[05:15:39] (03CR) 10Jdlrobson: [C: 04-2] "this reverts work we are doing to provide a mobile friendly history view. We override the watchlist special page on mobile and this should" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/96963 (owner: 10Legoktm) [08:35:48] (03PS1) 10Yurik: Optimizations for zero config editing [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/97484 [12:03:51] (03PS1) 10Hashar: browsertests: profile for default settings [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97497 [12:04:26] (03CR) 10Hashar: [C: 04-1] "Work in progress.. Similar to ULS change https://gerrit.wikimedia.org/r/#/c/97487/" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97497 (owner: 10Hashar) [12:32:29] (03PS1) 10Yuvipanda: Save scroll position when restoring [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97499 [15:22:10] is this a fatal from Mobile? https://bugzilla.wikimedia.org/show_bug.cgi?id=57540 [15:30:36] (03PS1) 10Kaldari: Adding styling for wikitables. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97518 [15:35:14] chrismcmahon: Looks like it might be already fixed in master... [15:35:21] It's looking in the wrong namespace [15:35:35] if ( class_exists( 'MobileContext' ) ) { [15:35:35] if ( MobileContext::singleton()->shouldDisplayMobileView() ) { [15:36:14] But beta is running head? [15:36:35] Not quite sure why it's using that namesapce [15:39:45] Reedy: beta should be running master, but I get the error there just now [15:42:43] https://github.com/wikimedia/mediawiki-extensions-GettingStarted [15:42:44] Ah [15:42:49] Hooks file has been updated since [15:42:50] And renamed [15:43:15] chrismcmahon: That page is full of potential fatals like that [15:43:23] Let me fix [15:44:47] Reedy: always nice to find fallout Monday morning from late Friday merges [15:46:25] Next question is if there's someone to review it [15:48:20] bleugh [15:48:23] aude: about? [15:48:38] Does class_exist check the global namespace? Or just the current? [15:51:50] Reedy: what is the problem? [15:52:41] The problem itself is fixed (fatal from missing class due to checking of the wrong namespace) [15:52:53] Just need someone to confirm if class_exists does the global namespace if in another? [15:52:59] Or do I need to prepend \ there too? [15:55:05] should work without \ [15:55:20] I guess we can soon see on beta ;) [15:55:25] ok :) [15:55:40] Fancy reviewing https://gerrit.wikimedia.org/r/97522 for me then please? [15:56:50] looking [16:02:59] ok, so they put stuff in a namespace on friday? [16:04:09] namespaces are stupid [16:04:13] (in PHP) [16:04:18] merged [16:04:37] :) [16:11:54] * aude kicked off the internet [16:12:13] Reedy: your patch looks fine, although they are doing things inconsistently [16:15:57] * aude kicked off internet again [16:58:04] chrismcmahonbrb: Should be fixed in master... Mayebe [16:58:11] Maybe on beta if jenkins is feeling nice [16:58:21] Reedy: it is, I just checked, thanks [16:58:27] great [16:58:45] chrismcmahon, anything else looking broken? [16:59:06] Reedy: while I have you here, any comments on this would be welcome: https://gerrit.wikimedia.org/r/#/c/96161/ [17:00:02] MaxSem: maybe so, I have some failures in the Mobile tests I didn't expect, looking now... [18:19:19] (03PS2) 10Jdlrobson: Hygiene: Simplify History error handling [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97145 [18:19:59] MaxSem: FYI I asked kenanwang and kaity to talk to you about things we need to improve on Special:History to make it more powerful to power users like yourself e.g. highlight things like deleted comments etc [18:20:58] MaxSem: should work now [18:21:35] (03CR) 10Jdlrobson: [C: 04-1] "Not a real -1. Code looks fine - I just think we can solve a bug in the process of fixing this :) \o/" (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97518 (owner: 10Kaldari) [18:22:14] jdlrobson, I was really worried that we'll have to constantly reproduce the core for watchlist, history, RC... [18:22:24] YuviPanda, thanks a lot:) [18:23:15] cool I would like to help! [18:23:32] (03PS2) 10Jdlrobson: Some fixes to the new overlay [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97213 (owner: 10Kaldari) [18:23:45] mhurd: connectivity card has been moved to sprint 19 [18:23:53] tfinc: ah thanks! [18:24:04] (03CR) 10Jdlrobson: Some fixes to the new overlay (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97213 (owner: 10Kaldari) [18:24:36] mhurd: YuviPanda brion that leaves us with just the code review column which is blocked on mails and review from brion. i want to close the board today so lets move those [18:24:45] whee [18:24:51] officially i'm going to say that we completed 26/29 story points [18:25:24] (03CR) 10Jdlrobson: [C: 032] Some fixes to the new overlay [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97213 (owner: 10Kaldari) [18:25:34] which means we got 89% of our stories done [18:25:39] thats our best average to date [18:26:01] kenanwang: when you trim sprint 19 let me know how many story points we have [18:26:28] we're at 26 [18:26:58] plus the 3 carrying over from sprint 18 [18:27:10] ok [18:27:17] (03Merged) 10jenkins-bot: Some fixes to the new overlay [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97213 (owner: 10Kaldari) [18:32:12] (03CR) 10Brion VIBBER: "some notes" (037 comments) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97112 (owner: 10Yuvipanda) [18:32:51] brion: looking at co,,ents [18:34:07] brion: re: the bug [18:34:16] brion: it's related, and I filed another bug but put the wrong number on it [18:34:26] brion: let me find it [18:35:41] brion: https://bugzilla.wikimedia.org/show_bug.cgi?id=57440 actually [18:35:47] * YuviPanda pokes MaxSem with https://bugzilla.wikimedia.org/show_bug.cgi?id=57440 [18:36:13] will do today [18:37:01] MaxSem: ty [18:37:08] (03CR) 10Brion VIBBER: Add a level of indirection when sending messages to WebView (031 comment) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [18:37:39] MaxSem: Is the existing history code for core modular enough that we could just use it for our needs? [18:37:56] or maybe tweak it so it is? I too would hate us to rewrite things [18:38:22] (I think it's fine to change the HTML markup but the underlying way to get at stuff should be identical) [18:38:34] jdlrobson, not very modular - I was thinking of making it "skinnable", but not sure how much work would it be [18:38:42] (03CR) 10Brion VIBBER: "I might recommend using a pingback from the webview to indicate when processing of the first message is complete, instead of a hardcoded 0" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97138 (owner: 10Yuvipanda) [18:38:45] (03CR) 10Yuvipanda: Add a level of indirection when sending messages to WebView (031 comment) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [18:38:57] MaxSem: I think that would be really really really good. Our Watchlist code also points out that this is much needed [18:39:46] (03CR) 10Yuvipanda: "I agree - it should ideally ping back after putting in each section and then we can give it the next section. But this makes the app so mu" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97138 (owner: 10Yuvipanda) [18:42:03] (03CR) 10Brion VIBBER: Add a level of indirection when sending messages to WebView (031 comment) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [18:42:21] (03CR) 10Brion VIBBER: "fair enough :D" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97138 (owner: 10Yuvipanda) [18:44:02] (03CR) 10Yuvipanda: Add a level of indirection when sending messages to WebView (031 comment) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [18:44:47] (03CR) 10Brion VIBBER: Add a level of indirection when sending messages to WebView (031 comment) [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [18:45:53] brion: ping? [18:46:04] pong YuviPanda [18:46:04] saw my messages about the bug? [18:46:14] YuviPanda: yeah [18:46:24] brion: so essentially, we always have to pass prop=sections [18:46:32] and that gives us extra empty sections that we discard [18:46:40] ok that explains it [18:46:45] brion: I can change the bug number in the code [18:46:45] wanna fix the comments? :D [18:46:48] yeah [18:46:48] thanks :D [18:46:49] doing [18:46:53] ok testing now... [18:47:33] rmoen: you there [18:47:34] ? [18:48:09] fun fact -- i can install the Glass app on my nexus 5 by mistake, but it doesn't, um, work :D [18:48:11] lemme close that window [18:48:32] i got too many projects open :D [18:48:49] brion: haha [18:49:17] so we've unpacked like one plate, and have no shower curtains yet [18:49:30] I also just noticed that this entire week has zero non-standup meetings [18:49:31] woohoo [18:49:36] \o/ [18:50:30] java: /Users/brion/src/wiki/apps-android-wikipedia/wikipedia/src/main/java/org/wikimedia/wikipedia/CommunicationBridge.java:113: cannot find symbol [18:50:31] symbol : class JavascriptInterface [18:50:31] location: class org.wikimedia.wikipedia.CommunicationBridge.BridgeMarshaller [18:53:08] trying to find the old 'import changes from maven model' button in case that helps [18:53:08] (03PS8) 10Mhurd: Customized the search box area to look more like designer mockups. [apps/ios/wikipedia] - 10https://gerrit.wikimedia.org/r/96956 [18:53:27] brion: ^ just amended with fix for ios6 [18:53:33] \o/ thx [18:54:02] brion: checking other issue real quick... [18:55:05] brion: ok to do the other fix as follow up a bit later> [18:55:06] ? [18:55:19] YuviPanda: ok i did a maven reimport, and it's still saying it doesn't know @JavascriptInterface [18:55:30] brion: hmm, what API level? [18:55:44] brion: can you try changing it manually to a newer one on IntelliH? [18:55:47] docs say it's new in 17... [18:56:10] brion: yeah, it's just ignored on older versions and needed on newer [18:56:12] project is set to build with maven android 4.1.2 [18:56:18] right [18:56:21] that needs to change [18:56:34] brion: let me do that [18:56:34] ok does that need to be fixed in the pom.xml? [18:56:43] brion: pom.xml and AndroidManifest [18:56:52] ok [18:56:54] brion: you can just change it in intelliJ too [18:56:57] brion: for the build [18:57:14] right but then it'l just break again later sometime :) [18:57:56] why can't the api levels just be android version numbers [18:58:03] this would save remember what the mappings are all the time [18:58:39] there we go, that's 4.3 [18:58:57] brion: i'll amend the change now [18:59:03] tx [18:59:20] brion: but do test still :D [19:01:01] * brion hmms [19:01:23] YuviPanda: what should the enter/done/next key do in the search field? just update the search, or drop focus, or jump to the first result? [19:02:05] brion: drop focus? [19:02:09] brion: it says 'done' [19:02:23] yeh then you see more results... [19:02:28] in which case we should ask for more than 5 :D [19:02:45] brion: would it be ok to do the search field tapping fix as a later ux polish? [19:02:52] mhurd: oh yeah [19:02:57] brion: ah, hmm [19:02:59] main one was the iOS 6 fix :D [19:03:07] brion: yeah that one was bad [19:03:11] otherwise things seem much improved :D [19:03:14] brion: I think 5 is still fine. perhaps some styling for the bottom, maybe [19:03:14] :) [19:04:00] YuviPanda: hmm the scroll position check doesn't seem to be working for me on back button [19:04:45] brion: ah, hmm. It's meant currently to work when you go somewere else and come back [19:05:00] shouldn't the magic of fragments Just Work tho? : [19:05:01] :D [19:05:02] brion: yay! i'll go see if mockups are ready... [19:05:47] brion: sortofnotreally :D [19:05:58] hehe [19:05:59] brion: the life cycle when things get destroyed recreated is different [19:06:06] brion: it's trivial to add, let me do that in a bit [19:06:09] ok we'll improve that later then [19:06:11] spiffy [19:06:31] brion: yeah another patch [19:09:21] YuviPanda, hmm [19:09:41] mhurd: nice! it's even got the translucent effect on 6 :D [19:10:22] something is icky about this... [19:10:46] (03PS1) 10Yuvipanda: Update target SDK to be able to find @JavascriptInterface [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97543 [19:10:51] brion: ^ [19:11:09] brion: my git is icky somehow about rebasing, so just added a new patch than update existing [19:11:25] (03CR) 10Brion VIBBER: [C: 032 V: 032] "fixes ios 6 :D" [apps/ios/wikipedia] - 10https://gerrit.wikimedia.org/r/96956 (owner: 10Mhurd) [19:11:30] ok [19:11:36] brion: it's pretty close :) [19:11:47] brion: ah! thanks! [19:13:28] YuviPanda, essentially you're requesting that §ions behave differently depending on whether prop=sections is specified or not [19:13:53] MaxSem: yes. [19:13:56] MaxSem: well [19:14:11] MaxSem: if it respects §ions... [19:14:11] that's even more misleading, no?:) [19:14:16] MaxSem: how so? [19:14:23] MaxSem: I ask for section 0 but get all of them back [19:14:44] but... they're empty! [19:15:01] brion: YuviPanda: katie's finalizing the history mockups right now [19:15:17] \o/ [19:17:01] jdlrobson: i forgot to ask during standup - were you able to get to any of the eventlogging related stuff on friday? [19:17:31] awjr: the fix for it or working out whether the data is useful? both counts no - but the latter I left with kenan [19:18:28] ok - you mentioned you were going to dig into eventlogging issues more generally as part of the hygiene day, did that happen at all? [19:20:09] jdlrobson: ^ [19:20:41] if not, we should make sure to tackle it sooner rather than later - it sounded like you and ori unearthed potentially problematic code, iirc [19:20:45] awjr: it's not really hygiene - there's just a bug somewhere in the code [19:20:53] in our code not event logging itself [19:20:57] right [19:21:00] brion: YuviPanda: should we move the spikes in sprint 18 code review column to done? [19:21:01] is there a bug open for that jdlrobson? [19:21:15] no because it's not a bug - we killed the code path [19:21:31] we should think about arranging a spike if anything [19:21:35] oh - you just said "there's a bug somewhre in the code" :p [19:21:45] mhurd: brion hasn't merged mine yet :D [19:21:59] (03CR) 10Brion VIBBER: [C: 032 V: 032] Display entire page than just first section [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97112 (owner: 10Yuvipanda) [19:22:07] there is/was/ [19:22:12] MaxSem: so if it can respect §ions no matter what... [19:22:22] hmm [19:22:35] "Can Merge No" [19:23:00] wat [19:23:11] says it needs a rebase [19:23:22] brion: weiird [19:23:37] brion: looking [19:23:55] CONFLICT (content): Merge conflict in wikipedia/src/main/java/org/wikimedia/wikipedia/PageViewFragment.java [19:23:57] jdlrobson: ok - if you want to put together a spike, maybe sync pu with kenan before the kickoff today [19:24:27] brion: ah, might be Tpt_'s patches [19:25:04] brion: rebasing [19:25:16] whee [19:26:14] brion: whaaat, it tells me there are conflicts in... .gitreview :| [19:26:22] o_O [19:26:52] * YuviPanda does it again [19:26:58] brion: wil you be coming into the office today ? [19:27:29] tfinc: not today, we have yet to unpack shower supplies :P [19:27:36] | ✗| → git rebase gerrit/master [19:27:37] i'm CRing from here for now [19:27:37] Current branch HEAD is up to date. [19:27:40] brion: k. can you mail out your schedule for this week? [19:27:45] will do [19:27:45] brion: great. cr is best now [19:27:46] * YuviPanda pokes around git [19:28:21] brion, YuviPanda: Hi! Please don't add api = ((WikipediaApp)getActivity().getApplicationContext()).getPrimarySiteAPI(); in the code, it break the hability to browse not English wp wiki. [19:28:38] Tpt_: yeah, don't worry about it, not going to add it back :) [19:28:45] Tpt_: oh wait [19:28:51] oh right [19:29:00] Tpt_: it should be all getPrimarySite() only [19:29:06] which I think is fine [19:29:49] No because you may want to browse a page of a wiki that isn't your primary one. [19:30:30] do this.page.getTitle().getSite() [19:30:39] what happens when you don't have a page [19:30:43] search primarily [19:31:11] getPrimarySite should eventually hook into sharedPreferences and a settings activity [19:31:13] In that case you use the default wiki [19:31:23] indeed, and that is what the code does now [19:32:18] I was talking about PageViewFragment that should allow to open any page on any wiki [19:34:18] Tpt_: oh that's just a merge conflict, it'll go away when I'm done rebasing [19:34:51] Ok. I understand now :-) [19:34:59] Thanks [19:35:50] (03PS2) 10Yuvipanda: Display entire page than just first section [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97112 [19:35:56] Tpt_: see ^ :) [19:36:38] YuviPanda: Thanks, it's nice now :-) [19:37:04] :) [19:37:07] (03PS2) 10Yuvipanda: Make the search textbox single line only [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97211 [19:37:08] (03PS2) 10Yuvipanda: Stagger messages to webview when restarting a pageview. [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97138 [19:37:09] brion: rebased! [19:37:09] (03PS2) 10Yuvipanda: Save scroll position when restoring [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97499 [19:37:10] (03PS2) 10Yuvipanda: Add a level of indirection when sending messages to WebView [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 [19:37:16] whee [19:39:46] YuviPanda, I'm having parse errors.... so what do you think prop=sections should control? [19:40:09] MaxSem: good question. I don't know [19:40:43] CLOSED INVALID: requestor doesn't know what he wants:P [19:40:44] MaxSem: maybe it could just go away? [19:41:06] MaxSem: well, what I want is that if I say §ions=0, I should get just section0 [19:41:09] not all of them [19:41:12] (03CR) 10Brion VIBBER: [C: 032 V: 032] Display entire page than just first section [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97112 (owner: 10Yuvipanda) [19:41:24] seriously though, I admit that I overcomplicated it:P [19:41:53] (03CR) 10Brion VIBBER: [C: 032 V: 032] Add a level of indirection when sending messages to WebView [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97135 (owner: 10Yuvipanda) [19:42:16] MaxSem: :D [19:42:30] (03CR) 10Brion VIBBER: [C: 032 V: 032] "Merging, but be warned this should be improved!" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97138 (owner: 10Yuvipanda) [19:42:30] MaxSem: can you tell me what prop=sections is supposed to do [19:42:45] (03CR) 10Brion VIBBER: [C: 032 V: 032] Make the search textbox single line only [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97211 (owner: 10Yuvipanda) [19:43:29] YuviPanda, I envisioned the use case like you could request basic information about all sections but text only for lede [19:43:53] yeah i think the idea of that was you'd grab the skeleton, then fetch sections later or on demand [19:43:59] yeah the current app does that [19:44:05] but new one... does not. [19:44:28] so I have to skip that extra info since I'm fetching it again anyway [19:45:11] brion: still two more, btw :D [19:45:18] patience :D [19:45:39] YuviPanda: so i notice the app seems to crash on rotation [19:45:49] brion: oh? [19:45:58] yeah [19:46:26] YuviPanda: https://gist.github.com/brion/7647565 [19:47:15] brion: what patch are you at? line 118 gives me nothign.. [19:47:24] https://gerrit.wikimedia.org/r/#/c/97499/ [19:47:45] hmm so am I [19:48:05] well lemme finish up these merges and then well track it donw [19:48:18] brion: ok [19:48:18] (03CR) 10Brion VIBBER: [C: 032 V: 032] Save scroll position when restoring [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97499 (owner: 10Yuvipanda) [19:48:29] brion: can you tell me what you see on PageViewFragment line 118? [19:48:48] // Enable Pinch-Zoom [19:48:51] which is all very odd [19:48:55] cause it's a comment O_O [19:49:29] brion: yeah same here :| [19:49:35] (03CR) 10Brion VIBBER: [C: 032 V: 032] Update target SDK to be able to find @JavascriptInterface [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97543 (owner: 10Yuvipanda) [19:49:37] brion: maybe you have an older version on the phone? [19:49:55] i'm clearing it out to be sure :D [19:50:12] brion: :D [19:51:55] https://gist.github.com/brion/7647696 ok now says like 123 [19:52:07] which is linkHandler = new LinkHandler(getActivity(), bridge, title.getSite()); [19:52:36] brion: ah that's better [19:52:51] brion: commiting a fix [19:52:57] brion: YuviPanda lunch and errands! [19:53:08] yay [19:53:27] back @1 [19:54:01] (03PS1) 10Yuvipanda: Ensure that we have a title before using it. [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97547 [19:54:04] brion: ^ [19:54:20] \o. [19:54:23] \o/ [19:55:15] so i got the android 4.4 update on my old nexus 7 [19:55:25] but it doesn't get the new launcher with google now integration, bummer [19:55:43] brion: nexus 4 also doesn't seem to get the GNow integration [19:55:54] that's lame! [19:55:57] well only slightly lame [19:56:05] cause really you can still bring it up with a different gesture [19:56:15] but... consistency man [19:56:16] it's nice [19:56:45] brion: heh [19:57:12] brion: I bet the fact that pre 4.4 is webkit and post is blink will bite us sometime too [19:57:12] (03CR) 10Brion VIBBER: [C: 04-1] "Ok, not crashing but it's popping the search overlay open again every time I switch orientations." [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97547 (owner: 10Yuvipanda) [19:57:19] heh [19:57:52] brion: that's an unrelated bug that happens if you have slowish internet. [19:58:24] weird, now it's not doing it [19:58:27] * brion blames comcast [19:58:44] (03CR) 10Brion VIBBER: [C: 032 V: 032] "Ok that seems intermittent and unrelated. Yay!" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/97547 (owner: 10Yuvipanda) [19:58:44] brion: so I was on a bus for 6 hours yesterday [19:58:49] brion: WITH GPRS [19:58:51] brion: not even EDGE [19:58:54] but GPRWS [19:59:02] daaaaaamn [19:59:03] YuviPanda, so what kind of requests do you want to make now? [19:59:05] brion: a search request for 'a' took 10s to return :D [19:59:15] MaxSem: let me give you the exact ones, moment [19:59:56] brion: and because of our two fetches, I would get the lead section first in about 10s, and then the rest after like 30 or so :D [20:01:11] owie [20:01:18] we'll have to practice low-bandwidth stuff :D [20:02:47] ok i think it's lunchtime [20:02:58] then i'll catch up on design-related emails and such :D [20:03:30] brion-away: :D [20:03:43] MaxSem: https://dpaste.de/4rYq/raw [20:03:58] MaxSem: first query has sections=0, second has sections=1- [20:04:36] mmm, so how will you know other sections' names now? [20:05:28] MaxSem: I.. don't? [20:05:40] MaxSem: I don't need to. I'm just displaying the lead first, then displaying the rest [20:05:47] MaxSem: and we aren't collapsing sections by default :) [20:06:04] mmm, what if lede is too short and user sees nothing at bottom? [20:06:10] MaxSem: spinner [20:06:25] if the lead is too short the rest of the article perhaps is too :P [20:06:52] nope, a lot of lists are huge but have only 1 sentence in section 0 [20:07:09] MaxSem: well, that's still fine. There'll be a spinner [20:08:18] grrrr, we forgot to mention that our task should be done without using any stupid frameworks [20:09:21] awjr|lunch, what do you think? ^^ [20:09:34] * YuviPanda rewrites Mediawiki in RoR [20:11:02] * MaxSem rewrites YuviPanda in /dev/null [20:11:54] :D [20:15:51] MaxSem: so, thoughts on the API? [20:16:08] YuviPanda, same question:P [20:16:30] MaxSem: make §ions be respected? [20:16:47] what about existing callers?;) [20:17:37] MaxSem: ah, hmm. [20:17:59] MaxSem: well, if prop=sections isn't specified, then only6 send back sections that are specified in §ions [20:18:08] MaxSem: else, keep current behavior? [20:19:27] MaxSem: that shouldn't affect current callers [20:19:41] O RLY? [20:19:47] * MaxSem fires up his grep [20:19:47] YARLY [20:20:09] if I specify prop=sections, I'll get back all section skeletons [20:20:12] if not... I don't [20:20:18] I get only the sections I asked for [20:21:44] YuviPanda, the first request I see in logs has no prop=sections. it's definitely not made by our app or MF - what if we break it? [20:22:15] MaxSem: does it also have sections= =all [20:22:35] then that won't change anything will it [20:22:42] it gets full content of all sections [20:22:57] MaxSem: what will break is no prop=sections and sections != all [20:24:16] grepping [20:39:31] mhm, YuviPanda - so for the partial day of today, we'd break 78997 requests:P [20:39:50] MaxSem: that have no prop=sections and sections != all? [20:39:56] yup [20:40:11] mostly sections=0 [20:40:21] hmmm [20:40:27] MaxSem: so how do we do this? [20:42:45] grrrr [20:43:29] MaxSem: have a 'besane=1' thing? :P [20:43:48] sane? nevar! [20:45:21] * Platonides puts MaxSem inside a jail for safety [20:45:41] but providing him a fiber connection and a laptop so he can continue hacking :) [20:46:25] * MaxSem hates laptops! [20:47:08] MaxSem: prop=justsectionsispecify? [20:47:15] MaxSem: or onlyspecifiedsections=1 [20:49:18] * MaxSem slips into his warm and cozy straightjacket [20:51:27] dr0ptp4kt, i have a number of new patches waiting for you :) [20:51:55] yeah, i'm reviewing 97484 [20:52:15] dr0ptp4kt, hehe, that's a complex one :) [21:02:57] dr0ptp4kt, btw, one thing that i can't figure out there is why empty saves do not get detected as such [21:27:48] YuviPanda: will i be able to close out https://trello.com/b/buxMRu19/mobile-app-sprint-18-search-connectivity today ? [21:28:03] tfinc: oh, yes. brion merged 'em all [21:28:05] tfinc: let me move them [21:28:31] YuviPanda: then all that you have to do is to mail mobile-tech with the answers to the questions we had in the spike and were done [21:28:31] tfinc: moved all android cards, there's an iOS one left in CR though. [21:28:36] tfinc: sweet [21:28:46] * YuviPanda digs up draft [21:28:52] mhurd can tackle the draft with you [21:28:58] since he said his findings were mostly the same [21:29:07] yeah, is he around? [21:29:10] * YuviPanda pokes mhurd [21:29:11] tfinc: YuviPanda ya [21:29:25] mhurd: want to etherpad it? [21:29:51] YuviPanda: ya i did the 2 requests - one for first section, one for remaining sections [21:30:02] mhurd: how much of a perf gain are you seeing? [21:30:21] (03CR) 10Jdlrobson: [C: 04-1] "Add a test please..." [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97080 (owner: 10MaxSem) [21:30:55] YuviPanda: when on super slow connection it seemed roughly in line with what you said you saw [21:31:06] i'd say 3 to 4x [21:31:22] (03CR) 10Jdlrobson: [C: 04-1] "Can we make the commit message more meaningful?" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/96962 (owner: 10Mgrover) [21:31:23] depending on article size [21:31:31] mhurd: was there any noticeable degradation when on a fast connection? [21:31:50] YuviPanda: no :) [21:32:02] mhurd: that's sweet. [21:32:20] mhurd: we're doing uncollapsed sections, right? [21:32:27] YuviPanda: yes [21:32:31] sweet [21:32:57] mhurd: so to recap: you're asking for section 0 (lead), display just that, then ask for sections 1- (everything else), then load it when it lands [21:33:17] and that gives us 3x-4x bump on slow connections, and no degradation on fast connections [21:33:19] yes with one more thing... [21:34:28] when i load section 0 i put a "loading the rest of the article..." message in a div at the bottom of the section 0 html. then when the remaining sections load i remove that div [21:35:26] mhurd: right, I should do that... with a spinner or so [21:35:37] mhurd: ok I'll write it up in a moment [21:36:18] YuviPanda: whether it's a spinner or text shouldn't matter too much - i'm sure designers will tweak [21:36:48] YuviPanda: thanks! oh, katie said mocks should be ready in half hour or so [21:36:54] nice [21:37:01] YuviPanda: can't wait :) [21:37:05] :) [21:38:05] YuviPanda: thanks for doing spike write-up! [21:38:10] :D [21:38:11] yw [21:38:26] mhurd: I'm going to boil some eggs now will write in a bit [21:38:40] YuviPanda: fooooooooood! [21:38:54] :D [21:38:59] mhurd: also have 2L of ice cream :D [21:39:08] YuviPanda: now you're talking! [21:39:35] i eat way too much ice cream [21:39:50] mhurd: we boiled cardamom and a bunch of other spices in water and made some really really great Tea [21:40:19] ooh sounds good [21:47:43] :D KitKat update [21:47:45] brion-away: --^ [21:48:17] YuviPanda: --^ Nexus4 [21:48:24] tfinc: pfft, I got it on friday :P [21:48:32] YuviPanda: how has it been so far ? [21:48:55] tfinc: not much of a difference. ART felt a little snappier than Dalvik. [21:50:49] YuviPanda: what are the user facing changes? given that you've had it since friday i'm sure you've had enough time with it [21:50:58] the joy of the early adopter [21:51:18] tfinc: honestly, nothing at all that I've noticed. [21:51:24] ok [21:51:27] tfinc: I didn't get the new launcher, which is the biggest change. [21:51:35] tfinc: idk why I didn't get it, need to i nvestigate [21:51:40] didn't get it ? [21:51:55] tfinc: yeah, my launcher is still the old one. dunno why. [21:52:16] tfinc: they changed the color of all the icons in the notification bar to be completely monochrome, me no likey. [21:52:36] i'm eager to see this now [21:52:36] tfinc: previously the phone signal icon was blue if you had internet access and white otherwise. now it is always white. [21:52:56] tfinc: and before when GPS was hunting for a fix there was an animation in the notification bar. no more. [21:52:56] YuviPanda: i just spoke with vibha about what should happen to the search box text when a search result is selected. she said for now just clear it out so i goes back to the placeholder text - i.e. "Search Wikipedia" [21:54:03] mhurd: ah, hmm ok [21:55:04] new bootup screen [21:55:10] ah that too :P [21:55:51] i'm surprised it can do all these updates without decrypting my files [21:56:00] i haven't put in my password for any of these steps [21:56:13] makes me think think the encryped portion is only user files [21:57:08] (03PS2) 10Kaldari: Adding default styling for tables [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97518 [21:57:22] YuviPanda: i had tried changing the text of the search box to the title of the selected article, but then it got screwy because when you navigate from within the article to another article the search box text would be out of sync with the article being shown. so then i thought just always change the search box text to the text of the currently shown article, but the problem with that is the search box's functionality gets obscured if it's always [21:57:23] article title - and just switching it to the default text "Search Wikipedia" makes all these problems go away [21:57:42] mhurd: yup, I AGREE WITH CLEARING [21:57:43] UGH [21:57:44] (03PS3) 10Kaldari: Adding default styling for tables [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97518 [21:57:51] mhurd: accidental caps sorry [21:59:23] jdlrobson, is nearby still needed on beta?:P [22:02:30] YuviPanda: haha [22:02:46] mhurd: how about this [22:02:51] mhurd: we clear it when they navigate away [22:02:56] mhurd: from the page they reached [22:03:11] YuviPanda: no when they make selection from search results [22:03:20] MaxSem: hmm? [22:03:30] mhurd: no, I'm suggesting an alternative [22:03:36] mhurd: which will also ensure they are never out of sync [22:03:39] Michelle already said yes:P [22:03:49] mhurd: and also gives me an opportunity to edit / change things if needed [22:05:05] YuviPanda: vibha said the use case for that (having the title in the search bar initially) was far less strong that that for having the reminder that the search box is used for searching (by showing default "Search Wikipedia" text) [22:05:45] mhurd: you mean I just searched there and got somewhere and then I forget I use that for search? [22:05:56] YuviPanda: yes [22:06:13] :P [22:06:25] haha [22:06:49] mhurd: that does not make sense to me. [22:07:47] (03CR) 10Kaldari: [C: 032] Remove infobox experiment from MobileFrontend [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/96397 (owner: 10Jdlrobson) [22:07:51] (03CR) 10jenkins-bot: [V: 04-1] Remove infobox experiment from MobileFrontend [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/96397 (owner: 10Jdlrobson) [22:08:02] YuviPanda: you are young and smart, i think about some of my older less tech-inclined relatives and have zero trouble imagining them being confused if they don't see the "search" prompt [22:09:14] mhurd: This feels very much like a bikeshed to me, and it also feels weird talking to her via a proxy. Can you ask vibha to mail out? [22:09:21] to mobile-tech? [22:09:39] I'm going to write out the spike email and then head to sleep now [22:10:02] YuviPanda: it's not a big deal for right now. just something to think about [22:10:21] mhurd: ok :) [22:10:51] YuviPanda: sorry about the weirdness, just me poorly summing up [22:11:36] YuviPanda: we'll have lots of little polish stories in later sprints :) those are fun [22:11:44] YuviPanda: get some sleep! [22:12:00] mhurd: I've the fear that we'll just end up bikeshedding a lot over minor issues. [22:12:33] like this one, for example. There is a problem - inconsistent text in search bar. Multiple solutions, are subtly different, none can be considered 'right' [22:12:59] and then we'll end up spending a lot of energy writing emails and talking about those [22:13:23] YuviPanda: that's why we have ux people :) their job is to make the call i guess [22:14:44] mhurd: I personally don't like the sound of that, since if writing code at any point feels like 'implementing what someone else made a call about', I guess I just tune out and then quality / enthusiasm suffers greatly [22:14:50] consensus is a nice thing :) [22:14:55] we'll get there when we get there I guess [22:15:22] YuviPanda: now you're bike shedding :) [22:15:26] but it's something I worry about a lot, since for most of the time I'm just a rectangular blob of pixels on a screen and communicating is hard at all [22:15:42] mhurd: i'm perhaps needlessly worrying, *not* bikeshedding :P [22:15:56] YuviPanda: of course there's discussion and attempt to reach consensus [22:16:14] lol [22:17:51] YuviPanda: katie attached browser history mockups to the trello cards for this sprint! [22:18:53] mhurd: sweet! I'll take a look [22:18:58] writing email now will head out in a bit [22:19:10] YuviPanda: sounds good! have a good night! [22:25:23] kaity: saw the mockup, looks great! [22:25:35] thanks YuviPanda! [22:25:50] I'll add some more to the acceptance criteria [22:26:32] kaity: one thing though - I'd want to follow Android design guidelines and keep the top to be an action bar. So when we are in browsing history, the search stuff will disappear and the title 'browsing history' will be there [22:27:25] kaity: so search bar won't appear on *all* screens, only some. [22:29:46] Yea I'm sure there will be changes to the top area thats fine [22:30:05] kaity: sweet :) [22:36:37] http://etherpad.wikimedia.org/p/MobileWeb-IterationName [22:40:17] I thought we had to stick to real animals [22:40:34] kickassaroo isn't a real animal? [22:42:09] Google Images gives some scary results [22:43:09] lololol [22:43:15] yes it does [22:43:22] I think I'll have nightmares ;) [22:44:28] (03CR) 10Dr0ptp4kt: [C: 04-1] Optimizations for zero config editing (031 comment) [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/97484 (owner: 10Yurik) [22:44:40] (03PS2) 10Cmcmahon: fixing automation tests [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/96962 (owner: 10Mgrover) [22:44:49] hehehe i hope it doesn't interefere too much with you adjusting back to pacific timezone [22:46:10] yurik, would you please clarify the meaning of "btw, one thing that i can't figure out there is why empty saves do not get detected as such" are you talking about the case where the user hits 'submit' without changing the data inside of a pre-existing blob? [22:46:55] awjr: nice. did anyone pitch komodo dragon ? [22:47:09] tfinc it was a close second: http://etherpad.wikimedia.org/p/MobileWeb-IterationName [22:47:29] i knew i should have stuck around to swing the vote [22:47:34] :) [22:47:39] hehehe [22:48:42] dr0ptp4kt, you just found the bug :) [22:48:45] commiting :) [22:49:00] (the bug that was causing it) [22:49:11] boolean logic. [22:51:01] (03PS2) 10Yurik: Optimizations for zero config editing [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/97484 [22:51:20] dr0ptp4kt, more like writing code at 3 am ;) [22:52:30] dr0ptp4kt, and fixed ^ [22:53:24] dr0ptp4kt, also, not sure what you meant by the side note [22:54:48] (03CR) 10Yurik: Optimizations for zero config editing (031 comment) [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/97484 (owner: 10Yurik) [22:56:40] yurik, any old legitimate looking username can be added to the admins field, even if it doesn't exist. not a huge deal, as a misentered _legitimate_ username would be equally problematic, but it opens up the possibility of a non-existing user being added to the list and then someone registering with that id later (granted, flaggedrevs will require oversight of any changes once we're ready for actual user-initiated config edits) [22:57:15] violetto: coffee in 15mins? [22:57:28] jdlrobson: sure [22:57:39] jdlrobson: heads up kaldari tried to +2 this but looks like needs a careful rebase: https://gerrit.wikimedia.org/r/#/c/96397/2 [22:57:39] dr0ptp4kt, are you saying we should add a check if the user exists? [22:58:07] dr0ptp4kt, into the validate() func? [22:59:29] btw, the reason the current admin is not re-added in validate() is because there is no context anywhere - i don't know who the current user is [22:59:50] and i don't want to use yet another global [23:03:06] yurik, yeah, a check for whether the user exists is probably a good idea. yurik, i understand what you mean about the username not being present in the context - clever hack to use getCanonicalName! [23:03:56] yurik, i also noticed that if a user in the 'admins' array is logged in, he can just remove the admins field altogether to get rid of it. do we want to prevent that? [23:04:36] that's not a security risk, per se, but i wonder if that behavior is more consistent with the current behavior where we don't allow him to remove his username from the array of users in the admins field. [23:04:55] dr0ptp4kt, will they be able to do it? Checking... [23:14:42] YuviPanda: I updated the trello card to include specs [23:15:07] Is that good info or unnecessary? [23:15:38] kaity: no I think that's good info! [23:16:11] kaity: I'll probably make the inter-image spacing 8px to match reccomendations at http://developer.android.com/design/style/metrics-grids.html [23:16:38] kaity: mentioning the colors + spacing is always useful! [23:17:37] awjr, yt? [23:17:41] yes MaxSem [23:17:43] what's up? [23:17:55] yurik, yeah, seems to work that way due to the empty array. alright, lemme know once you want me to look at that patchset again. going for a walk around the block. [23:18:22] dr0ptp4kt, i'm adding a few minor tweaks to it - to allow strings to be treated as arrays [23:18:27] checking in shortly [23:18:39] awjr, I remember we discussed requiring candidates to not use any framework, but it's not in task atm - do you support adding it? [23:19:03] oh yeah [23:19:04] hm [23:19:20] are we getting tasks completed with some kind of framework? [23:19:27] MaxSem: also can you link me to the current task? [23:19:31] yup (pm) [23:19:32] kaity: en.m.wikipedia.beta.wmflabs.org/w/index.php?title=Special:Random&campaign=random [23:20:48] jdlrobson: ping [23:21:04] time violetto ? [23:21:11] jdlrobson: now? [23:21:40] violetto: yup [23:21:42] where r u? [23:21:54] jdlrobson: 6th [23:22:00] meet you downstairs now? [23:22:09] K [23:51:25] (03PS1) 10Yurik: Config edit checks and visualizations [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/97651 [23:51:34] dr0ptp4kt, around? [23:51:44] yurik, yup [23:52:03] dr0ptp4kt, so i didn't make any changes to the first patch, instead i made another one based on the first [23:52:22] it checks that admni actually exists [23:54:43] yurik, ok, want me to look at ps 2 of the earlier one, then? [23:54:57] dr0ptp4kt, yep, i guess both are ready to go [23:56:25] yurik, ok, here i go