[00:01:01] ah yeah [00:01:24] awjr, MaxSem, i still don't see "read in another lang" on m, while i see it on desktop & zero [00:01:27] yurik: im not sure i understand the problem you're desribing? [00:01:32] i see it on m and zero [00:01:36] on the main page at least [00:02:01] awjr, testing in FF private mode [00:02:07] oh jeez im on actual wikipedia [00:02:09] one sec [00:02:16] http://en.m.wikipedia.beta.wmflabs.org/wiki/Main_Page [00:02:35] good to know it works on WP :-P [00:03:01] it works in beta mode on beta [00:03:11] and yeah, i see it in stable mode on beta [00:03:17] yurik: ^ [00:03:25] awjr, you see it working ok in all modes? [00:03:51] beta and stable on .m (didnt try experimental) [00:04:05] awjr, i'm in stable (no cookies - using privacy mode) [00:04:13] non-logged in [00:04:21] on firefox [00:04:47] http://en.zero.wikipedia.beta.wmflabs.org/wiki/Main_Page results in 503 for me [00:05:26] awjr, redirects to https then to real wp for me [00:05:38] o_O [00:05:39] try m. [00:06:15] aha, works [00:06:28] m. in private browsing, logged out, stable, in firefox is not fully loading for me [00:06:30] on my machine, both chrome & firefox in PRIVACY mode does not show any languages [00:06:33] DarTar: jgonera jdlrobson[dark] any of you able to come to the AB test meeting? [00:06:47] request to http://meta.wikimedia.beta.wmflabs.org/wiki/Special:BannerRandom?uselang=en&sitename=Wikipedia&project=wikipedia&anonymous=true&bucket=1&country=RU&device=unknown&slot=16 hangs though [00:06:58] kenanwang: first i've heard of said meeting but i'm working on this spike with dr0ptp4kt right now so i'm busy [00:06:59] MaxSem, awjr at this point i am ONLY testing http://en.m.wikipedia.beta.wmflabs.org/wiki/Main_Page [00:07:00] yeah that's what im seeing too [00:07:16] i wonder if that's blocking the js that turns the read in another language section into the blue botton [00:08:22] BUT, it does work if i try http://en.zero.wikipedia.beta.wmflabs.org/wiki/Main_Page?X-CS=250-99 -- IN PRIVACY MODE [00:08:48] if you just click it, you will have the "forceHTTPS" cookie set, which will bounce the whole thing [00:08:53] it looks like centralnotice is getting an exception [00:09:13] that request results in mw.centralNotice.insertBanner( false /* due to internal exception */ ); [00:09:27] gah and now betalbs 503d when i tied to log in [00:09:29] *trie [00:09:30] jebus [00:09:31] tried [00:09:44] i think the moral of the story is that betalabs is currently wonky [00:10:00] i am not convinced the there is a problem with the language button, but i cant be sure [00:11:25] awjr, betalabs has been very consistent for me - but i only use it from privacy mode to avoid any weird cookies [00:12:17] jdlrobson[dark], you sure you can't come? you implemented the AB test we're talking about... [00:12:18] awjr, can it be that you have the header setter still configured? [00:12:32] the little blue icon upper right corner? [00:13:08] no yurik [00:13:54] who's the main dev in fundraising? [00:13:56] yurik when dd you start noticing the issue? [00:14:14] yurik: matt walker i believe has been the one primarily working on centralnotice [00:14:26] awjr, half an hour ago when i started testing the new overlay [00:14:29] thx [00:14:34] i dont have my local instance configured for languages - maybe jgonera can dbl check that things are working ok locally? [00:15:40] but i think that would be a more reliable test atm [00:15:47] i keep getting 503s on betalabs [00:17:07] are you trying m or zero? wonder if its internal network or wifi... it has been wonky at the office lately [00:18:01] both work for me yurik [00:18:17] http://imgur.com/GB1JDcJ,hca0Ds8 [00:20:12] jdlrobson[dark]: did you get what you needed from kenanwang and kaity|away on https://wikimedia.mingle.thoughtworks.com/projects/mobile/cards/1405 ? [00:25:14] MaxSem: should this bug still be open? https://bugzilla.wikimedia.org/show_bug.cgi?id=50530 [00:25:32] it seems like what's still problematic is captured in https://bugzilla.wikimedia.org/show_bug.cgi?id=54616 [00:26:25] nope, we still strip .toc due to parser cache [00:27:41] i thought that was resolved with https://gerrit.wikimedia.org/r/#/c/80578/ ? [00:28:00] that's only part of it [00:28:08] there's parser cache [00:28:22] which forces us to continue removing .toc [00:28:38] >_< [00:29:06] MaxSem: would you mind elaborating on the bug? we discussed it during prioritization today and no one was clear on the current state [00:32:38] also MaxSem [00:32:43] why are you still awake?! [00:33:00] awjr, something is very wrong - i tried this url from an alpha version of chrome - no lang links [00:33:02] en.m.wikipedia.beta.wmflabs.org/wiki/Main_Page [00:33:30] could someone else try it from privacy mode without logging in [00:33:46] yurik: i just tried in private browsing in chrome and i am getting the same problem now [00:33:56] good [00:34:02] jgonera: yt? [00:34:03] finally there is a problem :) [00:34:35] awjr, sorry, was in a meeting [00:34:42] np jgonera [00:34:44] you free now [00:34:45] ? [00:34:57] more or less, depends how much time you need me for [00:34:59] yurik: i get a js error with debug=true but oddly that makes the button show up too :p [00:35:13] hehe :) [00:35:29] jgonera: yurik is reporting that in a fresh browser session, he is not seeing the 'read in another language' button on the main page of enwiki on betalabs (http://en.m.wikipedia.beta.wmflabs.org/wiki/Main_Page) [00:35:32] i can replicate the problem [00:35:50] and this is not zero related ;) [00:35:51] but it seems i have to be in a fresh private browsing session [00:36:14] oddly, i tried with ?debug=true and the button works, but it also results in a js exception :p [00:36:30] which appears to be the results of overlays/cleanup template missing [00:36:56] jgonera: if you have a minute can you dig into it? this would be bad to have go out with tomorrow's deployment [00:37:17] awjr, will try [00:37:23] ty [00:38:09] awjr, the second problem listed there means that it can be closed only when we switch to pageviews via Parsoid:P [00:42:28] (03CR) 10Kaldari: [C: 04-1] Don't handle new pages for files (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100753 (owner: 10AndyRussG) [00:48:52] (03PS2) 10JGonera: Story 1483: Make overlay headers fixed [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100515 [00:51:33] (03CR) 10JGonera: Story 1483: Make overlay headers fixed [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100515 (owner: 10JGonera) [00:51:34] (03PS1) 10Yurik: Made RTL common for the whole page [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/100941 [01:00:06] dr0ptp4kt, ^ [01:00:20] yurik, okay, will review tomorrow [01:01:31] dr0ptp4kt, sure. Its a tiny one (one line removed really). Will depl tmrw as well [01:01:45] i might have found another minor bug [01:23:24] awjr, I can't reproduce this on my local instance and have no idea how to debug it on betalabs if there's no error in JS [01:23:40] jgonera: hmm [01:23:44] did you see the error with ?debug=true? [01:24:16] awjr, yes, but I can't reproduce that error on my local instance either [01:24:32] jgonera: out of curiosity, do you have the zero extension installed? [01:24:37] awjr, no [01:24:49] the zero extension was causing similar JS errors last week [01:24:52] i wonder if it's related [01:24:58] yurik: jgonera ^ [01:25:19] let me try [01:25:36] jgonera, how can i help? [01:26:54] jgonera, btw, you can fairly easily change the code on betalabs, but ppl won't like you :) [01:27:29] they will complain that it should be distributed by puppet only [01:27:39] awjr, yurik I installed zero and it still works on my local instance [01:27:44] yurik, do I need cluster access? [01:27:53] i think so [01:27:59] you need to access beta cluster [01:28:03] not the main one [01:28:15] jgonera, do you have a labs account? [01:28:34] jgonera: huh [01:28:46] ok and you were using a fresh browser instance (eg new private browsing instance) to test? [01:28:54] yurik, maybe, I don't remember... [01:29:02] that was the only way i was to replicate it [01:29:13] awjr, yes [01:29:16] jgonera: ok [01:29:21] btw, one thing you could do is compare the output from the WP and beta [01:29:36] see which files are downloaded, etc [01:29:58] awjr, also, Jon can't reproduce it on his laptop even on betalabs [01:30:15] jgonera: worst case scenario, the problem shows up on test, test2, and mediawiki.org tomorrow and we fix it with a lightning deploy [01:30:17] jgonera, make sure you don't have any cookies - privacy mode [01:30:40] jgonera: thanks for digging into it [01:31:02] jgonera: YuviPanda or MaxSem should be able to help get you set up with betalabs console access if you're interested [01:32:04] honestly, I don't feel like digging into it right now unless you consider it supercritical [01:32:15] awjr, I'd wait for mediawiki.org, I don't trust betalabs [01:32:41] jgonera: i agree [01:32:57] jgonera: max and yuvi better be asleep right now anyway :p [01:36:20] awjr, i could probably add jgonera acct to beta [01:36:49] jgonera, but you need to make sure you have an account on labs, with a valid pub ssh key [01:37:19] yurik, how he's committing, then? [01:37:36] MaxSem is never asleep [01:37:41] it's up to him yurik, he may want it for something in the future, but i agree with jgonera that betalabs is untrustworthy and if it cant be reproduced locally, it should be ok to let it go out with deployment tomrorow [01:37:59] heh, sounds good to me :) [01:38:10] yurik, I vaguely recall that it's non-working time for you too [01:38:14] MaxSem, i don't think betalabs uses the same key storage as gerrit [01:38:31] MaxSem, nah, 8:40pm, as good as any [01:44:36] (03PS5) 10Jdlrobson: Experiment: Add Special:App to alpha to explore usage in Firefox OS [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [01:47:04] (03PS5) 10AndyRussG: Don't handle new pages for files [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100753 [01:48:16] (03CR) 10AndyRussG: Don't handle new pages for files (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100753 (owner: 10AndyRussG) [10:11:40] (03CR) 10Siebrand: [C: 031] "Partial i18n, and L10n reviewed." (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/98745 (owner: 10Kaldari) [10:16:28] (03CR) 10Siebrand: [C: 04-1] "The new special page needs an alias in MobileFrontend.alias.php." [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [16:12:10] (03PS6) 10Dr0ptp4kt: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [17:46:41] YuviPanda: hey! [17:46:46] heya! [17:47:15] saw email about api change being merged! woo! [17:47:46] YuviPanda: when will that deploy? [17:48:04] mhurd: today to testwikis, next thursday to enwiki [17:48:13] YuviPanda: ahhhh! awesome! [17:48:32] :) [17:49:09] YuviPanda: thanks yuvi and MaxSem! [17:49:14] :) [17:53:55] (03CR) 10Dr0ptp4kt: [C: 032] Made RTL common for the whole page [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/100941 (owner: 10Yurik) [17:55:34] (03Merged) 10jenkins-bot: Made RTL common for the whole page [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/100941 (owner: 10Yurik) [18:06:11] (03PS1) 10Yuvipanda: Add Save Page action [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/101035 [18:06:39] mhurd: around? [18:06:47] YuviPanda: ya! [18:07:42] mhurd: re: saved pages online/offline bits [18:07:47] that we talkeda bout yesterday [18:08:01] mhurd: did you and brion have any other conversations about it after I left? [18:10:12] YuviPanda: what did u propose again? (coffee hasn't kicked in yet) [18:11:28] mhurd: heh :D [18:11:50] mhurd: so even with saved pages, when someone asks for a page, and are online, we get the online version (and refresh the cached version) [18:11:56] mhurd: when they are offline we get the cached version [18:12:28] YuviPanda: ah. yes. we said we agreed and thought same behavoir could be nice for history items [18:12:57] mhurd: are you storing the contents of all items in history offline? [18:15:38] YuviPanda: ya both saved pages and history use common store: https://gist.github.com/montehurd/7663582 (<-- updated recently to simplify things a bit) [18:16:18] mhurd: ah, didn't realize that. on Android History *only* stores the page metadata (title, timestamp). Hitting back doesn't do a network request because those are in memory. [18:16:43] mhurd: I guess I now understand your confusion on why saved pages and history are being considered different features [18:18:48] YuviPanda: yeah on ios the webview keeps some amount of content in memory, but because it can clear things out whenever it decides to i'm bypassing its default data store and storing both html and image data in our own store [18:19:16] mhurd: ah, alright. Android swaps it out too, but you can tell it to keep things around in memory 'somewhere', and it'll figure out. [18:19:28] YuviPanda: luckily there's a convienient intercept point [18:19:53] mhurd: right. Android doesn't have that. [18:20:00] or rather it does, but we support old versions [18:20:12] YuviPanda: doh! [18:20:40] the other advantage of re-routing images to our data store is quick access for future fancy interface stuff :) [18:20:58] right [18:21:50] YuviPanda: today i'm updating the core data stuff to have better concurrency support - presently sporadic thread-related lock ups happen :( [18:22:05] heh, yeah, multithreading's a pita [18:22:15] YuviPanda: ya! always with the multi-threading! gah! [18:22:19] :) [18:22:45] YuviPanda: i'm getting good at handing multi-threading issues almost agains my will :) [18:22:51] mhurd: :D [18:23:17] mhurd: I spent 3 years or so doing desktop GUI development, so have dealt with them a bit before. [18:23:33] not that much, but as much as would be required for a GUI app [18:23:58] YuviPanda: it's good practice for sure [18:24:02] yup [18:33:39] dr0ptp4kt, are you joining? [19:02:20] any sign of brion [19:08:27] (03CR) 10Jdlrobson: [C: 04-1] Quick fix for styling issue with Issues overlay in Beta (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100910 (owner: 10Kaldari) [19:12:06] (03CR) 10Jdlrobson: Story 1483: Make overlay headers fixed (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100515 (owner: 10JGonera) [19:16:42] kaity: I haven't actually seen a roadmap [19:17:28] just forwarded [19:17:50] YuviPanda: I think it was sent to mobile-tech but might've missed [19:24:02] kaity: ah, yes, did miss it. Thanks [19:28:08] kaity: we do actually have a navigation drawer on Android for History and Saved pages :D [19:28:29] cool YuviPanda! [19:28:55] I do think though we should spend some time working on how it all fits together [19:29:00] kaity: agreed! [20:03:47] violetto: email them? [20:03:49] or something? [20:04:12] violetto: move it up? [20:06:29] brilliant.... staging hangs after apt-get autoremove and reboot [20:07:37] MaxSem: this is why you should puppetize everything :P So you can wait for a puppet run and then watch it hang [20:09:45] (03PS7) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [20:10:27] (03PS8) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [20:39:02] (03PS2) 10Yuvipanda: Add Save Page action [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/101035 [20:41:14] (03PS1) 10Yuvipanda: Ensure keyboard is always hidden when opening page [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/101066 [21:04:26] yurik: im looking at test2 post-deployment and am not seeing the problem you described yesterday with the read in another language button [21:04:28] can you confirm? [21:05:12] awjr, checking... [21:17:46] dr0ptp4kt, are you doing the VCL? [21:18:48] yurik: i take your silence as confirmation that you're not seeing the issue post-deployment? [21:19:07] awjr, sorry, yes, i don't see any issues so far [21:19:11] \o/ [21:19:14] that's great, thanks yurik [21:19:24] strang that beta showed them :( [21:19:37] eh, strange things happen on beta :) [21:21:46] mhurd: around? [21:21:57] YuviPanda: ya! [21:22:05] mhurd: :D [21:22:21] mhurd: so from what I understood, what you talked about was essentially a visual ToC [21:22:24] is that accurate? [21:23:05] yes, the reason i brought it up is it more or less fulfulls the requirements of that card we were chatting about [21:23:29] wan't me to step through it? [21:23:34] *want* [21:25:41] mhurd: yeah! [21:26:02] mhurd: and yeah, I think the use case there is a little confused (for the card) - visually navigate an article, or see all the associated images? [21:26:11] so clarity would be nice :) [21:26:14] both [21:26:17] mhurd: I'm also going to try to sleep in 30mins tho [21:26:34] so the idea was... [21:26:53] have quick way to present sections [21:26:56] YuviPanda: did you ever get the network failure design? [21:27:05] exposed by some gesture [21:27:11] jdlrobson: nope [21:27:20] YuviPanda: let me know when you do cos i need it as well :) [21:27:34] jdlrobson: will do :) I'll bump that thread [21:27:40] having image data presented with section titles gives nice context [21:27:44] thanks YuviPanda :) [21:27:52] mhurd: right. [21:28:12] but the section interface *floats* on top of the full web view and is partially transparent [21:28:42] mhurd: fully covering it? [21:28:46] so you can see the web view beneath - the reason being - as you scroll the TOC the middle TOC entry gets highlighted [21:28:51] (yes fully) [21:28:59] mhurd: also I'll note that translucency is kind of a no-go on Android [21:29:11] when that middle section gets highlighted it scrolls the web view beneath to that section [21:29:27] so as you scroll TOC you are navigating full html page [21:29:52] mhurd: right. [21:30:05] the transparency gives you that hint - although w/o transparency same effect could be achieved by having the TOC inset a bit [21:30:15] mhurd: hmm, so scrolling the overlay will scroll the underlying view as well, but at a much faster pace [21:30:15] so you could see around it some i suppose [21:30:19] yes [21:30:39] mhurd: so if we don't do transparancey and the phone is on portrait, it doesn't give us much space to do either [21:30:45] it will instruct the embedded web view to scroll to named anchor for that section [21:31:28] yes on landscape on small screen device the nav is less useful, but on tablets it will be nice [21:31:39] i think that will be the case for most nav schemes [21:31:52] so not necessarily a show stopper :) [21:31:53] how will you show it on portrait without transparency? [21:32:12] so no transparency effect on android? at all? [21:32:46] mhurd: it's expensive to do (not GPU'd), and even then you only get alpha channel transparency [21:33:53] mhurd: and with alpha channel transparency, visibility of things is a problem. If you make it too light you can't read what's on it nor what's underneath, and if you make it too dark you don't see what's underneath [21:33:57] ah i see. it would still work from a nav standpoint, you just would have one less visual clue, but as a nav interface making the selection can still dismiss the TOC interface *then* cause the web view to scroll to section selection [21:34:07] oh yeah, that completely works :) [21:34:10] and I quite like it [21:34:25] YuviPanda: true the alpha will need to be tuned pretty carefully visually [21:34:45] mhurd: yeah, but I think just having it be solid and tapping it taking you back to appropriate section on webview should be fine [21:34:54] i can give it a go iOS and if it just doesn't work or is too busy we'll still have option to use it opaque [21:35:07] ya [21:36:30] mhurd: right. so the sections will have some context around them (img thumbnails) rather than be just a pure ToC [21:36:37] yes [21:36:58] if a section has multiple images could present them as a little stack [21:37:30] tapping TOC image (or stack) would present full screen image [21:37:48] right [21:38:12] i'd prefer the full screen stack be vertically scrolling but that's just a personal preference [21:38:30] as long as the entire ToC can fit inside about 2screens... [21:39:03] the TOC text and thumbs can be as small as we need them to be i guess [21:39:18] within reason :) [21:39:41] i like the 2 screens target [21:39:53] mhurd: :D [21:39:57] means TOC browsable with a single vertical flick [21:40:16] yup [21:40:37] i think for most articles this should be the case [21:40:49] a few will have a bit more but that's ok [21:41:27] *can't wait to implement* [21:41:36] mhurd: :) [21:47:47] (03PS9) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [21:48:44] (03CR) 10Yuvipanda: [C: 04-1] "Can you use Special:WebApp? There's Extension:MobileApp that would serve styles and do other odd serverside stuff for the apps, and I want" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [21:48:49] (03CR) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [21:54:28] (03PS1) 10Yuvipanda: Draw drawer indicator icon next to the 'W' icon [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/101114 [21:55:06] mhurd: I'm going to sleep in about 5 minutes. Do let brion know he has a bunch of android patches waiting too, when he turns up :) [21:55:20] will do! sleep well! [21:55:30] yo YuviPanda [21:55:43] hey aharoni [21:55:44] 'sup [21:55:47] yurik: Zero-rated-mobile-access-home-page-selection - is the $1 there always the same language in which the message is shown? [21:56:36] aharoni, i think so - its the name of the language [21:56:55] (03PS1) 10Yuvipanda: Request only the sections we really want [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/101116 [21:56:57] aharoni: ah, I guess you wanted just yurik not me :P [21:57:23] it was not a mistake YuviPanda ! [21:57:32] mhurd: you should look at https://gerrit.wikimedia.org/r/#/c/101116/ too - the MobileView patch for requestedsectionsonly that went in last week has been deployed now, and would cut down transfer time at least a little [21:57:36] aharoni: :D ok! [21:57:41] yurik: so why does it have to be a parameter? [21:57:59] YuviPanda: no way! that's awesome! [21:58:14] mhurd: :) [21:58:26] YuviPanda: thanks for the head's-up! [21:58:34] mhurd: yw! [21:58:49] aharoni, i guess this way people don't have to re-type the language name? [21:58:52] yurik: if it is known that it points to a particular language, and it's supposed to be written in the same language, then it should just be translated literally. [21:59:12] also i think one carrier wanted to override the name of the language for some reasons [21:59:34] in Russian it will be "русский", but you actually want "по-русски" there. [21:59:42] and it's true for many other languages. [21:59:57] is it taken from Names.php? [22:00:15] if it has to be a parameter for carriers, then it can insert another message. [22:03:07] (03PS1) 10JGonera: Hygiene: Remove button.cancel in overlay header [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101121 [22:05:03] awjr, regarding the diff bug you reported yesterday, can you look at the diff HTML and tell me if each of these lines corresponds to a separate div? [22:05:52] MaxSem: sure - one min [22:09:24] aharoni, i guess we could come up with a different prompt there - what do you think should it be? [22:09:48] aharoni, it uses Language::fetchLanguageNames() [22:10:00] MaxSehttps://gist.github.com/awjrichards/4afe11b5987c16a223e2 [22:10:16] er [22:10:17] https://gist.github.com/awjrichards/4afe11b5987c16a223e2 [22:10:24] MaxSem: ^ [22:10:32] aharoni, it uses wfMessage( 'zero-rated-mobile-access-home-page-selection', ucfirst( $languageNames[$languageCode] ) )->inLanguage( $languageCode ) [22:10:52] (03PS10) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [22:10:53] (03PS1) 10Jdlrobson: Cache last call to PageApi in localStorage [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101123 [22:10:54] looks like no MaxSem, just each line wrapped in [22:11:17] (03PS11) 10Jdlrobson: Experiment: Special:App and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 [22:11:28] (03PS2) 10Jdlrobson: Cache last call to PageApi in localStorage [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101123 [22:12:28] awjr, eek [22:12:37] (03CR) 10jenkins-bot: [V: 04-1] Cache last call to PageApi in localStorage [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101123 (owner: 10Jdlrobson) [22:12:41] :( [22:13:02] hahahaha my dog seems to have just figured out how to play catch with herself [22:14:42] (03CR) 10jenkins-bot: [V: 04-1] Cache last call to PageApi in localStorage [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101123 (owner: 10Jdlrobson) [22:14:44] yurik: I'd just tell the translators to say that they are supposed their language name there and not as a parameter. [22:15:46] aharoni, ??? [22:15:57] supposed ^to write^ [22:16:10] (I miss a lot of words today for some reason) [22:16:55] (03PS1) 10Jdlrobson: Allow right clicking on images to avoid previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101124 [22:17:46] yurik: make it "English home page" in English, and in qqq say "replace 'English' with the name of your language". [22:18:30] mhurd: popped back in to say - think you can write up what we just discussed into an email to mobile-tech? [22:18:39] ya! [22:18:40] mhurd: would be easier to keep the conversation on [22:18:44] mhurd: sweeto! Thanks [22:18:51] welcome! [22:18:57] awjr, http://staging.wmflabs.org/wiki/Special:MobileDiff/12146...12147?useformat=mobile&mobileaction=beta [22:19:04] (03PS3) 10Jdlrobson: Cache last call to PageApi in localStorage [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101123 [22:19:21] aharoni, ah, gotcha. This is somewhat problematic though - there are some uncommon languages that will take a long time to translate - and they would show up incorrectly as "english" [22:19:29] MaxSem: is that with the latest patchset or have you made changes? [22:19:52] no, it's of a few days ago - I can't log into that instance now [22:20:00] aharoni, i would have to check if its the same as english, and fallback to the language name... [22:20:01] I am aware of a few such messages and I check their translations regularly. [22:20:16] aharoni, yes, that's you :) [22:20:21] doh [22:20:24] most other translators are not as diligent :) [22:20:34] I care about all extensions, not just those that I develop :) [22:20:56] awjr, but I have the same behavior locally.... [22:21:16] oh [22:21:18] strange [22:21:33] MaxSem: im helping milimetric out right now - i can take a look again in a sec [22:22:04] np [22:22:08] (03PS15) 10MaxSem: Inline diffs [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97871 [22:22:15] ^^ is not relarted:) [22:23:22] grr why do i always have trouble adding people to tools labs projects [22:23:37] MaxSem, YuviPanda can you help me ad milimetric to the bingle tools labs project? [22:23:59] awjr: heading to sleep now, sorry. MaxSem should be able to do it [22:24:05] trying to sleep before 4am :D [22:24:05] np YuviPanda [22:24:09] night [22:24:10] have no idea how to:) [22:24:11] good :) [22:24:15] MaxSem: congrats :) [22:24:26] thanks tfinc [22:24:37] yurik: ehh... Now I see that the message that I was about to give as an example actually has $1 now: [22:24:39] https://translatewiki.net/w/i.php?title=Special:Translations&message=MediaWiki%3AJquery-uls-ext-uls-display-settings-anon-same-as-content%2Fhe [22:24:45] (03CR) 10jenkins-bot: [V: 04-1] Inline diffs [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97871 (owner: 10MaxSem) [22:24:56] hehe :) [22:25:05] BUT, in the case of that message it's OK to have the simple name of the language. [22:25:27] In the case of this message you need something with a... падеж, предлог, и т.п. :) [22:25:31] grammar stuff. [22:25:35] awjr, how can I do this? [22:25:58] aharoni, btw, that page is an excellent example that dir= should not dictate alignment [22:26:12] MaxSem: nm i actually just figured it out [22:26:16] thanks though :) [22:26:27] thanks from me as well, to both [22:26:28] yurik: it isn't it. It's very comfortable for me. [22:27:37] aharoni, i don't think its a good idea to have empty string until much later - everything should have the same alignment, depending on the main language of the page [22:27:50] looks confusing, esp on wide monitors [22:27:58] a small laptop screen might be ok of course [22:28:25] Maybe. I read in an RTL language and I find it perfectly comfortable. [22:29:08]