[00:01:41] MaxSem: Please let me know when it's live [00:05:26] is in article photo upload going out today ? [00:05:50] no [00:06:04] k [00:06:32] preilly, here we go [00:08:11] MaxSem: is it live now? [00:08:19] yes [00:08:30] MaxSem: okay great thaks [00:08:34] s/thaks/thanks [00:10:11] arghhhh im entering the jetlag witching hour [00:10:42] awjr, have you tried melatonine? [00:11:11] MaxSem: yes but that will make me fall asleep, which i would have 0 problem doing right now :p i am trying to stay awake [00:11:35] my blood is basically pure cafeine at this point but im not sure it's totally working [00:11:39] well, you should have used it last night [00:12:55] i got a regular night sleep last night - falling asleep is definitely not the problem. i think it's a combintion of catching up on all the sleep i lost during 38 hours of travel and feeling like it's 4am right now [00:15:06] awjr: are you awake enough for 4:30 ? [00:15:22] tfinc: i think so but if i start sluring my words you know why [00:15:31] and no it's not because im drunk [00:16:00] MaxSem: What is the version number(s) that are actively deployed [00:18:36] preilly, your "fix zero banner" in both branches [00:18:43] 161548570e82b58f0463f73be8fd9e62be4da54b [00:18:56] wmf3 and 4? [00:19:10] yes [00:19:14] coolio [00:19:16] just checked [00:20:20] MaxSem: what is the site matrix url with version numbers? [00:20:59] awjr: that'll be amusing as i'm crashing too [00:21:05] awjr: perhaps we can just have coordinated nap time [00:21:07] https://en.wikipedia.org/wiki/Special:Version [00:21:09] hahaha [00:21:13] egh [00:21:14] MaxSem: thanks! [00:21:32] MaxSem: actually the other page [00:21:40] ??? [00:21:57] MaxSem: the one that lists each wiki and it's version [00:22:14] http://noc.wikimedia.org/conf/highlight.php?file=wikiversions.dat [00:22:21] MaxSem: bingo [00:22:26] MaxSem: thanks! [00:27:16] preilly: willing to take a quick look at https://github.com/wikimedia/WikipediaMobile/pull/314 ? fixes orientation issues and other minor iOS 6 compatibility buglets by updating iOS to cordova 2.2 [00:27:23] the main diff's a bit….. big of course ;) [00:28:16] brion: I can in a little bit [00:28:20] thx [00:30:17] brion: np [00:32:19] tfinc: im experiencing some difficulty loading the hangout - im getting the error 'It's taking too long to connect you to this hangout. Try again in a few minutes.' [00:33:21] same here [00:33:49] i do like how it shows a live preview of my camera [00:33:52] but then doesn't load anymore [00:33:58] :( [00:33:58] skype? [00:35:51] what's the difference between "u( document.body ).hasclass( 'beta' )" and "M.setting( 'beta' )"? [00:36:45] hmm, well one is checking for classes on the document and the other is checking config variables… but in theory they should be checking for the same thing [00:36:50] seems icky having two ways of doing that though [00:37:03] * brion looks at jon's empty chair and stares [00:37:39] jgonera: M.setting( 'beta' ) is better IMO [00:38:03] separate application logic and application state from the DOM as much as possible [00:39:31] ok, I thought so, thanks [00:40:29] ori-l, what do these session keys in event logging feed denote? [00:41:14] MaxSem: they're timestamps (microseconds since epoch, so divide by 1000 to get standard unix time format). the timestamp is set at page load [00:41:24] so it gives you a way of grouping multiple events that happen on a single page view [00:42:01] so there's no way of tracking individual users, got it... [00:43:50] yeah. it helps answer jon's chief request, which is to know how sections are getting used [00:45:37] i'm going to take down the websocket event stream now -- the data is getting slotted into a mysql instance for data analysis. exposing the event stream as a page was just for debugging purposes. Maryana has access to the data. [00:46:59] it's on stat1 - if anybody is interested in getting access, ping dario for the generic researcher login creds [00:49:58] ori-l - is eventlogging camel case? [00:50:06] just want to make sure i have it right in the email :) [00:50:24] yup [00:52:11] im about to take off - anybody need anything from me before i go? [00:53:57] mustache [00:54:34] i feel like i'm talking to your evil twin now, awjr [00:55:42] Maryana: maybe you are! [00:55:55] ahhh! [00:56:06] just remember: true mustache is a state of a mind [00:56:47] ok im off to collapse. see you all tomorrow! [06:28:30] New review: Jdlrobson; "please revert taken care of in" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/34428 [16:10:40] New patchset: MaxSem; "introduce styling for code pre tags" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/30811 [16:11:25] Change merged: MaxSem; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/30811 [17:34:32] greetings all [17:34:40] hey [17:37:55] hey MaxSem how did you jet lag do this time? [17:38:15] well, I was only 1.5 hours apart [17:38:22] who's around and brave enough to review https://gerrit.wikimedia.org/r/#/c/32560/ ? [17:38:39] the lag was mostly due to my weird sleep schedule in Russia [17:38:46] but it went pretty well [17:40:19] MaxSem: your becoming a seasoned traveler then [17:41:12] * MaxSem kicks gerrit-wm [17:42:09] awjr: MaxSem up for a review of the change that i mentioned ? [17:42:21] i can do it shortly [17:42:33] tfinc, already -1'd [17:42:35] oh looks like MaxSem already did it, tfinc [17:42:39] thanks [17:43:24] I have some stuff I want to see reviewed, too. my list is longer though:P [17:43:45] were pretty behind on our review. awjr MaxSem how can we get better about it? [17:44:00] do we have any visuals of just the mobile review queues ? [17:44:11] most of stuff that remains is too frontendy for me [17:44:42] awjr: would you be able to review the rest for jon ? [17:45:05] can we throw Brion at it, pretending he's giving a CR introduction to Juliusz? [17:45:16] :P [17:45:25] lolz [17:45:34] i'd like to leave for the holidays knowing our review queue is empty [17:45:52] s/id like/we need to [17:45:52] hola. [17:45:56] hey aharoni [17:46:10] is it just me or is mobile-feedback exploding in the last few days? [17:46:30] "Technical problem" in many languages. [17:46:44] we rerouted some sewage into it:] [17:46:57] tfinc: we have 16 unreviewed change sets in the master branch, i intend to keep working at it as im doing the other things on my plate for today. the trick is some are short and easy, others are big and hairy which take a long time and a lot of focus. there's no way i'll get thorugh all of them today [17:47:00] Is this automatic subject a feature of the app, or of some mobile OS? [17:47:31] awjr: how about we a dedicated time boxed hour for the backlog between everyone ? [17:47:46] imagine coming back on monday with a zero backlog [17:48:00] hehehe that would require jon to put his laptop awway while he's travelling [17:48:23] thats impossible and we like that. zero queue as of today [17:48:33] tfinc i'll bring it up at standup [17:48:37] k [17:48:47] * tfinc goes to setup the standup area [17:51:15] and were up [17:56:28] Most people complain about screen rotation on iPad. But there was also somebody writing in Indonesian and asking how to add information to Wikipedia. I replied with a link to the welcome page of the Indonesian Wikipedia. [17:59:15] awjr, mtg [17:59:28] aharoni: thanks for the heads up. we have a bug and parts of a fix for the landscape issue [18:10:13] New review: MaxSem; "Related: https://gerrit.wikimedia.org/r/#/c/33399/" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/33396 [18:19:35] * tfinc ponders an irc bot that reminds us of the cr backlog [18:19:48] merging https://gerrit.wikimedia.org/r/#/c/32548/3 would allow us to submit a few already reviewed revisions [18:22:46] what's blocking a merge? jons last comment says that its a refactor and doesn't require product [18:23:06] the dependencies [18:23:30] that particular changeset didn't necesarily need to be run by product, but the group of dependent changesets did [18:23:38] but jon's already done that and product seems happy about them [18:23:58] brion: did juliusz say he was getting late today? [18:24:18] he was looking at apartments after work yesterday, don't recall anything about being late today [18:25:12] awjr: i started on our Open Source/Free Knowledge Culture slides. did you get the share? [18:25:24] tfinc i did thanks, haven't had the chance to look yet [18:25:51] tfinc i may have time to add some stuff today but probably not until monday [18:26:44] New review: awjrichards; "This commit is part of a larger group of commits that has a significant impact on how tables work in..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/32548 [18:27:41] MaxSem: where is GeoData at? [18:27:52] at code review:P [18:28:38] MaxSem: did you not get niklas to review the changes you needed at the decamp ? [18:28:40] devamp ? [18:28:44] yes [18:28:52] but only to the +1 extent [18:29:02] who else do you need on thes ? [18:29:42] someone from our team on PHP side ,some op on puppet side [18:30:23] i am happy to look at the php side but not sure when i can get to it [18:30:32] brb [18:36:45] MaxSem: is solr set up on staging? [18:37:27] yes, staging uses solr from mobile-solr2 [18:37:34] sweet [18:37:52] MaxSem: i can try to look at the changes this afternoon [18:38:00] cool, thanks [18:39:04] New review: awjrichards; "We can revert this when the comments are addressed from https://gerrit.wikimedia.org/r/#/c/30309/2" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/34428 [18:46:15] brion: with https://gerrit.wikimedia.org/r/#/c/29621/ how do i get to the diff view? [18:46:41] awjr: tap on the list items, it should navigate in [18:47:40] brion hmm it's just showing me info about the user, im not seeing a diff view [18:47:56] it should show a mini diff in there... [18:47:59] do i need to make a config change or something to see the diff/ [18:48:04] maybe the styles are borked again lemme check [18:48:09] kk [18:50:14] k look: [18:50:27] this just arrived to mobile-feedback: [18:50:29] This is more of a correction with an article about the city I live in "Rosenberg, Texas." Our Fort Bend County Fair happens in September not October as listed in the "Rosenberg, Texas" article. For the past 26 years it has always been in September. Usually, the last week of September. [18:50:53] Am I the only one thinking "Article Feedback"? [18:51:09] awjr: yeah styles are borked [18:51:12] the user info is covering the diff [18:51:12] brion doh! [18:51:20] ahha ok [18:51:38] brion i saw some more issues in that changeset - one sec i'll add my comments [18:51:43] tx [18:51:55] New review: awjrichards; "* After drill down, back does not go back to the special page but rather the article you were previo..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/29621 [18:52:21] aharoni: until yesterday the contact page in the mobile view provided a way for people to give article-specific feedback that would not go to mobile-feedback [18:52:35] yeah i'm not sure how to handle that back button [18:52:45] Oh. And why was that changed? [18:52:53] aharoni: now we only display a section for technical feedback, it's possible that that's going to get used for all sorts of non-technical feedback too [18:53:18] It IS used for this, as the last couple of days demonstrate. [18:53:21] aharoni: it was a request frmo OTRS to reduce the huge amount of spam/noise they were receiving from the page. so we disabled everything else until we can implement a better soluiton [18:53:34] O I C. That explains. [18:54:51] Handling it in OTRS actually makes more sense here - not because I want to drop the work on other people, but because one-click automatic replies. It's probably the only good feature of OTRS, actually. [18:59:25] awjr: ok the problem appears to be the 'position: relative' on content_wrapper [18:59:30] brion: i spy a jgonera in the room :) [18:59:36] if i remove that, i get the formatting i expected again [18:59:47] want to run that by jon, see what it was added for [18:59:47] aharoni: here's the relevant bug: https://bugzilla.wikimedia.org/show_bug.cgi?id=41502 [18:59:52] tfinc: \o/ [19:00:14] Maryana: do we have a story in the backlog for better addressing https://bugzilla.wikimedia.org/show_bug.cgi?id=41502 ? [19:00:53] brion is there a way to work around for now? it would be sweet if we could deploy the watchlist stuff next week and jon won't be back til next weds [19:00:57] no, but we probably should [19:01:13] New review: Brion VIBBER; "Style problem is caused by "position: relative" on #content_wrapper." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/29621 [19:01:14] Maryana: yeah, because now we've likely shifted the noise/spam problem to oureslves :p [19:01:23] hehe [19:01:32] awjr: i could remove the position: relative but i'm not sure what it was added for yet [19:01:58] awjr: i'm sure that will help accelerate finding a solution :) [19:02:05] heh [19:04:53] ok it was added in commit 61ffcc69e120b0d0716f7a5959d11d13d2d5c57f "add edit link to pages in beta mode" [19:04:54] sigh [19:05:32] Maryana: will you being using your 3rd floor desk today? [19:05:37] New review: Brion VIBBER; "The position: relative was added in commit 61ffcc69e120b0d0716f7a5959d11d13d2d5c57f "add edit link t..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/29621 [19:05:40] * awjr shakes fist [19:06:01] tfinc: nah, working from the berkaberk today [19:06:13] berkaberk? [19:06:22] brion: i need you to show jgonera what a missed standup email entails [19:06:26] oh, berserkley? [19:06:27] berkeley :) [19:06:42] berserkley. i like that! haven't heard that one before. [19:06:46] awjr: you know … bezerkeley [19:06:50] hehehe [19:06:51] jgonera: you missed the standup, I am forced to make you write an email now :) [19:07:22] jgonera: send a quick note to mobile-tech about what you've been working on, and call out anything you might need help on etc [19:07:33] jgonera: let us know what you did since the last standup, what you're up until the next one, if you have any blockers, and any other news pertinent for the team [19:08:00] Brion: I've got a wiki-history question for you, if you don't mind. When did watchlists become a part of the MediaWiki software? [19:08:13] brion, awjr, ok, mobile-tech is a mailing list, right? mobile-tech@wikimedia.org? [19:08:21] jgonera: yeah [19:08:30] i…. think [19:08:32] * brion checks [19:08:35] yes :) [19:08:40] yes :D [19:08:54] * brion always uses autocomplete in email, never remembers domains for sure ;) [19:08:54] ok, I'll write it in a few minutes [19:08:57] thanks! [19:13:13] New patchset: Brion VIBBER; "Mobile replacement for Special:Watchlist, initial version." [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/29621 [19:14:30] New review: Brion VIBBER; "Patch set 5:" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/29621 [19:15:55] New review: MaxSem; "watchlist.js is empty, is it intended?" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/29621 [19:21:15] New review: Brion VIBBER; "Watchlist.js is indeed empty, I never got round to putting anything in it. Could take it out, but I ..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/29621 [19:22:14] tfinc, how do I know if I have access to WMF Engineering Google calendar? [19:22:32] you…. should i think [19:22:36] should be able to select it somehow [19:22:43] jgonera: you would open up google calendar. under other calendars search for it [19:23:25] tfinc, it's not there [19:23:41] tfinc, I only have my own calendar and Tasks [19:23:59] jgonera: i just added you. you should get a notification [19:24:34] tfinc, thanks, got it! [19:28:17] brion: will we be prepping a new ipa for release before the holidays ? [19:29:00] that's up to someone to review https://github.com/wikimedia/WikipediaMobile/pull/314 :) [19:29:04] preilly: poke ^ [19:29:36] ponders bricks for his office storage http://www.thinkgeek.com/product/edac/ [19:29:47] nice [19:29:50] build a fort dude [19:31:38] this would work well too http://www.thinkgeek.com/product/eded/ [19:31:58] brion: it seems okay in my limited testing — I really don't like the iOS 6 issue with FB [19:32:23] [WikipediaMobile] preillyme pushed 4 new commits to master: http://git.io/VJk1LQ [19:32:23] WikipediaMobile/master 2c6e611 Brion Vibber: Upgrade Cordova for iOS from 1.7 to 2.2... [19:32:23] WikipediaMobile/master 4094859 Brion Vibber: Adjust fb key placeholder [19:32:23] WikipediaMobile/master d462cbf Brion Vibber: Fix for Facebook sharing on iOS 6.... [19:32:27] brion: I merged commit da90865 into wikimedia:master from brion:cordova22-ios-update just now [19:32:30] whee [19:33:02] Project WikipediaMobile - Nightly builds build #456: SUCCESS in 16 sec: https://integration.mediawiki.org/ci/job/WikipediaMobile%20-%20Nightly%20builds/456/ [19:33:02] * Brion VIBBER: Upgrade Cordova for iOS from 1.7 to 2.2 [19:33:03] * Brion VIBBER: Adjust fb key placeholder [19:33:03] * Brion VIBBER: Fix for Facebook sharing on iOS 6. [19:33:08] thanks preilly [19:33:18] brion: np [19:33:25] tfinc: willing to do a 3.3 push before the holiday? or want to do some betas first? [19:33:39] we should beta as were picking up a cordova update [19:33:46] *nod* [19:33:49] if it was just a landscape fix i'd be less worried [19:33:53] brion: whats your take? [19:34:01] i'd go for some betas first [19:34:07] might shake out a few surprise issues [19:34:35] oh and this restores the armv7s build for iPhone 5, since we're building CordovaLib ourselves now [19:36:27] nifty. more fun for Jcmish to test for us [19:37:05] :D [19:54:40] why does the beta version make so many AJAX requests to get the article content instead of just fetching it with the page like the normal version? [19:56:13] jgonera: loading is optimized for initial toc and first article section. anything else is lazy loaded after behind the scenes [19:56:19] that way the user sees a page as fast as possible [19:56:32] it seems it's not like that [19:56:50] the whole article is loaded in a single additional AJAX request just after the page is loaded [19:56:56] is it finished? [19:57:24] i'd have to check the beta but we may be in the middle of this transition where each section is loaded dynamically on request. the case that i mentioned above is what were transitioning to [19:57:46] brion: is that the chrome vs article loading that jgonera is seeing ? [19:58:05] as chrome and article load are seperate [19:58:13] we don't like to reload the chrome as its likely has not changed [19:58:17] ok, I see, so I guess that's just a temporary solution [19:58:29] jon would know better the details of how this works [19:58:36] what do you mean by "chrome"? :) [19:58:57] jgonera: main menu, search bar, etc [19:59:07] items that show up for every single article and never change [19:59:19] reloading those is a waste of time as they don't change per article [19:59:27] ok, so I'll just try to fix that bug without changing too much since Jon is probably going to be working on that [19:59:37] sure, I understand [20:00:07] jgonera: i think what you'll need to do is wait until the page is fully loaded by ajax, then jump to the hash [20:03:41] brion, yeah, I know, the problem is that the code that is supposed to do that doesn't know where to look for the info about sections, it still "thinks" the page structure is like in the non-beta version [20:03:53] hmm [20:04:02] possibly we need to adjust how we insert the data [20:04:06] but I'm getting to that [20:04:10] when i looked it looked like the anchors are missing [20:04:44] exactly, they don't exist anymore until the section is loaded :) I just have to find out how to get to the section data loaded through AJAX [20:05:00] whee [20:05:10] let me know if i can help you with anything [20:05:21] sure, thanks [20:19:47] brion, how do I run the test suites? I saw there's some JS tests and something in PHP too [20:20:25] jgonera: in LocalSettings.php set $wgEnableJavaScriptTest = true; [20:20:32] and go to Special:JavaScriptTest (?) [20:21:02] not sure about the php tests in MF, lemme check [20:21:52] allegedly run 'make tests' but … that doesn't work for me, it can't find it's bootstrap die [20:23:12] there we go, had to have directories in the right order [20:23:56] trying JS tests, it keeps saying "Loading Special:JavaScriptTest" and nothing happens, seems way too long for JS tests [20:24:34] hmm [20:24:56] I'm running it in mobile beta [20:25:35] try it in desktop mode, i'm not sure if it's supposed to run in mobile mode that might break some of the other tests [20:44:51] pastry run, brb [20:56:32] awjr_interview: http://www.businessinsider.com/github-office-tour-2012-8?op=1 [20:57:20] brion, it works in desktop, thanks. one more thing: when we have a subsection title, e.g. "Flora and fauna" how is the id (anchor) generated? just by replacing spaces with _? is there any standard function somewhere? [20:57:51] there's a transformation function... [20:58:30] is there one in JS as well or only in PHP? [20:58:37] brion: same article that i just sent to arthur but this seems like you http://static4.businessinsider.com/image/5037d6a1eab8ea9c77000003-590/this-techno-wonderland-desk-is-reserved-for-the-king-or-queen-of-developersthe-title-for-the-developer-assigned-that-month-to-coordinate-the-squashing-of-bugs.jpg [20:58:54] hehe nice [20:58:59] jgonera: looking [20:59:57] hmm [21:00:08] ideally we'd get the appropriate anchor sent to us along with the pretty name [21:00:25] Sanitizer::escapeId() appears to be used in the PHP side to make the actual anchors out of sanitized section names [21:00:27] from the AJAX api you mean? [21:00:34] yeah [21:00:45] but i think right now all we get is the sanitized name [21:01:31] for the moment, a first-order approximation of encodeURIComponent(name.replace(/ /g, '_')) will probably do [21:03:08] actually I just checked and it's already implemented, I can ask the api for an anchor for each section too, yay! [21:04:01] \o/ yay [21:39:33] tfinc: i like the octocat portrait over the fireplace [21:39:49] aaaand the hidden bottle of whiskey [22:10:54] * chrismcmahon was pairing with jcmish on our first sanctioned browser test of the mobile site and found a real nice bug, she should be writing that up any minute here [22:11:10] \o/ yay bugs [22:15:28] hehehe sweet [22:16:22] since we were starting from scratch, I was all "hmm, the test failed, must have made a coding error... whoops, no that's a real bug." [22:19:25] New review: awjrichards; "This changeset seems to break section toggling and it is not loading tables in an overlay for me." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/32552 [22:22:56] New review: awjrichards; "Captions in tables look a little weird (which is not a symptom of this changeset, just something I n..." [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/32548 [22:22:56] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/32549 [22:22:56] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/32551 [22:22:56] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/32548 [22:47:26] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/32558 [22:48:36] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/32559 [22:50:21] New review: awjrichards; "This seems to cause the random page text to appear before you've selected whether or not to read the..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/32561 [22:57:15] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/34064 [22:57:50] are patch sets in gerrit commits in git? [22:58:10] New patchset: awjrichards; "center the heading relative to the back arrow" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/34065 [22:58:27] jgonera sort of [22:58:47] i'm going to start winding down to leave for my drive to LA. does anyone need anything ? [22:58:56] jgonera, well actually yes - but additional changesets in a commit are the result of amending the initial (or previous) commit [22:58:58] jgonera: brion MaxSem Jcmish awjr --^ [22:59:05] tfinc: im good [22:59:08] i'm good [22:59:11] k [22:59:16] be safe! [22:59:21] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/34065 [22:59:31] have good traffic tfinc ! [22:59:52] New review: awjrichards; "per ori's comments" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/30309 [23:00:48] I just want to be sure I get it right: 1) create new branch 2) commit changes 3) git review [23:01:08] yeah, that will push your commit into gerrit [23:01:16] so I shouldn't have more than one commit per git review? should I keep amending the first one? [23:01:41] jgonera: if you want more clarity and/or to get more confused, take a look at http://www.mediawiki.org/wiki/Git/Workflow [23:01:56] jgonera: that depends [23:02:12] awjr, that's what I'm looking at and that's why I have questions ;) [23:02:26] jgonera: you would only amend the first one if you needed to make a change (either because someone spotted something bad in review, or you just wanted to fix something) [23:02:31] the only thing I used so far were github's pull requests and there multiple commits with different messages were fine [23:03:08] yeah you can do multiple commits, but gerrit turns them into multiple change sets with dependencies and it's kind of a pain in the ass [23:03:27] jgonera: but if you wanted to move on to work on something else, you would create a new commit (not amend). if you wanted to create smoething that was dependent on a previous commit, you would probably make that change on the same branch as the previous commit (but it's best to avoid this kind of situation if possible and try to keep atomic commits on their own branch to help avoid nasty dependency trees/merge conflicts) [23:03:32] easiest thing is to squash to a single commit, push that up into gerrit, and then amend and repush the same commit if need to revise it [23:03:33] hella [23:03:38] just type in whatever that page says:) [23:03:46] jgonera: if you see ways to improve/clarify that document, go ahead and edit it :) [23:04:07] ok ;) [23:07:31] jgonera, there's a test repo if you want to try things out [23:08:09] that's a good idea, where can I find it? [23:10:02] ok, I see, patch sets are like subsequent revisions of a change, right? [23:10:31] yeah that's a good way to put it [23:11:14] test/mediawiki/extensions/examples [23:13:15] happy thanksgiving all! [23:13:29] i'll check in on CR over the weekend if we've got any left :) [23:14:57] New review: awjrichards; "COOL! this seems to work now. My comments from patch set 5 still apply. It would be nice to get the ..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/29621 [23:15:18] New review: awjrichards; "er... i meant my comments from patchset 4" [mediawiki/extensions/MobileFrontend] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/29621 [23:18:09] brion can you review jon's photo upload related changes when you get to code reviewin? there's enough js there to really scare me and im feeling braindead from the other cr i've done today [23:18:54] jgonera, want a screen sharing demo of committing? [23:19:59] oh brion is not here anymores :( [23:20:11] MaxSem, I'll experiment with the test repo first, thanks [23:21:07] MaxSem: do you know if there's a way i can move commits in gerrit that have been made to the master branch into a remote branch? [23:21:20] i suspect there isn't but that would be amazing. [23:21:44] ehm, I don't know. try asking Chad [23:28:07] ok, just one last thing: I create a branch, make a single commit then git review, then I make another commit and git review again. the second git review wants to add both commits, is there a way to "reset" it? [23:28:58] yes, get the first commit merged:) [23:29:15] or use branches just for one commit each [23:30:27] in principle, if it asks you whether you want to push more than one revision and you answer yes, it will simply ignore the changes it already has [23:30:51] hm, ok, I'll try that in the test repo [23:31:05] there can be problem when remote branch has dependencies on master and vice versa though [23:31:25] if there can be only one commit per branch though it kind of feels like it defied the purpose of branches a bit [23:32:53] who said that branching is easy in git? he lied:P [23:42:25] New review: awjrichards; "Looks OK but I would like someone with stronger JS chops to review the JS. Also you should go ahead ..." [mediawiki/extensions/MobileFrontend] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/32683 [23:59:37] jgonera, have we lost you?:P [23:59:59] MaxSem, I'm here