[00:05:09] bawolff: thanks [07:12:43] New review: Siebrand; "Add some requests for priority." [integration/jenkins-job-builder-config] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/38555 [09:30:41] New patchset: Krinkle; "Sync: Commit live changes by hashar to jobs/MediaWiki-GIT-Fetching/config.xml" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48104 [09:30:55] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48104 [09:33:46] New patchset: Krinkle; "Sync: Remove old files already removed by hashar on live" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48105 [09:34:17] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48105 [09:35:39] New patchset: Krinkle; "Sync: Commit submodule update tools/codesniffer to latest master" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48106 [09:35:50] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48106 [09:42:31] New patchset: Krinkle; "Sync: Commit .gitignore changes for jenkins-job-builder" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48107 [09:42:47] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48107 [10:04:12] New patchset: Krinkle; "Sync: Commit move of tools/codesniffer submodule" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48109 [10:05:27] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48109 [10:05:54] New review: Krinkle; "FYI: http://stackoverflow.com/a/13942859/319266" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48107 [10:13:15] New review: Krinkle; "* Why docs.mediawiki.org instead of doc.wikimedia.org? Also in some comments you say doc.mediawiki.o..." [integration/jenkins] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/39212 [10:18:39] New patchset: Krinkle; "Build script for mediawiki documentation" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/39212 [10:18:42] New patchset: Krinkle; "Build script for mediawiki documentation" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/39212 [10:18:50] New review: Krinkle; "* Rebased" [integration/jenkins] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/39212 [12:21:34] hello [12:21:42] * OldPanda waves at hashar [12:22:38] YuviPanda: just discovered your SignPost app :-] [12:22:45] tch tch :P [12:22:45] late [12:22:46] :) [12:22:47] YuviPanda: any plan to have it officially endorsed by the WMF? [12:22:56] hashar: does it need to be? [12:23:00] It doesn't use any trademarks... [12:23:13] I do need to put out a blog post [12:23:20] well that might attract more users [12:23:30] I noticed there Google Play claims "only" 50+ downloads [12:23:41] having it under the wikimedia umbrella might attract more users [12:23:47] we steal apk's :) [12:23:55] hashar: true, i've done zilch marketting [12:24:04] anyway, whoever write the signpost could also add a story about the app :-D [12:24:13] endorsement is not important, getting the word out, is [12:24:31] hashar: so... they did :) [12:24:32] we could even have a banner running on en.wikipiedia.org : "Get our weekly newspaper on your phone!" [12:24:42] on the signpost [12:24:50] and that's why it has 50+ useres (about 70 to be more exact) [12:25:13] most of them are internal I think - Sue loves it. [12:25:32] hashar: also, there was an RFC going on about having community apps go through an approval process [12:25:37] as it currently stands, this will fail it [12:25:41] since it uses a proprietary service [12:25:46] Google's Cloud messaging [12:25:49] to send 'push' notifications [12:25:58] (Never miss another signpost issue!) [12:26:28] but I have doubts about that RFC - as it stands, it disallows iOS apps :P [12:26:50] ohhh [12:26:53] yes, and my cyanogenmod has no google account - I guess it won't work? [12:27:07] saper: well, it needs you to have Play [12:27:09] not a google account [12:27:15] (assuming you are not on CM7) [12:27:20] saper: if it is open source we can publish the files in the F-Droid repository of Android OSS [12:27:34] hashar: the code itself is completely open source. [12:27:39] but any proprietary brick will be rejected by f-droid :-] [12:27:41] the GCM client libraries are open source. [12:27:47] no proprietary bricks in the code [12:27:49] no Play - you need google xmpp services I guess? [12:27:55] saper: yup. [12:28:06] need to check GCM out [12:28:13] saper: so currently it should crash. But I can easily put a guard in there so that it works, but you just don't get push [12:28:15] been quite happy without google I must say [12:28:27] there is no alternative to efficient push I'm afraid. [12:28:35] it is nice to see on Google Dashboard that last time my device report to Google was in Nov 2012 :) [12:28:59] YuviPanda: I can check if you give me a link to .apk :) [12:29:00] saper: what device, btw? [12:29:04] saper: oh sure. one moment [12:29:29] Jimbo Wales Endorsed(tm) Huawei IDEOS U8050 for Africa [12:29:36] ah [12:29:43] Original Safaricom Edition in Black [12:29:58] goddamn, the Android Commons app won't support that phone. [12:30:03] * YuviPanda reconsiders. [12:30:11] oh yes [12:30:12] I can test [12:30:19] we bought those phones like crazy in Kenya [12:30:27] one guy bought three and he was not the only one [12:30:31] hmmm [12:30:56] saper: the reasoning being in India it is impossible to buy 2.2 phones now [12:30:59] I might be biased [12:31:00] in April last year I fold tfinc about it already - go for it and test there [12:31:17] it ran a decent recent 2.x Android line, don't remember which one [12:31:26] now I am running CM7 on it [12:31:40] CM9 or even CM10 can be tested [12:32:20] saper: the app will definitely run on CM7 [12:32:23] CM7 is based on 2.3 [12:32:39] yes [12:32:55] don't remember what was stock but it was recent [12:33:18] * YuviPanda googles [12:33:29] although one guy gave me his Chinese-Polish tabled he bought mid-December in Poland and stock kernel date (Android 4.1) was 5-Dec-2012 :) [12:33:37] by default it ships with 2.2 [12:33:44] rly [12:33:48] * YuviPanda makes a build of the signpost app [12:33:50] one moment. [12:34:21] saper: https://dl.dropbox.com/u/8768784/WPSignpost-1.1.apk [12:34:34] it should be relatively little work for me to make it work on non-google devices if it does crash [12:34:42] and i'll be happy to do that [12:35:00] heh I copied your link on laptop #1 and wanted to paste on laptop #2 [12:35:11] and I'm happy to test [12:35:30] I am not sure I have a stock ROM yet (sure it's available via xda-devs) but should be tested as well [12:35:58] whenever somebody preaches me about "global outreach" and other shit like this I take out this phone and test their webpage [12:36:19] saper: :) [12:36:34] saper: the question of the commons app is media availability + support [12:36:48] hmm [12:36:56] but the Jimbo phone does have a 3.x MP camera [12:37:06] but I don't have one to test one [12:37:07] *on [12:37:34] hm where's my "adb lolcat" cable [12:37:38] hah [12:37:54] camera is shit on ideos [12:38:05] I barely can read barcodes or qrcode [12:38:12] almost never succeeds [12:38:18] indeed. That's one of the reasons I'm wary of spending time on supporting it for Commons Image Upload [12:39:15] photos can be taken, most of the time [12:39:27] would you consider them commons acceptable quality? [12:40:04] not sure I uploaded something [12:40:11] but yes, for quick stuff why not [12:40:24] hmmm [12:40:38] saper: and audio / video? [12:40:43] I love this phone because unlike most smartphones it feels smaller in my pocket than my old candybar Nokia 6233 [12:40:58] need to try, but works [12:41:03] oh I miss my old candybar and flip phones [12:41:08] the durability [12:41:16] the physical keyboard [12:41:45] and I like it has green and red buttons [12:42:26] hardware [12:43:49] my Nokia is still working very well and is used for my car (handsfree set) and I used it in Serbia where SIM card could not log into the network because CyanogenMod SIM services implementation crashed on something that was on that card; weird. [12:44:06] (so avoid Telenor when in Serbia:) [12:49:38] saper: :D [12:49:42] saper: were you able to try the signpost one? [12:50:32] trying right now [12:50:47] btw Xabber got opensourced recently \o/ [12:51:04] in.yuvi.signpost crashed on startup [12:53:55] do you want "unknown permission" messages on install too? or only the crash [12:55:02] <^demon> Speaking of permissions, I still haven't gotten over the fact that Android has a BRICK permission. [12:55:54] <^demon> "Required to be able to disable the device (very dangerous!)." [12:57:22] YuviPanda: dpaste.de/J8V2r/ sorry for lame line endings [12:57:32] saper: ah :) [12:58:12] ok gotta go [12:59:01] saper: thanks. Expected error. Shall I email you when I've a 'fixed' build available? [13:37:35] saper: sent you a new build. Thanks :) [13:37:38] * YuviPanda runs off now [14:10:38] ^demon: since rename-project won't make into our gerrit upgrade ... should I send you the plugin, so we can at least use that for now? [14:10:50] <^demon> Yeah, send it to me. [14:10:54] k [14:11:20] <^demon> Have you started working it into core yet? [14:12:01] Yes of course I started :-) [14:12:09] It's way easier there [14:12:20] Just one schema ... no injection of schema handlers etc [14:13:07] <^demon> *nod* Sooner that's on gerrit-review, we can start the nitpicking by upstream ;-) [14:19:06] hashar, ^demon - We should have Jenkins run Scribunto unit tests one of these days. [14:20:21] oh true [14:22:23] anomie: I got some jobs generated for it [14:22:35] but I guess it requires us to install the php extension [14:22:54] (the php extension does have a build job attached to it iirc) [14:22:59] <^demon> Well building the php extension might make a nice job too :) [14:23:11] or maybe I just dreamed about it [14:23:23] hashar- Well, most of the tests are run twice, once for each engine. If you don't have the PHP extension, it *should* just skip all those tests. I should check that though. [14:23:28] <^demon> Ah, php-luasandbox-build. [14:23:37] ah [14:23:44] But it would be nice to run them against both engines, Just In Case� [14:24:05] <^demon> Is it packaged yet? [14:24:14] <^demon> I assume Tim packaged it for putting on the apaches. [14:24:19] so yeah the luasandbox is being build https://integration.mediawiki.org/ci/job/php-luasandbox-build/11/console [14:24:49] ideally I would have a post merge job that build the extension and publish it somewhere [14:25:00] the run the Scribunto unit tests using that extension buil [14:25:01] d [14:27:28] <^demon> Since it's all being run by the cli, you could get by with having it dl() the library before running tests. [14:27:35] yeah [14:27:49] or php -d/path/to/luasandbox.so phpunit.php [14:28:06] <^demon> That too. Plenty of options. [14:28:21] Yes, if the PHP extension isn't installed, it correctly skips the tests. So that's good, at least. [14:28:24] lets get the unit test to be triggered [14:29:01] <^demon> fwiw, we *do* have php-luasandbox in the wikimedia apt repos. [14:29:04] <^demon> Just checked. [14:29:34] New patchset: Hashar; "test pipeline for mw/ext/Scribunto" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/48125 [14:29:44] Change merged: Hashar; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/48125 [14:30:31] ^demon: ah so we can test out on the production one [14:30:37] <^demon> Yup. [14:30:41] ideally we could run the tests using the production one AND the latest master version [14:31:42] LUA / Scribunto is definitely a project I should have looked at [14:32:26] anomie: Jenkins claims to test something https://integration.mediawiki.org/ci/job/mwext-Scribunto-testextensions-master/2/testReport/(root)/ [14:32:57] retriggering against latest master [14:33:12] console https://integration.mediawiki.org/ci/job/mwext-Scribunto-testextensions-master/3/console [14:33:26] Tests: 401, Assertions: 372, Skipped: 44. [14:33:42] test report https://integration.mediawiki.org/ci/job/mwext-Scribunto-testextensions-master/3/testReport/(root)/ [14:33:50] anomie: RESOLVED FIXED. [14:33:58] anomie: still have to find out why some tests are skipped [14:34:22] Scribunto_LuaSandboxInterpreterTest::testRoundtrip with data set #6 (array()) [14:34:22] 14:33:24 interpreter not available [14:34:23] ah [14:34:23] hmm [14:34:37] It didn't find the luasandbox extension, for whatever reason [14:34:53] oh [14:34:56] So it skipped running the tests that need that. [14:35:01] interpreter == PHP Extension ? :-] [14:35:18] For Scribunto_LuaSandboxInterpreterTest, yes [14:35:35] (there's also Scribunto_LuaStandaloneInterpreterTest, of course) [14:36:16] !g Id938b8e7fef6465b41d74c12d86fd86c385c683b [14:36:16] https://gerrit.wikimedia.org/r/#q,Id938b8e7fef6465b41d74c12d86fd86c385c683b,n,z [14:36:33] YEAHHH [14:36:54] Someday maybe I should refactor the interpreter tests like I did the engine tests, so it will skip them all with one message instead of 40-some. [14:40:19] I have installed the PHP extension [14:44:39] OK (716 tests, 743 assertions) !!! [14:44:46] https://integration.mediawiki.org/ci/job/mwext-Scribunto-testextensions-master/4/console [14:44:55] though the stupid job still run the tests twice [14:44:56] grmblblb [14:46:16] ahh [14:46:20] siebrand doing style stuff [14:46:24] on DefaultSettings.php [14:46:25] https://gerrit.wikimedia.org/r/#/c/45165/4/includes/DefaultSettings.php,unified [14:46:28] \O/ [14:55:32] <^demon> This always makes me happy when I'm compiling gerrit. [14:55:39] <^demon> [INFO] --- maven-compiler-plugin:2.3.2:compile (default-compile) @ gerrit-gwtui --- [14:55:39] <^demon> [INFO] Nothing to compile - all classes are up to date [15:00:11] ^demon: seeing this especially with gerrit-/gwtui/ is nice. Since we talked about speeding up compiling those permutations ... [15:00:42] <^demon> I think the only way to limit permutations is limit the number of platforms you target. [15:00:48] Adding in gerrit-gwtui/src/main/java/com/google/gerrit/GerritGwtUI.gwt.xml turn 6 permutations down to 1. [15:01:16] <^demon> Well, wouldn't that break gerrit for lots of users? [15:01:27] I use it just for testing ... locally [15:01:33] <^demon> Ah, ok. [15:01:38] We should not use that for production. Obviously :-) [15:01:45] <^demon> That's what I thought ;-) [15:02:07] New review: Hashar; "I am bringing the issue on wikitech-l so the community agree on the URL we wanna use." [integration/jenkins] (master); V: 0 C: -2; - https://gerrit.wikimedia.org/r/39212 [15:02:14] But since I had to to quite some gwtui rebuilding, I looked it up and thought to let you know nontheless [15:05:25] <^demon> Good to know. I think there's a command line option you can use to force a single platform. [15:05:55] <^demon> https://gerrit-dev.wmflabs.org/r/#/admin/plugins/ - installed move-project [15:06:29] \o/ [15:07:47] <^demon> Just tried it out, moved Foobar to Foobaz. [15:08:36] Looks good to me [15:09:23] Btw. do we also want a ui for gerrit core implementation of project renaming? [15:09:50] <^demon> UI would be nice, yeah. [15:10:00] Ok. More gwt fun for me :-) [15:10:58] <^demon> For core, we should probably make it it's own permission rather than use AdministrateServer. I dislike those kind of catch-all permissions (Project Owner is another). [15:11:16] Yes. I'll adapt that. [15:11:35] But currently there was just CREATE_PROJECT (or something the like) [15:11:40] and that was too weak. [15:12:03] So I thought better to be safe for now and fine-tune later [15:12:13] <^demon> Yeah, for the plugin it's fine. [15:12:25] Do we want events for renaming projects as well? [15:12:55] <^demon> Hmm, don't know. I could see the argument for it. Maybe leave the question unanswered (but ask) when committing. [15:13:11] * qchris puts that on his notes [15:13:19] <^demon> "Hey, do we want events on this while we're here?" :) [15:14:08] I can see someone answering "yes" already. [15:14:27] <^demon> Bah, the reviewnotes stuff was just merged. [15:14:43] <^demon> Will need to deploy that plugin when we update core past that point. [15:19:43] New review: Hashar; "Note, I CR-2 to prevents me from merging that change in :-]" [integration/jenkins] (master); V: 0 C: -2; - https://gerrit.wikimedia.org/r/39212 [15:21:39] New review: Hashar; "Yeah!! Thank you to have taken care of that madness. When I cloned codesniffer I should have really..." [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48109 [15:26:46] ^demon: so earlier I was talking about how OpenStack has automatized a bit their Gerrit system. [15:26:56] ^demon: one fun thing is that they have the Gerrit per project access list in puppet [15:27:00] ^demon: example: https://github.com/openstack-infra/config/blob/master/modules/openstack_project/files/gerrit/acls/openstack-infra/git-review.config [15:27:19] ^demon: so if one want to change the Gerrit rights, he simply edit the file and submit the change for review :-] [15:29:53] <^demon> Or, you can just edit it from the gerrit UI and submit for review. [15:29:57] <^demon> (From 2.5+) [15:32:46] ahh [15:33:04] :-D [15:33:13] no need for puppet thus [15:33:18] i can't wait for the deployment [15:40:37] New patchset: Hashar; "GlobalFunctions.php conditionally uses bcmath functions in wfBaseConvert since 9b9daad" [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/42962 [15:41:08] Change merged: jenkins-bot; [mediawiki/tools/code-utils] (master) - https://gerrit.wikimedia.org/r/42962 [16:11:35] damn [16:11:35] forgot to rekey jenkins [16:24:19] ori-l, why am I a reviewer on https://gerrit.wikimedia.org/r/#/c/48066/ ? since it's a cherrypick of an already merged revision you should feel free to merge and push it whenever you can [16:24:56] New patchset: Hashar; "globa ExtraSettings.php was not being loaded" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48132 [16:24:56] New patchset: Hashar; "(bug 43059) $wgShowExceptionDetails = true" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48133 [16:25:32] Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48133 [16:25:32] Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48132 [16:41:55] New patchset: Hashar; "actually inject ExtraSettings.php" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48135 [16:42:10] Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48135 [16:43:07] I am sometime not paying attention :( [16:49:42] New patchset: Hashar; "ant: fakelocalsettings start with ' Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48136 [16:52:24] New patchset: Hashar; "ant: get rid of empty newlines in fake localsettings" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48139 [16:52:53] Change merged: Hashar; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/48139 [17:17:24] heading out, have a good weekend [18:20:38] ori-l: What is the status of: https://gerrit.wikimedia.org/r/#/c/48052/ ? [18:23:33] hey jdlrobson got a moment? maybe you are midmeeting or something [18:32:08] Krinkle|detached: What's the status of: https://gerrit.wikimedia.org/r/#/c/42329/ [18:44:50] ori-l: What is the status of: https://gerrit.wikimedia.org/r/#/c/45257/1/includes/OutputPage.php [20:21:19] preilly: thanks for the reviews! [21:22:13] ori-l: np [21:22:18] ori-l: Are you in the office today? [21:56:18] preilly: yes, hi [21:56:40] ori-l: cool can you answer the questions that I asked earlier? [21:57:22] preilly: You mean re: the status of various patches? [21:57:33] What specifically? [21:57:50] yes status of patches? [21:58:05] Right, but what do you want to know that isn't being indicated in Gerrit? [21:59:42] Sorry, I guess it's not entirely clear what you're asking. [22:06:26] ori-l: I was just looking at: https://gerrit.wikimedia.org/r/#/q/owner:+%2522Ori%2522+status:+%2522open%2522,n,z is all [22:06:49] Search for owner: "Ori" status: "open" [22:06:54] I think you answered your own question :) [22:08:04] preilly: re: the python-jsonschema merge, did you also upload the package to the reprepo queue? [22:08:47] ori-l: not yet was waiting for paravoid [22:09:01] K, np, thanks! [22:28:34] ori-l: I can poke faidon if you want [22:28:44] ori-l: I need to catch up with him for another python module packaging [22:29:34] ok I am half an hour late :-] [22:29:36] damn irc [22:29:38] *wave* [22:36:11] * kaldari waves [22:36:31] * rmoen waves [22:36:45] jsduck config: https://gist.github.com/Krinkle/912693e9ed199029eae5 [22:36:49] hello? [22:36:52] am I in a vacuum? [23:03:27] jsduck jquery types: https://gist.github.com/Krinkle/88158a6dea278e59cbde [23:03:28] spagewmf: