[11:30:03] (03CR) 10JGonera: ""for instance it doesn't tell me what to do with the last modified bar"" (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103618 (owner: 10Jdlrobson) [12:10:35] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [13:47:14] (03CR) 10JGonera: [C: 04-1] "Just a minor concern." (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103948 (owner: 10Jdlrobson) [13:51:00] (03CR) 10JGonera: [C: 031] Replace usage of SpecialPage::getTitle with getPageTitle [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103953 (owner: 10Legoktm) [13:54:37] (03CR) 10JGonera: [C: 04-1] "What about EditorApi?" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103887 (owner: 10Kaldari) [14:02:27] (03CR) 10JGonera: [C: 04-1] "Two minor things, other than that looks OK." (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103570 (owner: 10Jdlrobson) [14:07:00] (03CR) 10JGonera: [C: 031] "Fewer files, less confusion." (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/102853 (owner: 10Jdlrobson) [14:09:44] (03CR) 10JGonera: [C: 04-1] Remove tap highlight on section headings (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103767 (owner: 10Jdlrobson) [14:10:05] (03PS7) 10JGonera: Remove tap highlight on section headings [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103767 (owner: 10Jdlrobson) [15:30:18] (03CR) 10Siebrand: "i18n reviewed. Left inline comments." (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/102043 (owner: 10Jdlrobson) [16:01:50] (03CR) 10Brion VIBBER: [C: 032 V: 032] Make it possible for results to be read multiple times [apps/android/java-mwapi] - 10https://gerrit.wikimedia.org/r/103891 (owner: 10Yuvipanda) [16:03:32] (03CR) 10Parent5446: Tool: Blow Kaldari's code review mind (034 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103884 (owner: 10Jdlrobson) [16:14:23] (03CR) 10Brion VIBBER: "Ah that explains it :D" [apps/android/wikipedia] - 10https://gerrit.wikimedia.org/r/103626 (owner: 10Yuvipanda) [16:19:46] jgonera, what's the difference between modules/editor and modules/editorNew? [16:20:30] MaxSem, editor is the old overlay design + common files such as api abstraction, editorNew is what we have only in beta/alpha, i.e. new overlay design and visual editor [16:21:05] in near future I hope they'll both get merged (when we push new overlays to stable) [16:21:22] aka new will become the only [16:23:47] yes, plus the common stuff from old (api abstraction) [16:42:39] (03PS1) 10MaxSem: Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 [16:43:28] (03CR) 10jenkins-bot: [V: 04-1] Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 (owner: 10MaxSem) [16:43:38] dafuuq? [16:44:40] MaxSem, onevar [16:44:54] I'm not a fan of it, but that's what we use [16:45:01] stupid conventions of a stupid language [16:45:06] ;) [16:45:20] it's arguably less stupid than PHP [16:45:31] (03PS2) 10MaxSem: Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 [16:46:00] even if it is, the margin is really small;) [16:46:27] (03CR) 10jenkins-bot: [V: 04-1] Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 (owner: 10MaxSem) [16:47:27] (03PS3) 10MaxSem: Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 [17:32:01] (03CR) 10JGonera: [C: 04-1] "Happy to merge when small issues from OverlayNew.less are resolved." (034 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101428 (owner: 10Jdlrobson) [17:37:57] (03CR) 10Jdlrobson: [C: 04-1] "-1ing till I hear back from @Hooman to push down the queue (also this patch needs a rebase)" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/102004 (owner: 10Hoo man) [17:39:33] (03CR) 10JGonera: [C: 04-1] "Very small nitpicks. Also (just suggestions, not required for this to be merged):" (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103854 (owner: 10Jdlrobson) [17:40:20] (03PS2) 10Jdlrobson: Replace usage of SpecialPage::getTitle with getPageTitle [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103953 (owner: 10Legoktm) [17:41:27] (03CR) 10Jdlrobson: [C: 032] Hygiene: Replace usage of SpecialPage::getTitle with getPageTitle [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103953 (owner: 10Legoktm) [17:42:01] (03Merged) 10jenkins-bot: Hygiene: Replace usage of SpecialPage::getTitle with getPageTitle [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103953 (owner: 10Legoktm) [17:43:31] (03CR) 10MaxSem: [C: 032] Replace usage of SpecialPage::getTitle with getPageTitle [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/104007 (owner: 10Legoktm) [17:43:49] (03Merged) 10jenkins-bot: Replace usage of SpecialPage::getTitle with getPageTitle [extensions/ZeroRatedMobileAccess] - 10https://gerrit.wikimedia.org/r/104007 (owner: 10Legoktm) [17:44:38] (03CR) 10MaxSem: Tablet detection support (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/92036 (owner: 10MaxSem) [17:46:34] (03PS3) 10MaxSem: Tablet detection support [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/92036 [17:47:41] (03CR) 10Jdlrobson: [C: 032] Use main page mode when appropriate for previews (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 (owner: 10MaxSem) [17:48:16] (03Merged) 10jenkins-bot: Use main page mode when appropriate for previews [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104052 (owner: 10MaxSem) [17:53:14] (03CR) 10Jdlrobson: "Thanks for the code reviews - I just knocked this up with not much thought about code quality but I'll make sure it gets polished up as su" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103884 (owner: 10Jdlrobson) [17:59:16] (03CR) 10JGonera: [C: 04-1] "Thanks for working on this, looks interesting. Comments inline." (034 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [18:16:01] hey violetto are you around? [18:17:10] jdlrobson, kaldari, MaxSem do you need anything from me today? [18:17:29] jgonera, I wanted to discuss templates [18:17:29] nope [18:17:40] although any +2s you have left would be appreciated ;-) [18:19:15] jgonera: are you working on the header update? [18:19:21] jdlrobson, jgonera, kaldari - any comments on templates? any questions why I chose Twig?:) [18:19:25] kaity, jdlrobson is working on it [18:19:46] cool thanks [18:19:48] jdlrobson, there's a few patches close to a +2, but they require small fixes [18:19:52] MaxSem: did you see the email about Swig? [18:20:00] yep [18:20:09] weapons race escalates [18:20:09] MaxSem, are the results somewhere? [18:20:39] jgonera, I emailed the code and performance results [18:21:01] jdlrobson, I'll give your gerrit script a try on Monday ;) [18:21:18] MaxSem, where to? [18:21:23] jdlrobson: moiz and I will update that header spec today to include a search icon [18:21:28] thanks kaity [18:21:34] jgonera: you'll love it :D [18:21:44] MaxSem: I only saw the results for Twig. Have you also profiled any PHP implementations of Mustache? [18:21:56] jdlrobson: we're also proposing to replace just the number for notifications with a bell icon + number [18:22:09] also, kaity, I guess the hamburger icon changed slightly too, can you provide an SVG with that icon too? [18:22:21] jgonera: sure [18:22:46] MaxSem, can't see it on mobile-tech or inbox [18:23:23] jgonera, "Preliminary research results: Twig" [18:25:43] MaxSem, ok, I see it [18:25:57] We should probably profile one of the PHP implementations of Mustache as well for comparison. [18:26:01] so, MaxSem, jdlrobson, kaldari should we do a similar POC with mustache? [18:26:04] right [18:28:51] MaxSem: any thoughts on that? [18:28:51] kaity: a change to echo was not a part of the original spec… you know how i hate surprises ;-) [18:29:04] jgonera: MaxSem yeh a similar POC with mustache would be great [18:29:17] that's why I asked "any questions why I chose Twig?" [18:29:19] ;) [18:29:28] would give a great comparison [18:29:35] (03PS28) 10Dr0ptp4kt: Experiment: Special:MobileWebApp and manifests for offline FF & Chrome, FFOS + CWS. [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [18:30:15] jdlrobson: true! you can say no :) [18:30:30] Mustache looks more limited, and even though in theory simple is good this doesn't work in practice for a whaladon like MW [18:31:44] jdlrobson: but changing search is also kind of a big change [18:31:49] In what ways is it more limited? It would be good to list these in the RfC [18:32:10] moizsyed: you should have an invite to mobile-tech now [18:32:11] jdlrobson: should we push changing search and echo to another iteration? [18:32:30] jdlrobson: thanks! [18:33:37] Tyler mention mustache being limited as well but didn't offer any specifics. Obviously any templating language is going to be limited and we don't necesarily want the one that is least limited :) [18:34:47] kaldari, actually it's in the first line of http://mustache.github.io/ :P [18:34:48] MaxSem, yes, it is more limited, but there's no Twig for JS, or is there? [18:35:03] Swig, apparently [18:35:06] jgonera, https://github.com/justjohn/twig.js/ [18:35:25] 72KB, minified... [18:35:34] that's over 7x more than what we have right now [18:36:33] MaxSem: well, mustache isn't technically "logic-less". I just want to know what are the actual differences. [18:36:50] and which of those differences are actually needed [18:36:57] SWIG is 35KB, not as horrible but still more [18:37:35] kaldari, imagine you have to split a template in 5 parts because it doesn't have {{ if }} and then glue them together in PHP [18:37:56] true, I agree we need if/else + foreach + escpaing. What else? [18:38:01] yeah, unless there's something in Twig that somebody really needs, I'd rather follow the "less is more" mantra [18:38:14] MaxSem, mustache has if, a primitive one but it does [18:39:29] kaldari, extensibility - without having an integration with MW i18n it's gonna look awful [18:39:59] that's a good point [18:40:22] any other ideas for requirements? [18:40:29] MaxSem, that's true, but why would Mustache not be extensible with i18n? it has lambdas [18:40:54] I'm just adding to requirements [18:41:05] I think we should list all the requirements and "could haves" [18:41:16] and benchmark mustache too [18:41:25] add JS version size as one of the arguments [18:41:33] do we have a wiki page for it already? [18:44:59] greg-g: https://office.wikimedia.org/wiki/User:Hfung/Product_Manager_Biweekly_Meeting [18:46:09] MaxSem, jgonera, jdlrobson: I've started listing the requirements and nice-to-haves here: https://www.mediawiki.org/wiki/Requests_for_comment/HTML_templating_library#Templating_library_options_and_considerations [18:51:46] (03CR) 10Siebrand: "It's completely unknown to me what the change is that has been made between version 27 and 28. I see it rebases and changes in here all th" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [18:52:15] (03PS1) 10Jdlrobson: Hygiene: Gerrit tool [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104122 [18:53:27] (03CR) 10Jdlrobson: "Siebrand this patch is highly experimental and will continue to change until we have some sort of stability. It might be worth removing yo" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/100919 (owner: 10Jdlrobson) [18:53:31] MaxSem, jgonera, jdlrobson: http://bits.shutterstock.com/2013/03/07/mustache-vs-swig-templating-shootout/ [18:54:11] kaldari, thanks, will read up on it later [18:54:15] i agree that less is more is good [18:54:19] greg-g: sorry about that curve ball :) [18:54:30] templates shouldn't need to be as powerful as jinja [18:54:40] you just end up writing code in a template language which is far worse [18:54:45] greg-g: i am serious that we need to understand who the customer/s of this are. past ops i've never really heard a compelling case [18:55:13] OK, moving on. I have some questions about bug 58979 (goecoordinates)... [18:56:50] MaxSem, jdlrobson: Since it looks like the GeoData extension that Max wrote actually stores geodata in the database and makes the data available via an API, why would Nearby be dependant on a wiki template's HTML output? [18:57:19] Sorry I don't know the history here [19:00:25] kaldari, we invoke geonotahack upon seeing coordinates in the upper right corner, not actual GeoData [19:00:28] tfinc: right channel? i don't see any greg-g / tfinc conversation here? [19:00:45] i can tab complete and see greg-g :) [19:01:11] there's a difference between these: sometimes people don't want title coordinates [19:01:13] MaxSem: ah, because otherwise we would have to make an API request on every page load to see if it had geodata or not [19:01:39] (03CR) 10Jdlrobson: "Follow up in https://gerrit.wikimedia.org/r/104122" (036 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103884 (owner: 10Jdlrobson) [19:01:46] jdlrobson: backchannel from a hangout meeting ;) [19:02:05] kaldari, in principle nothing stops us from sticking primary coordinate ffrom GeoData into page HTML as a data value [19:02:18] MaxSem: Yeah, that's what I was thinking [19:02:50] or even just a simple global variable flag [19:02:51] tfinc: yeah, my working definition of the customer(s) is/are: myself, erik, other decision makers. The more we share what we're doing, the more we find out that we need to make changes to what we're doing based on what other's in those roles think. [19:02:59] * greg-g is tearing down the walls [19:03:01] ;) [19:04:47] awjr, MaxSem, jdlrobson: so card 1583 sounds like it is actually a story rather than a bug. Would anyone object to me creating a story for it? [19:04:57] https://wikimedia.mingle.thoughtworks.com/projects/mobile/cards/1583 [19:05:41] nope [19:05:50] kaldari: sure thing - create a story and whack it in the backlog [19:06:02] i wouldn't say it's a high priority right now [19:11:11] (03PS8) 10Jdlrobson: Update page previews and geonotahack to use new overlays [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101428 [19:13:57] (03PS9) 10Jdlrobson: Update page previews and geonotahack to use new overlays [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101428 [19:14:29] (03CR) 10Jdlrobson: "Remember this is all in beta and none of the associated styles need to be finalised or even promoted to stable." (033 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/101428 (owner: 10Jdlrobson) [19:15:28] new story created: https://wikimedia.mingle.thoughtworks.com/projects/mobile/cards/1586 [19:20:01] (03PS2) 10Jdlrobson: In alpha mode, save and restore the set of open sections in the history [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:20:29] (03CR) 10Jdlrobson: "Thanks for the patch but yeh a few issues. I've added the relevant bug number to this!" (033 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:20:56] jdlrobson: technically mustache also supports filters through it's lambda support [19:22:21] (03CR) 10Jdlrobson: In alpha mode, save and restore the set of open sections in the history (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:22:29] (03CR) 10Jdlrobson: [C: 031] "See comments on PS1" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:22:43] (03CR) 10Jdlrobson: [C: 04-1] "Whoops that should have been a -1 :)" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:22:57] kaldari: it doessss?! oh nos [19:23:21] jdlrobson: although I don't think we've ever used it. jgonera might know [19:24:59] (03CR) 10Jdlrobson: [C: 032] Remove tap highlight on section headings [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103767 (owner: 10Jdlrobson) [19:25:56] so MaxSem when do you want to talk SkinMobile / SkinMinerva >? [19:26:21] yeah. kaldari, have you seen it? [19:26:34] (03PS3) 10Jdlrobson: Update Talk overlay to use new overlays [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103854 [19:28:33] MaxSem: The SkinMinerva changes? Yes, but haven't looked at it extensively. [19:29:19] I still need to review it [19:29:41] my problem is that the mess just changes its shape but doesn't go away - now instead of overcomplicated inheritance there's if ( #isMobile ) everywhere [19:29:42] kaldari: would help to get your view on whether we should take that approach or not [19:30:22] MaxSem: What if there was one single mobile function that only runs if isMobile ? [19:30:39] I would looove to see the number of differences between mobile reduced instead:) [19:31:10] Even though it is still a bit messy, it seems more in line with the general architecture of how skins are supposed to work and be used. [19:31:21] MaxSem: I'd love that too, but I think some are inevitable for example will section collapsing ever be in core? [19:31:54] well, minerva itself isn't going in core either:) [19:32:19] but MaxSem it's a skin that could be supported by core [19:32:26] and I think we all agree that we want to move in the direction of the mobile interface being part of a skin implementation, right? [19:32:34] also MaxSem I feel like it's acceptable to have a different skin for a mobile vs desktop site [19:32:36] yep [19:32:38] or more so, I mean [19:32:45] yep was to kaldari [19:32:52] the fact we have one file might encourage us to kill isMobileMode stuff more [19:33:00] for instance I have no clue why footer needs to be different [19:33:46] kaity: hey! [19:33:57] hey monte [19:34:11] kaity: are you remote today? [19:34:23] mhurd: yes Im in Ohio! [19:34:28] kaity: whoa! [19:35:28] kaity: time to do a hangout later for table of contents stuff? [19:35:31] (03PS4) 10Jdlrobson: Update Talk overlay to use new overlays [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103854 [19:35:46] (03CR) 10Jdlrobson: Update Talk overlay to use new overlays (032 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103854 (owner: 10Jdlrobson) [19:36:18] yea gimme 10 minues mhurd [19:36:25] kaity: awesome! [19:37:53] kaldari: grr i now have your "You are about to submit multiple commits" bug [19:38:15] git checkout master / git checkout -b foo / git review / "You are about to submit multiple commits" < there should be nothing to review!! [19:38:46] (03CR) 10jenkins-bot: [V: 04-1] In alpha mode, save and restore the set of open sections in the history [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103845 (owner: 10Pcc) [19:39:05] jdlrobson: yeah, I've tried everything to get rid of that, but haven't figured out the problem. Seems to be a bug with git review. [19:39:23] i'm gonna ask in wikimedia-dev [19:39:37] (03Merged) 10jenkins-bot: Remove tap highlight on section headings [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103767 (owner: 10Jdlrobson) [19:41:53] mhurd: I have 20 mins before a meeting, is that enough you think? [19:43:20] MaxSem: kaldari Sooo.. SkinMobile / SkinMinerva what's the way forward? [19:43:54] JeffHall: chrismcmahon what's the state of https://gerrit.wikimedia.org/r/#/c/97497/ ? [19:45:33] kaity: would you have a bit of time after> [19:45:34] ? [19:46:15] mhurd: yes Im free at 12:45 or so [19:46:25] kaity: that works! :) [19:47:01] jdlrobson: I think this is probably in the right direction. I'll review it further today. [19:47:08] mhurd: do you have a current mockup [19:47:50] jdlrobson: I know we're working on the FIXME aspects of that, but we probably won't do much right away with Zeljko and Antoine both on holiday [19:49:14] kaity: i have my orig mocks, and may's iteration, but today i sketched some revisions which would be fun to chat about [19:49:35] mh [19:49:48] whos mhurd? [19:49:52] mhurd: cool! can you send me may's iteration? I haven't seen [19:50:08] thats Monte! [19:50:11] monteeeeeeeee [19:50:20] sorry, im still a noob [19:50:36] hehe ;) [19:51:41] kaity: ya i'll take snapshot of my stuff and get may's for you by 12:45 [19:51:48] thanks [19:53:27] jdlrobson, I don't like it but defer to others [19:54:07] MaxSem: okay thanks :) [19:54:25] or, rather, I see no improvement [19:54:52] (03PS2) 10Jdlrobson: Update wgPageName on refresh [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103948 [19:55:27] kaldari: run git fetch --all [19:55:35] that fixes the review issue [19:59:25] Hi, I'm a GCI student, and I was wondering which part of MobileFrontend handles non-existentent files (e.g. https://en.m.wikipedia.org/wiki/File:Foozajashsaash.png) [20:03:05] (03PS9) 10Jdlrobson: Hygiene: LESS file reorg (killing enwp.less) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103570 [20:03:18] (03PS10) 10Jdlrobson: Hygiene: LESS file reorg (killing enwp.less) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103570 [20:06:05] (03CR) 10Jdlrobson: i18n: consistency tweaks (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103643 (owner: 10Shirayuki) [20:09:11] jdlrobson: https://gerrit.wikimedia.org/r/#/c/104128/ is the core patch to get language variants [20:13:34] jcf2000: well the nonexistent page message is coming from includes/skins/SkinMobileBeta.php [20:13:37] handleNewPages() [20:15:09] hey jcf2000 - have you tried finding out yourself?:) [20:29:08] (03CR) 10Jdlrobson: [C: 04-1] "Thanks for this - looking a bit closer to what can work." (036 comments) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/74922 (owner: 10Pcc) [20:35:20] (03CR) 10Jdlrobson: [C: 032] Tablet detection support [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/92036 (owner: 10MaxSem) [20:35:56] (03Merged) 10jenkins-bot: Tablet detection support [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/92036 (owner: 10MaxSem) [20:36:29] (03CR) 10Jdlrobson: [C: 04-1] "chrismcmahon: jdlrobson: I know we're working on the FIXME aspects of that, but we probably won't do much right away with Zeljko and Antoi" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/97497 (owner: 10Hashar) [20:36:48] dr0ptp4kt: yurik care to merge https://gerrit.wikimedia.org/r/#/c/103658/ ? [20:40:02] (03CR) 10Jdlrobson: [C: 04-1] "grep getToken . -R" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103887 (owner: 10Kaldari) [20:41:08] (03CR) 10Jdlrobson: "Waiting on Kaldari's thoughts..." (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/102853 (owner: 10Jdlrobson) [20:42:11] jdlrobson, hang on, will test [20:42:45] * jdlrobson grabbing lunch [20:51:49] MaxSem: Yes, I did try to find it myself [20:52:36] theopolisme: Thanks [20:56:43] (03CR) 10Dr0ptp4kt: "See SkinMobile.php question." (031 comment) [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103658 (owner: 10MaxSem) [20:57:42] lol. [20:59:31] hiya, I need to see a beta-labs log. chrismcmalunch said they're available on deployment-bastion:/data/project/logs , but that host is unknown (and I probably don't have rights). I'm in the deployment-prep project but not an admin of it [21:04:38] ^ignore that, sorry [21:10:20] (03CR) 10Brion VIBBER: "A good start! A few quick notes on things that need tweaking:" [apps/ios/wikipedia] - 10https://gerrit.wikimedia.org/r/103858 (owner: 10Mhurd) [21:10:38] (03CR) 10Brion VIBBER: "Love the googly eyes on the debug dump menu item ;)" [apps/ios/wikipedia] - 10https://gerrit.wikimedia.org/r/103858 (owner: 10Mhurd) [21:19:09] MaxSem: are you following https://gerrit.wikimedia.org/r/#/c/104128/1 ? [21:19:28] now i do [21:22:01] jdlrobson: thanks, having someone other than the patch author comment is helpful [21:22:20] brion: woohoo! [21:22:40] kaity: chatted with jared! good for now :) [21:22:53] mhurd: ok cool [21:23:09] could you still send what you were working on? [21:23:23] it sounds interesting! [21:23:32] theo|away: i'm hoping MaxSem will have more useful input [21:25:08] kaity: i'll condense the sketches down. should be a fun goal for a first pass at TOC [21:25:52] i'll be home this evening and back in the office next week [21:26:06] see ya guys! will review a little more over the weekend and start poking more android as well :D [21:26:18] brion: awesome! thanks again! :) [21:27:44] kaity: it may actually take me less time if i just implement a first TOC pass and send screenshots... that ok? [21:28:38] jdlrobson: heh, great. [21:28:56] So I guess that's a no-go... [21:29:46] sigh welcome to the wonderful world of developing for MediaWiki core. [21:30:26] :p well I only wasted a few hours of my life on that.. [21:31:21] oh well, ugh... [21:32:41] theo|away: i'm interested to see Max's input [21:32:58] I still think there is an issue here - we can't truly be expected to do 3 ajax requests to construct a language list containing variants [21:32:58] Max says that anomie have solved this [21:33:45] Anomie's patch seems to make sense. So now we just need to fetch that info as well and then do some parsing on the client side [21:33:51] https://sr.wikipedia.org/w/api.php?format=jsonfm&action=query&meta=siteinfo&siprop=general&titles=%D0%92%D0%B8%D0%BA%D0%B8%D0%BF%D0%B5%D0%B4%D0%B8%D1%98%D0%B0&prop=langlinks < MaxSem this is the nastiest api response I have ever seen :) [21:34:02] sorry for not replying sooner, was busy benchmarking [21:34:22] * jdlrobson hates the API [21:34:31] every time an API response is shown to someone a fairy dies [21:34:36] Well, no one will have to look at it I guess -.- [21:35:29] * MaxSem roasts a bunch of jdlrobson's dead fairies and eats them [21:35:59] theo|away: I guess how you can help here then is to alter how we fetch languages using this horrible API request [21:36:07] if you're not completely demotivated by this [21:36:46] MaxSem: where are languages in this response? https://sr.wikipedia.org/w/api.php?format=jsonfm&action=query&meta=siteinfo&siprop=general&titles=San%20Francisco&prop=langlinks [21:37:18] "variants": [ { "code": "sr" [21:37:21] etc [21:37:40] Brad's https://gerrit.wikimedia.org/r/#/c/104140/ will andd variant names to codes [21:38:12] MaxSem: we need a full comprehensive list of languages - is there a way to override the limit of the response? [21:38:18] we really need sensible api defaults [21:38:47] Fairly demotivating yeah, but i guess that's open source for ya -- it's a mixture. [21:39:10] http://sherman.blogs.com/snapshots/images/2008/05/15/dsc_0092_blog.jpg < this is what i see when i look at the api [21:39:48] lunch! brb [21:39:54] Client side we'll aslo need to strip out the one that has a code == lang [21:40:33] you mean https://sr.wikipedia.org/w/api.php?format=jsonfm&action=query&meta=siteinfo&siprop=general&titles=%D0%A1%D0%B0%D0%BD_%D0%A4%D1%80%D0%B0%D0%BD%D1%86%D0%B8%D1%81%D0%BA%D0%BE&prop=langlinks&lllimit=max ? [21:41:01] MaxSem: No, we need the full names of each language [21:41:23] theo|away: we currently do another api request to get this information :( [21:41:28] Yep -.- [21:41:30] see _getAllLanguages() [21:41:32] I know :( [21:41:43] MaxSem: dealing with languages from the API is a huge mess [21:41:45] mhurd: yea thats fine [21:42:01] you have to do action: 'query', meta: 'siteinfo', siprop: 'languages' first [21:42:56] you can combine it with the rest [21:42:58] oh wait theo|away it seems you can add two siprop values [21:42:59] wow [21:43:05] Indeed [21:43:10] The API http://sherman.blogs.com/snapshots/images/2008/05/15/dsc_0092_blog.jpg [21:43:19] i wish this stuff was much more logical [21:43:35] that is the ugliest mess ever [21:43:40] but *shrug* [21:44:14] see yurik MaxSem it's not just me!!! Any newcomer to the API sees this kind of thing and it's a huge demotivater… ^ [21:44:36] we need to build a simpler more logical API2 asap! [21:44:44] jdlrobson, yurik wrote it, not me:P [21:44:46] * MaxSem hides [21:45:03] yeh that's why i'm cc'ing him. He's heard many of my moans [21:45:19] theo|away: i hope you're willing to don armor and attempt this :) [21:45:53] Eh, big red X's aren't really big motivators ;) but i suppose i'll work on it, maybe later though [21:52:38] jdlrobson, MaxSem we have OPW student writing tutorial :) As far as langs go, I couldn't have predicted how wikidata would develop and how they would want to extend the api. And yes, in the hindsight i wolud have done a few things differently :D [21:52:46] if only i had a time machine ... [21:52:57] yurik: a good IP should not have a tutorial :) [21:53:00] *good API [21:53:06] have/need [21:54:09] jdlrobson, yes and no - it all depends on the approach and the amount of data. Try facebook api - its huge. We provide ALL the data in our databases + parsing + tons of other actions. This has to be at least as complex as SQL :) [21:54:37] and SQL takes some time to learn... like a semester :) [21:54:41] the API is written from an optimization pov not from a human point of view [21:54:48] that's my main grievance with it [21:57:29] I don't think we should significantly sacrifice performance for convenience. Doing a busy loop is easier than writing a system that wakes up on a timer, yet its not the right approach [21:58:06] yet of course we shouldn't require people to rewrite tcp stack to interface with the api :) [21:58:58] in any case, I totally agree that we should simplify it - it has been 8 years ago when i wrote it - we should have learnt a few things by now :) [21:59:17] jdlrobson, ^ [21:59:26] like DON'T ADD NEW FEATURES BECAUSE IT'S ENOUGH [21:59:42] MaxSem, not gona happen :) [22:00:01] not because of me mind you [22:01:30] some additions are dictated by new functionality like wikidata, and some - to help with ajaxy approach for a number of tasks [22:12:20] jdlrobson: i'm not going to be able to do this til anomie's patch is merged will I? [22:14:29] theo|away: jdlrobson just stepped away for a bit. he should be back in few [22:14:43] Thanks :) [22:14:53] theo|away: thanks for being so awesomely productive so far [22:15:13] theo|away: its great to see volunteers like you just in and get so much done this quickly [22:15:13] I really enjoy it! [22:15:16] really impressive [22:15:20] theo|away, you can checkout that change and continue developing [22:22:51] (03PS1) 10MaxSem: POC: Mustache in UserProfile [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104150 [22:23:07] here we go:) [22:23:18] (03CR) 10jenkins-bot: [V: 04-1] POC: Mustache in UserProfile [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104150 (owner: 10MaxSem) [22:42:33] (03CR) 10Kaldari: "Did either of you read the commit summary?" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103887 (owner: 10Kaldari) [22:45:29] jdlrobson: I guess my understanding of card 1451 was different than that of you and jgonera. [22:45:59] although since there are no acceptance criteria I guess there's a lot of room for interpretation [22:47:30] kaldari: yeh i think all 3 of us are interpreting this bug in a different way. "We should probably now be able to reduce or kill our getToken code.." < I guess I focused on the reduce bit [22:47:40] I think it's bad to have two methods of getting tokens though [22:47:44] we should have one entry point [22:47:56] which seems like it would need to be api.getToken [22:48:21] yeah, we can't get rid of api.getToken unless we expand user.tokens in core [22:49:00] jdlrobson: plot twist, you know how anomie just said to use variantarticlepath? it's by default just set to false O.o [22:49:34] jdlrobson: although for all of the instances where I changed it in that changeset, apt.getTokens is overkill. [22:49:56] kaldari: possibly - but it will be confusing to see two ways to get tokens - it will not be clear which one to use [22:50:00] theo|away: o_O [22:50:14] Do you want me to just add a short-curcuit to api.getTokens so that it uses user.tokens unless other types are needed? [22:50:21] theo|away: i think you need to set the default language [22:50:34] theo|away: try $wgLanguageCode = "sr" [22:50:48] kaldari: yup api.getTokens should call mw.user.getTokens [22:50:52] it gives us a clearer path to upstreaming [22:51:02] even so our own wrapper around getToken is useful [22:51:30] true, although we do need to upstream it at some point [22:51:52] there's no inherent reason why mobile should have its own api wrapper [22:52:02] theo|away: you may need to set $wgVariantArticlePath = "/$2/$1"; in your local settings [22:52:18] kaldari: well there is at the moment since all our stuff uses M.require :) [22:54:52] jdlrobson: so for now, do you want me to not use user.tokens.get at all in MobileFrontend except for in api.getTokens or is it OK to go ahead and migrate it where possible? [22:55:12] kaldari: I would prefer we continued to use api.getTokens [22:55:15] ok [22:55:18] until there is a clear path to completely removing it [22:55:25] it keeps all our token code in one place [22:55:31] and it will just confuse us [22:55:41] (about when to use mw.user.getToken over api.getToken) [22:55:48] do you agree? [22:56:00] sounds reasonable [23:11:56] jdlrobson: Our current api.getToken code specifically disallows anonymous users from retrieving tokens. Is it OK to get rid of that? It seems to violate separation of concerns. [23:12:53] kaldari: sure but i'd be super careful removing it - it might break other legacy things that might rely on it to work out if a user is anon or not. If it was me i'd do that in a separate patchset [23:13:07] good idea [23:26:08] (03PS1) 10Kaldari: Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 [23:26:38] jdlrobson: ironically, it made api.getTokens longer rather than shorter :P [23:26:46] (03CR) 10jenkins-bot: [V: 04-1] Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 (owner: 10Kaldari) [23:26:51] kaldari: mm that is ironic lol [23:27:31] wonder what jenkins is complaining about [23:28:22] ah, I see... [23:32:50] we return a deferred, not a string [23:33:13] (03CR) 10Kaldari: [C: 04-1] "need to return a deferred, not a string" [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 (owner: 10Kaldari) [23:39:42] (03PS2) 10Kaldari: Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 [23:40:17] (03CR) 10jenkins-bot: [V: 04-1] Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 (owner: 10Kaldari) [23:43:12] (03PS3) 10Kaldari: Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 [23:43:47] (03CR) 10jenkins-bot: [V: 04-1] Add short-curcuit in api.getTokens for tokens availble in user.tokens [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/104160 (owner: 10Kaldari) [23:44:47] (03PS3) 10Jdlrobson: Story 1512: Apply new chrome in beta [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103618 [23:47:16] kaity: what should the header look like on special pages with the new chrome? Like edit overlay - bold text etc? [23:48:04] kaity: currently looks like http://imgur.com/gBWwn4N [23:49:39] (03PS4) 10Jdlrobson: Story 1512: Apply new chrome in beta [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103618 [23:50:41] (03CR) 10Jdlrobson: "Made a second pass. Had to guess on the header styling for special pages again as was unable to get hold of Kaity on IRC." [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103618 (owner: 10Jdlrobson) [23:51:00] (03PS5) 10Jdlrobson: Story 1512: Apply new chrome in beta [extensions/MobileFrontend] - 10https://gerrit.wikimedia.org/r/103618 [23:51:14] kaldari: SkinMobile - can you review it? [23:51:19] Keen to wrap that up asap [23:51:31] kaldari: https://gerrit.wikimedia.org/r/102853 [23:51:49] yeah, I started, but got distracted :)