[09:12:17] hullo [10:28:27] gonna take a break [10:30:55] o\ [10:30:57] o\ [10:30:58] o/ [11:58:37] bmansurov: yt? [12:01:33] https://gerrit.wikimedia.org/r/#/c/232222/ was rebased by jdlrobson after you +2'd it [12:04:51] phuedx, yes [12:05:00] phuedx, i'll take a look [12:05:10] i'm looking at it now too [12:05:18] ok [12:05:22] (mainly because i haven't taken a look at it) [12:07:00] phuedx, i've +1'd it, feel free to merge if it passes your judgment [13:31:53] coreyfloyd: hey, if you're around: does this bug also affect iOS? https://phabricator.wikimedia.org/T108333 [13:51:14] Bmansurov : hi [13:51:21] hello [13:54:13] I was little confused about your suggestion . [13:54:36] Can you explain it again [13:54:39] ? [13:58:26] I am asking that you suggested to seach placeholder to fill the empty space in js enabled browsers or js disabled browser [14:03:45] sorry, i meant the search input to fill the empty spac [14:03:46] e [14:04:14] if we can make the input width bigger than 80px, then there will be enough space for the placeholder text [14:04:43] adisha, js disabled browsers as the bug indicates [14:05:51] so I'd remove the 80px rule and play around with the display property, i.e. table-cell and expand the form [14:06:32] Other than search input we also need space for hamburger menu icon and search button as js is disabled [14:07:22] And if increase input size it will create for smaller size devices. For eg 240 px screen [14:08:22] adisha, yes, but currently the input size is at 80px even though there is enough space to stretch it [14:11:47] Yes we can do that also [14:12:49] So what you suggest [14:14:41] bmansurov: But what if my project name is something bigger which can not accomodate in 80px or a little more than than [14:14:45] that [14:15:32] For example: my placeholder was showing 'Search wikimedia' as wikimedia is my project name [14:15:49] Then I think Project name will cutoff.. [14:18:46] adisha, it does if you're on a super small screen [14:24:01] adisha, to clarify, i don't think you can do much in that situation other than shrinking the font size [14:24:52] and to my cut off text is better than no text [14:25:01] So what you suggest..because i was tring to do what you suggested [14:26:00] adisha, the text input and the button should use the whole space available [14:26:13] adisha, try removing 80px and see how things change [14:26:23] No when I tried it. It was difficult to identify where is input text [14:26:40] adisha, what do you mean? [14:27:16] I meant when I tried to remove placeholder from search input [14:27:35] It was difficult to identify where is search form [14:27:38] adisha, yes exactly [14:27:56] adisha, so i wouldn't suggest removing it [14:28:02] ok [14:29:38] bmansurov: I think you can have a look what i have done uptill now. Here is the link http://198.199.96.81/core/index.php/Main_Page [14:30:07] So that you can easily discuss what changes shuld I do [14:30:13] Is it ok? [14:30:59] adisha, try looking at the git history and find out why 80px was added in the first place [14:31:37] In the ui.less it is mentioned that : [14:32:12] Assume the smallest possible screen size. We want people to be able to search we don't care if it looks pretty. [14:33:05] as we can see that didn't work well [14:33:11] so try removing it [14:33:14] hehehe [14:33:17] ok [14:33:17] and setting the width of the input [14:34:17] larger than 80 px [14:34:33] in percentage [14:34:52] and float the button [14:35:00] or max-width [14:35:07] i'm not sure exactly to what [14:40:44] I tried all the solutions blutton is already float:left [14:41:53] i remember seeing a similar code in alpha Special:MobileMenu page some time ago [14:42:03] i'll try to find it and share it with you later today [14:42:16] ok [14:42:32] But all this solutions are not working [14:43:32] bmansurov: So should I change status again to WIP: [14:43:36] :( [14:44:17] you could, or you can just leave it as is because it has a -1 [14:44:48] ok [14:45:27] Actually I discussed the solution I implemented with FlorianSW and jdlrobson [14:46:29] ok, let me get back to you about it [14:46:40] ok [15:01:04] I want to participate in upcoming round of outreachy and want to contribute for wikimedia. So do Extension: MobileFrontEnd has a project that can I choose for outreachy internship? [15:08:46] adisha: possibly... let me think about that. Keep submitting patches so I can get a better idea of what you are interested in and understand your strengths! [15:23:22] jdlrobson: thanks and please think about it as I am interested in MobileFrontend extension. [15:29:35] dr0ptp4kt bearND: hey! i had a question on the tech lead sync etherpad. should i just copy whatever i put down for scrum of scrums? i'm not really sure what info this should have [15:30:48] mdholloway: on west coast time… but no - it doesn't affect us now (we don't have a dark mode) [15:47:16] mdholloway bearND: for restbase cards, are we just moving those straight to the "Done" column after they pass code review? [15:54:01] niedzielski: the tech lead sync stuff can sometimes be more details than the scrum-of-scrums. It could include things or ideas that are internal to the Reading team [15:54:59] bearND: should it be a superset of scrum of scrums? [15:55:11] niedzielski: Yes, on the Mobile Content Service board we move them directly to Done. On the Android board I usually move the to Ready for Signoff but a case could be made to move it to Done as well. /cc:mdholloway [15:55:31] niedzielski: yes, superset [15:55:37] bearND: ok will move to done on android unless there are objections [15:55:49] thanks! [15:56:04] coreyfloyd: cool, thanks. one reason i was asking is because there [15:56:51] need to be a small change to night.less in the MobileApps extension [15:57:56] gotcha [16:02:07] ping kristenlans [16:14:08] So...this is the mobile^Wreading team channel? [16:18:07] Can someone talk about whether Media Viewer support is at all on your radar? [16:22:45] marktraceur: What do you mean by "Media Viewer" support? [16:23:34] Deskana: Every possible meaning, but there's a particular bug I have in mind [16:23:37] In MultimediaViewer [16:25:25] Oh, never mind, MatmaRex closed it. I guess he fixed it. [16:51:35] bmansurov: I will do it asap.. [16:53:35] dbrant: hey! quick question: when did we want to remove the link preview a/b testing again? [16:56:55] niedzielski: hey! i would say, as of the next beta release. we just need a little more data, and we're done. [16:57:13] and sorry for Phab spam. [16:57:30] ah, cool. thanks :) [17:16:29] mhurd_afk: bearND standup [17:27:53] mbinder: dbrant coreyfloyd don't leave! [17:27:55] come back! [17:29:14] whoops, sorry bgerstle what's up? [17:29:27] mbinder: bearND and i were going to talk about the link preview stuff [17:29:41] we can also schedule a dedicated meeting for this if that's easier [17:29:55] dbrant: coreyfloyd ^ [17:30:01] but we'd like to chat about it now both to settle the issue and to make sure dbrant is in the loop so as to continue the convo after bearND leaves [17:30:19] niedzielski: you got SoS covered? if not, bearND could you cover? just say for the other platforms "no update", then cover the android + mobile content service [17:30:38] niedzielski & bearND i'm in a retrospective, so can't go to sos [17:30:51] dr0ptp4kt: sure i can go [17:31:26] niedzielski: thanks [17:32:21] dr0ptp4kt bearND: so i should update the etherpad for web and ios, right? [17:32:56] niedzielski: i removed the web and ios stuff. no need to speak to it today [17:33:12] dr0ptp4kt: ok cool [17:37:20] dr0ptp4kt bearND: ok reading has been represented :) [17:43:00] coreyfloyd: etonkovidova also, looks like we did mess up crash reporting, but for alpha only [17:43:24] easy to fix, i'll submit a PR shortly once i confirm [17:44:22] bgerstle: coreyfloyd: interesting...I will check today the latest alpha. if everything is fine. [18:20:09] Taking a moment to stand after hours of 2.33 hours of meetings. brb [18:21:17] yeah -- gonna have a stretch and give the kids a kiss goodnight [18:41:32] ping dr0ptp4kt [18:50:58] worked out why our tests are failing. Perf team broke stuff but yay our browser tests caught stuff and this is exactly why we need to do that spike around deploying only green code [18:56:22] rmoen: phuedx anything you need before i head to office? [18:56:35] jdlrobson: in a meeting [18:56:37] also, no [18:59:24] :) [19:07:38] jdlrobson: also, nice catch on the broken stuff -- how did you dig it out? [19:08:58] i haven't had the time to look at the failure too closely [19:35:52] ok -- i'm out for the night [19:35:53] <3 [19:39:47] phuedx|zzZ: in terms of debugging - rmoen|lunch told me that the browser tests were failing locally so i just ran them and it was kinda obvious it was broken. Then just git bisected my way to victory [19:43:59] bulk de-assigning myself from old firefox os and win8 tasks; may flood channel for a few mins :D [19:44:53] brion: What about Google Glass ones? ;-) [19:45:00] (Is it now Alphabet Glass?) [19:45:02] :D [19:45:27] "Glass Alphabet" is a cool product name, actually. [19:45:34] James_F: there's a rumor they've got a new version in testing, we'll see if it becomes a fun fad too or if it goes to practical uses :D [19:45:55] ah looks like the flood's just over in #wikimedia-dev heh [19:48:45] * James_F grins. [19:55:45] thanks rmoen for the hygiene cleanup [19:57:23] jdlrobson: np [19:58:02] AC repair man is here, probably afk for a few minutes [20:08:34] FlorianSW: hey, around? [20:09:02] * FlorianSW thinks about running away, if benestar asks "around" :D [20:09:14] benestar: just a joke, yeah, I'm here :) [20:09:34] bd808 are you around the office? [20:09:49] cool, so lets discuss https://phabricator.wikimedia.org/T110069#1597787 :) [20:10:19] how will we change the api module in mobile? [20:10:30] * api query [20:11:29] benestar: currently, it's not possible, for that we have https://gerrit.wikimedia.org/r/#/c/234730/2 and iirc jdlrobson want to think about this solution. [20:11:56] yes, I know... [20:12:23] so the options we have is either a config variable or a hook in another place [20:13:14] when using a config variable, we need some placeholder for the search term and the limit [20:13:22] benestar: I feel a bit unconfortable putting "real" code into a config variable, even if it's "only" an api query :( [20:14:01] FlorianSW: well, it's not really "real" code but only a json object [20:15:54] benestar: that's why I put it into " :P Please don't misunderstand me, but in my opinion it doesn't make sense to put something into a configuration, which isn't really a configuration. And a query isn't a configuration, it's something where it should be possible to hook in (without needing to set a configuration variable) [20:18:16] FlorianSW: something like this? http://pastebin.com/t4eCyNwq ^^ [20:18:27] I agree that this isn't really a configuration [20:18:48] hmm, convert : to => xD [20:19:00] benestar: ehhh, this is exactly that, what I want to avoid :) [20:19:20] better: http://pastebin.com/VdMq1TaT [20:19:29] FlorianSW: what is your alternative? [20:19:52] having a hook that is fired in this method? https://github.com/wikimedia/mediawiki-extensions-MobileFrontend/blob/master/resources/mobile.search.api/SearchApi.js#L35 [20:20:10] rather here: https://github.com/wikimedia/mediawiki-extensions-MobileFrontend/blob/master/resources/mobile.search.api/SearchApi.js#L180 [20:20:39] benestar: the change I posted above, or (depends on the needs of Wikidata and what we, in MobileFrontend, are willing to give access/a way to change too) a hook which is fired, when the the query is requested (to change only that) [20:20:52] benestar: exactly :) [20:21:09] jdlrobson: can you comment? [20:21:30] that would make it fully changeable (like a config var) and is targeted to do only what it is supposed to (without a way to change _anything_) [20:22:28] it would however require some mobile specific js code in Wikibase [20:22:44] but we don't heavily depend on some private functions which are subject to change [20:22:49] as in our first proposal [20:38:04] benestar: hey sorry catching up [20:41:02] getApiData = return $.extend( {..}, mw.config.get( 'wgMFDefaultSearchData' ) ? < benestar seems like a good first step [20:41:42] jdlrobson: Florian doesn't like the idea of storing an api call in the config [20:41:52] benestar: that's why I put it into " :P Please don't misunderstand me, but in my opinion it doesn't make sense to put something into a configuration, which isn't really a configuration. And a query isn't a configuration, it's something where it should be possible to hook in (without needing to set a configuration variable) [20:42:08] what api query do you want to change from / to ? [20:42:12] and before: FlorianSW> benestar: I feel a bit unconfortable putting "real" code into a config variable, even if it's "only" an api query :( [20:42:26] we have this line for instance: pithumbsize: mw.config.get( 'wgMFThumbnailSizes' ).tiny, [20:42:45] jdlrobson: thumbnails will not be supported in the first version, but may follow [20:42:49] generator: mw.config.get( 'wgMFSearchGenerator' ) maybe [20:42:52] so we need a way to inject that data [20:43:06] something like this config: http://pastebin.com/VdMq1TaT [20:43:18] and then in javascript walk through all params and replace $query by the query string etc. [20:44:01] benestar: i'm confused. [20:44:09] :( [20:44:25] i thought the idea was for you guys to create a new api that takes the same parameters and deals with the conversion of titles on the server side? [20:44:36] so i was expecting this to be a simple case of swapping out the generator [20:44:40] or something similar [20:44:59] jdlrobson: the draft I have made is a completely new api module which however produces the same output [20:45:13] because we aren't yet at the point where we can use mediawiki generators and stuff [20:45:30] Cirrus integration needs more work but we need a workaround now to make the transition easier [20:45:58] hmm, maybe we can actually create a generator... [20:46:03] * benestar will think about this [20:46:10] back in ~30 mins [20:46:21] basically the less code in MobileFrontend the better - since we're actively trying to kill Mobilefrontend and make it a skin :) [20:46:22] k [21:07:00] jdlrobson: I'm back :) [21:07:21] so to summarize, I'm about to create an api module called wbmwsearchentities or something like that [21:07:37] mdholloway bearND dbrant: hey! good news is the CI job seems to be working. bad news is that it's failing on a false positive. please ignore jenkins _test_ failures for a few minutes while i put it in a temporary workaround. sorry about the hassle [21:07:47] so the api query would look like ?action=wbmwsearchentities&query=&limit= [21:08:14] benestar: sounds good. I would expect a config variable in MobileFrontend that sets wgMFSearchAction [21:08:22] so you can change it from query to wbmwsearchentities [21:08:34] it keeps the changes in MobileFrontend minimal [21:08:58] jdlrobson: well but your query currently uses pssearch= [21:09:07] and pilimit= [21:09:17] would those just be ignored by your module? [21:09:22] yep [21:09:53] I guess it wouldn't make much sense to accept pssearch as that prefix is strongly bound to prefixsearch :S [21:10:45] we really could do that but it somehow feels wrong [21:10:49] rmoen: can you review my patch around desktop quick surveys? [21:11:15] super keen to get that wrapped up so we can start finalising the extension and you are my only hope right now during pst hours :) [21:11:24] benestar: a lot of this feels wrong :/ [21:11:27] jdlrobson: I got u [21:11:58] jdlrobson: I think the best thing would be to have a generator=wikibasesearch [21:12:08] but then the prefixes for search and limit would be different [21:13:13] jdlrobson: we could introduce another config variable for the prefix to be used before query and limit ... [21:14:32] /msg jdlrobson You're talking about https://gerrit.wikimedia.org/r/#/c/235618/ ? [21:14:39] lol what? [21:15:04] no the only patch in code review column [21:15:11] https://phabricator.wikimedia.org/T107603 [21:21:26] benestar: can you update the phabricator task? there's lots of patches hanging aroun and it's unclear on the way forward. I think this could do with writing pseudo code firsy [21:21:28] *first [21:21:32] https://phabricator.wikimedia.org/T110069 [21:21:55] niedzielski: Excellent! It's really nice to see tests running. What do you plan to do about the TranslationTests failures? [21:23:12] bearND: https://gerrit.wikimedia.org/r/#/c/235622/ is the workaround patch to unblock folks (currently waiting with fingers crossed that CI passes). the proper fix will come in T110243 [21:37:29] bearND mdholloway dbrant: well, CI sort of worked \o. https://gerrit.wikimedia.org/r/#/c/235622/ is ready to review and merge but i gotta figure out why Lint is declaring success when it has failures [21:44:17] coreyfloyd: kaity there's no save button on the search result cell mocks: https://phab.wmfusercontent.org/file/data/rowd2ivayvuxeyv35obt/PHID-FILE-h5proub7acgtqdi4s7yh/search-29.png [21:44:20] https://phabricator.wikimedia.org/T110725 [21:44:22] is that right? [21:44:31] oh right, kaity you're on the road [21:44:44] bgerstle: yeah - there is supposed to be a seed button on the right… [21:44:51] figured [21:44:53] you can just add and use standard spacing [21:44:54] but what does seeding something do? [21:45:01] seeding? [21:45:03] ^^ [21:45:13] just pullin' your leg [21:45:34] what?!? - you joke is over my head [21:45:46] "coreyfloyd> bgerstle: yeah - there is supposed to be a seed button on the right…" [21:46:29] thaks rmoen [21:47:09] I like the improvements jdlrobson, had to play with them a minute before merging :) [23:03:15] niedzielski: any chance of getting my app patches merged today? [23:04:18] bearND: sure thing. do you mind if i spend a little time testing these final versions? [23:04:44] niedzielski: Sure. I just rebased the follow-up patches BTW. [23:05:16] bearND: ok awesome. i'll ping you on irc if any issues come up [23:06:02] niedzielski: thanks! [23:06:55] bearND: np, thanks for getting these in before your family +1! :) [23:07:51] rmoen: your accessibility patch needs a slight tweak. I know it's out of the sprint but would be great to get that merged it's been bugging me for some time. [23:08:00] jdlrobson: actually working on it [23:08:08] sweet [23:08:32] jdlrobson: its been bugging you for some time? its a 1 day old patch [23:08:38] oh [23:08:41] the card i see [23:08:44] yeh :) [23:08:51] i'm not thattttt impatient ;-) [23:08:55] hah [23:09:11] ok yeah i think its a small css tweak. should have it in a sec [23:13:22] bearND: i left some comments on ps1 for https://gerrit.wikimedia.org/r/#/c/235524. can you check those out? [23:19:03] marxarelli: around? You're not allowed to edit interface messages (protectednamespace-interface) (MediawikiApi::ApiError) [23:19:10] keen to think of a good way to work around this [23:19:16] https://integration.wikimedia.org/ci/job/mwext-mw-selenium/713/console < context [23:20:11] best thing i can think of is $wgGroupPermissions['*']['edit-interface'] = true; [23:23:16] niedzielski: ah, thanks. I missed those comments. Will fix shortly [23:24:31] bearND: thanks, i think i added them without -1. sorry about that [23:25:04] yeah, I usually get my queue from the -1 [23:41:36] bearND: so everything seems to test fine \o/ [23:42:27] niedzielski: about to push the changes for the last patch but having some gerrit troubles. It claims I made no changes. [23:42:37] niedzielski: feel free to merge the first three [23:48:31] niedzielski: thanks. I'll see if merging those three has help with the fourth patch [23:49:14] bearND: cool, maybe rename a variable or something to trigger a change [23:49:18] bearND, you forgot a git add ? [23:49:56] niedzielski: I've tried multiple changes. Getting: "(W) No changes between prior commit 59ca2ac and new commit 6205b15" [23:50:05] and "! [remote rejected] HEAD -> refs/publish/master/service (no changes made)" [23:50:18] Platonides: git status is empty. So, no [23:51:28] I also I'll just wait a few minutes until the prior patches are done with the gate-and-submit job [23:51:59] If necessary I start a new patch by changing the change-id [23:52:35] bearND: weird. so if i pull your patchset, i see 4b72d9b [23:53:41] bearND: which isn't either of the commits you mentioned [23:55:59] niedzielski: yeah, that's weird [23:59:00] \o