone on iOS 6? [17:26:42] preilly: --^ [17:48:29] yuvipanda: can't work it out - but it seems to be the combination of position fixed and display box [17:48:42] if I switch to display block or switch to position absolute the layout doesn't jump [17:49:00] jdlrobson: I can understand position: fixed being the culprit, but I'm baffled about box [17:49:11] they just don't play well together it seems.. [17:49:57] jdlrobson: is it always 1px? [17:50:06] I believe so [17:50:22] it's always the same amount of pixels on the left and top [17:50:28] jdlrobson: margin-top: -1px? margin-left: 1px? [17:50:43] noooo because it only happens when the layout changes [17:50:58] yes, I guess it is a variant of http://ejohn.org/blog/sub-pixel-problems-in-css/ [17:51:01] e.g. i click the back button and get a new overlay or I click on search and the search gets focused [17:51:26] anyway brb must make a call [17:51:33] jdlrobson: okay, poke when back [18:15:03] * yuvipanda pokes preilly [18:15:27] yuvipanda: what's up? [18:15:52] preilly: hey! [18:15:55] preilly: was https://bugzilla.wikimedia.org/show_bug.cgi?id=37761 on iOS 6? [18:17:32] yuvipanda: no iOS 5.1 [18:18:40] preilly: steps to repro? Same as http://www.mediawiki.org/wiki/Wikimedia_Apps/Pre_launch_checklist#On_iOS? [18:24:34] yuvipanda: yes [18:29:38] preilly: still waiting on a mail to mobile-tech for which perf graphs to focus on post deployment [18:52:22] hmm, okay. Reproducible on device now - looks like a regression from a prior commit [18:54:22] k [19:00:36] http://www.theverge.com/2012/6/21/3108520/android-4-1-jellybean-confirmed [19:01:01] Android 4.1 Jelly bean, coming soon to the galaxy nexus! :D [19:01:40] tfinc: Asher wanted to create a new set of graphs for it [19:01:56] tfinc: but I guess I'll just mail out what we've got now [19:02:30] if those graphs are going to take more then day then just create a page that links to resources for deployers [19:02:33] and you can update it later [19:02:55] Reedy: nice. but i'd rather just see 4.x on more devices [19:03:20] That is certainly something Apple is better at [19:05:09] tfinc: iOS app seems to lose saved pages across restarts?! [19:05:11] 3.1.2, that is [19:05:27] it does explain why you weren't getting any of your saved pages when you upgrade [19:09:58] works fine on iOS 4.x [19:12:46] yuvipanda: but not on iOS 5.1? [19:12:52] * preilly is confused [19:12:58] preilly: seemingly. [19:13:59] preilly: i think i just confirmed. our app on iOS 5.x didn't have a working saved pages feature at all :| [19:14:48] yuvipanda: hmm [19:19:37] [WikipediaMobile] yuvipanda force-pushed 1.7-migration from 89737e9 to 259bd9b: http://git.io/XMTNfw [19:19:37] [WikipediaMobile/1.7-migration] initial migration attempt - Patrick Reilly [19:19:37] [WikipediaMobile/1.7-migration] fix access origin - Patrick Reilly [19:19:37] [WikipediaMobile/1.7-migration] initial migration attempt - Patrick Reilly [19:37:24] stabstabstab mysql [19:37:31] simply sql too [19:37:52] * yuvipanda stabs self too [19:38:11] but you're not sql? [19:38:35] MaxSem: no, I'm nowhere near as useful [19:40:10] MaxSem: just grabbing some lunch but when I get back could you help me with my comments on - https://gerrit.wikimedia.org/r/#/c/10452/7/specials/SpecialMobileOptions.php - would be keen to get this merged [19:41:36] no, it doesn't redirect [19:41:59] you can use just "$code:" [19:42:12] so no MainPage? [19:42:19] yes [19:42:39] excellent [19:43:11] the second one is simple: no such guideline, just look at how form fields are named [19:43:47] even HTMLForm automatically prepends 'wp' [19:44:18] isn't wp wikipedia specific though? [19:45:09] also where you say use SiteMatrix what are you referring to - the http://www.mediawiki.org/wiki/Extension:SiteMatrix extension? [19:46:01] yes [19:46:19] ok will take a look at that [19:46:32] if we are using wp prefix on inputs we should also change them for beta and images yes? [19:47:15] anyway quickly going to grab some lunch [20:29:54] tfinc: around? [20:30:01] yuvipanda: yes [20:30:32] tfinc: so, I figured the issue. [20:30:42] tfinc: essentially, on 5.1 saved pages has not been saving pages *at all* [20:30:54] tfinc: they've just been disappearing on app / phone restart [20:31:01] tfinc: which is why you were not getting any on upgrade [20:31:22] yuvipanda: that would do it [20:31:40] tfinc: so, I merging preilly's 1.7 branch fixed it [20:31:42] i'm really glad we didn't push it out in its current state [20:31:51] yuvipanda: thats *a lot* of chnage [20:31:52] change* [20:32:32] tfinc: it's much smaller on iOS than on Android - we've only 3 plugins vs 9 on android [20:32:38] tfinc: but yes, it still is not a 'small' change. [20:34:04] yuvipanda: can we not fix it without the 1.7 branch? [20:34:21] if not then we'll have to do another round of testing [20:34:23] tfinc: so, there was a workaround available for 1.4.1 [20:34:30] tfinc: it was applied before v3.1.2-iOS went out [20:34:41] tfinc: and it was tested before it went out [20:35:06] tfinc: but it seems to have had no effect. either a point update to iOS broke it, or it has been broken from the beginning. [20:35:16] tfinc: either way - the workaround doesn't work, and I think 1.7 is necessary [20:41:01] tfinc: the new changes are in master. [20:41:30] yuvipanda: so that sounds like we'd bump up to 3.3 then [20:42:01] tfinc: ? [20:42:18] tfinc: shouldn't they be in sync with android releases? [20:43:53] tfinc: i see that they're based off different versions of phonegap, but I think feature-wise they are pretty much the same, so a version bump would just cause confusion [20:44:44] yes and with this change they will no longer be in sync anymore [20:45:53] tfinc: they will be in sync feature wise [20:46:29] sure but under the hood things will be different [20:46:35] lets broaden this decision out to the team [20:46:52] jdlrobson: preilly : what are your thoughts about fast tracking 1.7 branch and version bumps for tagged releases ? [20:47:23] tfinc: seems okay with at least a full round of testing [20:48:26] fine with me [20:48:42] preilly: and what about versioning? we have the choice of either doing it be feature making this next iOS release 3.2 or by significance in changes making this 3.3 [20:49:34] tfinc: I'm inclined with sticking with by feature [20:50:02] tfinc: even though I understand the issue with the mass amount of under the hood changes [20:50:40] k. barring jdlrobson thinking otherwise we have consensus at versioning by feature [20:50:54] and not by under the hood changes [20:51:44] philinje: preilly did you guys get an invite from open path? i didn't and don't have a call in # [20:52:24] tfinc: so, do I cut an RC4 or do we just test internally and push out? [20:52:43] hmm [20:53:07] we've already done two test passes over 1.7 and have marked it as ok [20:54:36] tfinc: hmm, so do an internal round? [20:55:45] yes [20:56:41] tfinc: okay, let me make an IPA [20:56:54] tfinc: don't think so [21:01:36] tfinc: http://dl.dropbox.com/u/8768784/Wikipedia-iOS-3.2RC4.ipa [21:01:53] ARRG old mac crashed [21:02:27] MaxSem: so SiteMatrix extension might not be installed.. what happens in that situation? [21:02:51] then this functionality should disappear [21:04:19] jdlrobson-shinym, did I understood it right that this thing should just switch between wikis in different languages? not using interwikis or anything? [21:04:28] Correct [21:05:05] meeting time .. [21:07:50] yuvipanda: http://dumps.wikimedia.org/iOS/Wikipedia-3.1.2-RC4.ipa [21:09:58] tfinc: sweet. I'm testing on my devices, do you think you can pull in anyone else? [21:12:00] yuvipanda: mail out to mobile-tech [21:12:04] wiktionary, wikinews and wikisource are now automatically redirecting to mobile for mobile devices [21:12:06] i'm in a meeting and have to focus here [21:12:10] preilly: woot! [21:12:11] tfinc: okay! [21:14:14] curl -A "Mozilla/5.0 iPhone AppleWebKit/420+ (KHTML, like Gecko) Version/3.0 Mobile/1A542a Safari/419.3" -I http://en.wiktionary.org/wiki/Wiktionary:Main_Page [21:14:15] HTTP/1.0 302 Moved Temporarily [21:14:15] Location: http://en.m.wiktionary.org/wiki/Wiktionary:Main_Page [21:21:59] So yuvipanda doesn't seem to be any designs for how search should look [21:22:02] :( [21:22:10] it doesn't seem to be on the workflows? [21:24:15] i can't believe that after this many years of android … users are still asking for "Exit functio would be really nice. What are your devs thinking? I have to use another app to close/kill your app. WAKE UP." [21:25:18] jdlrobson-shinym: hmmm, yes. I assumed it would work like how it is in GMail. [21:25:31] i think linSmith is updating them today [21:25:52] I just left it functional for the time being [21:25:58] but new pull should be rebased [21:26:46] yeah, I'm working on designs [21:26:51] jdlrobson-shinym: why exactly is 'name' and 'timeout' declared at that scope? [21:27:25] jdlrobson-shinym: they could be declared inside #filter-results ' keyup, no? [21:27:37] jdlrobson-shinym: they aren'te being used anywhere else [21:29:44] they should be declared outside the keyup [21:29:47] but maybe in the click [21:30:12] actually no [21:30:15] the rebase went weird [21:30:38] i see no click [21:30:41] hah [21:30:46] dropped some commits unintentionally? [21:30:55] no.. it's fine [21:31:19] it's just the timeout and name should be defined in the #country-list .country-search click handler [21:31:50] 1s I need to rebase this again [21:32:44] okay! [21:32:57] jdlrobson-shinym: github pull reqs doesn't like force pushes, so open a new one [21:34:17] this is driving me mad..\ [21:34:25] git? [21:36:53] try the latest one [21:38:06] jdlrobson-shinym: checking [21:38:32] actually don't think this is right either [21:38:33] grrr [21:39:28] yuvipanda: think this is still messed up.. 1s [21:40:21] ah, then ignore the comments :) [21:40:47] You and your refactoring combined with too much context switching today has confused me ;-) [21:41:47] so yuvipanda the timeout thing has to live outside keyup [21:42:16] as we want other keyup events to be able to access the existing timeout [21:42:30] jdlrobson-shinym: like? [21:42:51] which other timeout? [21:42:53] Well.. the whole point of the timeout is to make sure only one ajax event occurs at once [21:43:04] so say I type Hello [21:43:18] and I pause after the 1st letter l [21:43:24] it will only trigger the one search for the string 'Hel' [21:43:34] instead of one for 'H', one for 'e' and one for 'l' [21:43:58] jdlrobson-shinym: okay, I just realized that was a pretty big brainfart. Nevermind :) [21:44:06] ok cool.. just checking lol [21:44:28] my brain doesn't seem to be working today so I thought it was me ';-) [21:44:45] but yes you are right about the this [21:45:07] yeah, so I guess a var that = this; before should work [21:47:18] k so updated [21:48:19] jdlrobson-shinym: looks good to me, will test and then merge :) [21:49:57] phew [21:50:00] my head hurts [21:50:09] so I see camera based code.. is that functional? [21:50:25] jdlrobson-shinym: yup, it is! [21:50:32] how do I use it? [21:50:34] jdlrobson-shinym: both the camera based code and the gallery based code are [21:50:41] jdlrobson-shinym: http://test.wikipedia.org/wiki/File:Part_of_Meerdaalwoud_(woods)_1340050397664.jpeg test uploads :P [21:50:56] jdlrobson-shinym: so, in the monument-detail page, one of the buttons in the action-bar is 'upload' [21:51:04] they're all placeholders, so you look at the code :P [21:51:10] yeh that's what I figured [21:51:16] any reason just placeholders for time being?\ [21:51:25] jdlrobson-shinym: waiting for assets from linSmith [21:51:38] k [21:51:39] jdlrobson-shinym: so placeholder graphics till then. [21:52:41] do button in the top right is upload? [21:52:46] *so button in the top right.. [21:53:17] jdlrobson-shinym: yup [22:01:16] so yuvipanda are you aware that mw-messages-ready is running twice? [22:01:55] jdlrobson-shinym: on the browser? [22:02:02] yup [22:02:07] jdlrobson-shinym: onDeviceReady runs twice [22:02:19] which means everything runs twice [22:02:52] I might put a flag in there then… as I like debugging in the browser [22:03:20] jdlrobson-shinym: please do. It's a symptom of an underlying problem (that PhoneGap's onDeviceReady doesn't quite like working with AMD) [22:05:48] https://meta.wikimedia.org/wiki/Mobile_Projects/WiktionaryMobileAndroidFeedback [22:13:05] hi yuvipanda [22:13:46] hey pfhayes [22:14:05] yuvipanda: you said you had some existing code for working on shake for random article? [22:15:12] pfhayes: ah, no. I was mentioning that it would be easier for you to start off with the cordova-1.7 branch - i think that version has some improvements for the accelerometer api [22:15:20] pfhayes: but, it just got merged into master, so doesn't matter [22:15:24] yuvipanda: ah i see. thanks [22:16:45] [22:19:10] pfhayes: thanks for the pull request! [22:19:15] pfhayes: i've made a few comments [22:19:32] yuvipanda: thanks [22:26:41] jdlrobson-shinym: mergint it! [22:26:51] but i'll note that it needs to cancel not just the last time out, but also the last request! [22:26:55] sos that the callback is not fired [22:27:41] hmm, actually [22:27:45] jdlrobson-shinym: do you want to fix that before I mereg? [22:27:47] *merge? [22:28:44] jdlrobson-shinym: left another comment, will merge when that's done :) [22:35:11] preilly: so with the first batch of redirects done past messaging and main page creation is there anything else that will slow down the next batch? [22:36:42] tfinc: nope [22:37:04] yuvipanda: how much longer have I got of you? [22:37:22] excellent [22:37:47] jdlrobson-shinym: ~15 mins? [22:37:57] * yuvipanda is testing the app on his two devices [22:38:05] ok… well I had one thought about the back button - it should be a page-link [22:38:16] and we should set the data page attribute of the back button for things such as login [22:38:20] jdlrobson-shinym: i think it is, no? [22:38:23] jdlrobson-shinym: only login is different [22:38:24] rather than unbinding all the time [22:38:29] jdlrobson-shinym: and I'm wondering how to fix that [22:38:33] yeh login shouldn't be different - this is what I'm saying [22:38:35] I have a proposed fix [22:38:35] jdlrobson-shinym: 'back' just applies styles [22:38:47] jdlrobson-shinym: a page stack? [22:38:53] but I'm in a meeting now so it's probably not going to be with you until 30 minutes time [22:39:07] I was just setting data('page') inside the click handler [22:39:37] hmmm, there's prevPage being set in showPage that we might perhaps use [22:40:03] jdlrobson-shinym: but yes, the login back button is an anomaly that should be fixed [22:44:25] Amgine: hi [22:44:39] [22:45:08] just checking, has there been any announcement to wiktionary-l or the wiktionary village pump? [22:47:36] Yes for wiktionary-l, and it has been discussed a couple of times on the several en.WT discussion pages. [22:48:07] But not the latest official release, which was awaiting the blog entry and so is now overdue. [22:49:18] pfhayes: which of us should do so? [22:49:41] Oh, heh... it's being discussed on el.wt. [22:49:56] Amgine: you can do ti [22:50:09] kk. Will get it in the next hour or two. [22:52:22] great, thanks [22:52:50] Which languages should I focus on, philinje? [22:53:59] let me grab the top 10 languages, but if you want we could do an automatic post to all languages - might take a little while to get done [22:55:08] top 10 languages: en, fr, de, ru, pl, pt, ja, zh, es, it [22:56:25] we recently used Global Message Delivery to put a message about mobile homepage formatting on all language versions of Wiktionary [22:56:46] that might be overkill in this case [22:56:48] [23:04:00] Have a good weekend everyone. [23:06:10] weekend already? [23:06:56] Family gathering in NC. [23:07:06] ah, have fun caxthelm [23:07:48] Thank you, I plan to spend the entire time eating boiled peanuts. [23:17:58] jdlrobson-shinym: https://github.com/wikimedia/WikipediaMobile/pull/259 before you leave, if possible :) [23:18:11] sure [23:19:14] jdlrobson-shinym: thanks! [23:20:25] hmm, i think the sun rising should be a good indicator for my 'time to go sleep'. [23:20:28] so... not yet :) [23:21:14] :) [23:25:35] [WikipediaMobile] yuvipanda pushed 1 new commit to master: http://git.io/7cvvJQ [23:25:35] [WikipediaMobile/master] Removing 'Select Text' menu option on Android 3.0+ - Patrick Hayes [23:26:05] pfhayes: merged :) [23:26:13] yuvipanda: thanks! [23:35:00] alright, i think it is bright enough [23:35:02] good night everyone [23:35:11] * yuvipanda hopes that at least some people other than him test out this RC [23:35:25] gnite jdlrobson-shinym, and good luck with your apartment search :) [23:35:33] night yuvipanda