[00:06:41] New patchset: Krinkle; "phpcs: Clean up settings" [mediawiki/tools/codesniffer] (master) - https://gerrit.wikimedia.org/r/58637 [00:07:38] New patchset: Krinkle; "phpcs: Ignore *.i18n.php (installer)" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58638 [00:08:00] Change merged: Krinkle; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58638 [00:08:43] Change merged: Krinkle; [mediawiki/tools/codesniffer] (master) - https://gerrit.wikimedia.org/r/58637 [00:13:51] New patchset: Krinkle; "Update mwcodesniffer submodule" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/58639 [00:14:03] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/58639 [00:20:34] New patchset: Krinkle; "Gitignore: Add beta- and mediawiki-vagrant- workspace" [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/58640 [00:20:52] Change merged: Krinkle; [integration/jenkins] (master) - https://gerrit.wikimedia.org/r/58640 [01:57:51] New patchset: Krinkle; "Add mwext-VisualEditor-doc-test to check pipeline" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/58653 [01:58:01] Change merged: Krinkle; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/58653 [02:03:09] New patchset: Krinkle; "Revert "Add mwext-VisualEditor-doc-test to check pipeline"" [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/58654 [02:03:50] New review: Krinkle; "And reverting again.. doc-test needs to install MW and exec a local shell file stored in the ve repo..." [integration/zuul-config] (master) C: 2; - https://gerrit.wikimedia.org/r/58654 [02:03:50] Change merged: Krinkle; [integration/zuul-config] (master) - https://gerrit.wikimedia.org/r/58654 [08:01:13] New patchset: Zfilipin; "Updated watir-webdriver Ruby gem" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58672 [08:41:15] New review: Hashar; "Probably fixed https://bugzilla.wikimedia.org/show_bug.cgi?id=46286" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58638 [08:42:15] New patchset: Hashar; "stop cleaning mw/core workspace to save I/O" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58502 [08:42:59] New review: Hashar; "deploying" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58502 [08:54:34] New review: Krinkle; "It sounds safe in theory, but in principle it feels wrong." [integration/jenkins-job-builder-config] (master) C: -1; - https://gerrit.wikimedia.org/r/58502 [08:55:51] New review: Krinkle; "[jenkins]" [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58502 [09:13:22] I think I am going to restart Jenkins since it is really quiet right now. Takes roughly half an hour. [09:24:27] New review: Hashar; "deployed" [integration/jenkins-job-builder-config] (master) C: 2; - https://gerrit.wikimedia.org/r/58502 [09:24:28] Change merged: Hashar; [integration/jenkins-job-builder-config] (master) - https://gerrit.wikimedia.org/r/58502 [11:28:50] hashar: do you have a minute? [11:29:46] zeljkof: kind of :-] [11:29:56] just a quick question :) [11:30:24] how come page that does not exist at test2 has link "create"? for example http://test2.wikipedia.org/wiki/Hashar [11:30:42] but it does not have the link at wmf labs? [11:30:50] example: http://en.wikipedia.beta.wmflabs.org/wiki/Hashar [11:31:08] I guess it has different configuration [11:31:14] by default there is a [create] tab [11:31:25] on the english wiki anonymous users are not allowed to create a new page [11:31:28] can you change the configuration? [11:31:33] test2 != enwiki [11:31:38] so test2 uses the default setting [11:31:46] I see [11:31:50] and enwiki on beta uses the english wikipedia setting which is to disable anonymous page creation [11:32:03] makes sense [11:32:20] but we need the link for tests [11:32:26] though that might be a bit troublesome for you :-] [11:32:26] what should we do? [11:32:39] need them where? on the beta english wiki ? [11:33:00] since the conf is the same as in production, that is not really possible or beta will diverge from prod [11:33:07] what do you want to test? [11:33:08] ok, makes sense [11:33:28] we have a test for creating pages by anons [11:33:38] and it works on test2 [11:34:01] can you create a wmf labs site that we will be able to customise as needed for tests? [11:34:47] selenium.beta.wmflabs.org or something like that [11:36:27] that would be essentially like test2 isn't it ? [11:37:44] for start :) [11:37:58] but we would need all kinds of extensions there in the future [12:59:16] can it be possible that position => top for resource loader is broken? [12:59:22] I'm seeing flash of unstyled content [14:01:51] can any one here set up a new labs project ? [14:10:21] Oren_Bochman: I can, if you ask me in #wikimedia-labs :) [15:10:46] New review: Cmcmahon; "maintenance" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58483 [15:10:46] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58483 [15:11:33] New review: Cmcmahon; "maintenance" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58672 [15:11:33] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58672 [17:06:02] Project browsertests-commons.wikipedia.beta.wmflabs.org-linux-chrome build #1: FAILURE in 2 min 54 sec: https://wmf.ci.cloudbees.com/job/browsertests-commons.wikipedia.beta.wmflabs.org-linux-chrome/1/ [17:21:51] Project browsertests-en.wikipedia.beta.wmflabs.org-linux-chrome build #1: FAILURE in 6 min 44 sec: https://wmf.ci.cloudbees.com/job/browsertests-en.wikipedia.beta.wmflabs.org-linux-chrome/1/ [17:21:58] New patchset: Zfilipin; "Deleted unused page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58710 [17:21:59] New patchset: Zfilipin; "Extracted common step into background" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58711 [17:21:59] New patchset: Zfilipin; "Whitespace improvements" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58712 [17:21:59] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58713 [17:21:59] New patchset: Zfilipin; "Simplified code for going to login page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58714 [17:22:00] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58715 [17:22:00] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58716 [17:22:00] New patchset: Zfilipin; "Refactored pages to use only one MediaWiki URL" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58717 [17:22:00] New patchset: Zfilipin; "Simplified checking for page text" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58718 [17:22:01] New patchset: Zfilipin; "Added new MediaWiki sites" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58719 [17:22:56] chrismcmahon: the first complete test run at wmf labs: https://wmf.ci.cloudbees.com/job/browsertests-en.wikipedia.beta.wmflabs.org-linux-chrome/lastCompletedBuild/testReport/ [17:23:20] looks like exactly 50% of the tests fails (22 of 44) [17:24:38] zeljkof: thanks! that's great [17:24:53] chrismcmahon: I have pushed the commits to gerrit [17:25:09] I will add you to reviewers as soon as I create translatewiki job [17:25:41] by the way, commons job is created: https://wmf.ci.cloudbees.com/job/browsertests-commons.wikipedia.beta.wmflabs.org-linux-chrome/ [17:25:56] and the first bug is created: https://bugzilla.wikimedia.org/show_bug.cgi?id=47129 [17:29:30] Project browsertests-sandbox.translatewiki.net-linux-firefox build #1: SUCCESS in 1 min 18 sec: https://wmf.ci.cloudbees.com/job/browsertests-sandbox.translatewiki.net-linux-firefox/1/ [17:30:26] chrismcmahon: and translatewiki tests are all green :) https://wmf.ci.cloudbees.com/job/browsertests-sandbox.translatewiki.net-linux-firefox/1/testReport/(root)/ [17:31:02] \o/ \o/ \o/ \o/ \o/ \o/ zeljkof [17:31:29] I think that is all sites we use (for now) [17:31:44] now to get all tests to green again... [17:48:36] New review: Cmcmahon; "delete unused page (legacy from original PoC)" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58710 [17:48:36] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58710 [17:49:42] New review: Cmcmahon; "refactor, DRY" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58711 [17:49:43] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58711 [18:02:28] zeljkof: Hi! Are you available now to work with me? [18:04:07] rachel99: sure, will send you private message [18:16:50] zeljkof: the upload wizard bug you opened is because you points to commons.wikiPedia.beta.wmflabs.org [18:17:12] hashar: let me check :) [18:17:19] zeljkof: should be wikimedia :-] [18:17:19] https://bugzilla.wikimedia.org/show_bug.cgi?id=47129 [18:17:36] my Safari console gave me an error message I have pasted in bug report [18:17:46] hashar: ooops, chrismcmahon is to blame, he gave me the url :) [18:17:57] but how does the site even open? [18:18:04] feel free to open another bug report to get the Pedia one to redirect to Media :-] [18:18:27] there is a wildcard DNS entry so anything that is beta.wmflabs.org is served [18:18:49] internally I guess we just check for the lang and if lang == commons we consider we are on commons regardless of the project name (wikimedia, wikipedia, wikisource ) [18:19:13] hashar: but now http://commons.wikipedia.beta.wmflabs.org/ did redirect me to http://commons.wikimedia.beta.wmflabs.org/ [18:22:13] oops [18:22:16] RoanKattouw or anyone, I'm confused by addModule{s,Script,Styles}(). If a page has CSS that works in the absence of JS, addModules() does not seem to load it. [18:22:19] maybe some config issue so [18:22:51] New review: Cmcmahon; "whitespace fixes" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58712 [18:22:51] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58712 [18:22:55] Should I then call both addModuleStyles() and addModuleScripts loading the same module, or should I have two separate modules? [18:23:27] spagewmf: Ideally have separate modules [18:23:39] zeljkof: I need to test again [18:23:39] If the CSS is intended to work even when there's no JS, put it in its own module and use addModuleStyles [18:24:25] hashar: anyway, upload wizard works now on commons, I have marked the bug as resolved [18:24:55] New review: Cmcmahon; "use proper RandomPage in language test" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58713 [18:24:58] RoanKattouw, thx, I will add something to mediawiki.org. is there a naming convention for css/js only modules. [18:25:18] Not that I know of [18:30:06] zeljkof: and I found another bug on beta :D http://bug-attachment.wikimedia.org/attachment.cgi?id=12083 [18:31:07] hashar: good one :) [18:31:23] zeljkof: I love how beta let us catch more and more bugs. [18:31:59] hashar: yes, beta is great [18:32:27] zeljkof: specially now that a matrix job upgrade all the wikis for us :-] [18:35:08] zeljkof hashar getting a false failure from lint and jenkins https://gerrit.wikimedia.org/r/#/c/58713/1 [18:35:33] lint misread ":class" [18:36:17] features/support/pages/random_page.rb:8: warning: else without rescue is useless [18:36:17] features/support/pages/random_page.rb:16: syntax error, unexpected keyword_end, expecting $end [18:36:18] chrismcmahon, hashar: I have noticed that, but I did not understand why it is failing [18:36:31] different ruby 1.9 version? [18:36:38] gallium should use 1.9.3 [18:36:41] zeljkof hashar that ":class" identifies a page element, it is not the beginning of a Ruby class() with no "end" statement [18:36:54] just a dumb regex mistake from lint [18:37:13] that is ruby -c [18:38:19] hashar zeljkof the line causing the false failure in lint is "a(:uls_trigger, class: 'uls-trigger')" [18:39:10] it is not a Ruby class, it is an identifier for a page element. [18:41:23] and where is your begin ? :-] [18:41:23] you get else end [18:41:23] but no begin [18:42:50] zeljkof: ^^ [18:43:15] hashar, chrismcmahon: I am pairing with rachel99, can not look now :) [18:43:25] New review: Hashar; "(1 comment)" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58713 [18:43:34] filled my comment on the change [18:46:29] New patchset: Zfilipin; "Fixed link to Commons" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58733 [18:47:06] thanks hashar [18:47:44] if you want to help, feel free to donate to Wikimedia 8-]] [18:47:57] chrismcmahon: honestly the syntax check might sometime be hard to catch [18:48:13] since they would report an unrelated line number (often at the end, far from the initial error) [18:48:21] I am sure you can do something different than begin / rescue [18:48:34] maybe env['FOOBAR'] || 'default there' [18:48:40] my ruby is rusty [18:49:40] hashar: I was committing a lot of changes one by one, looks like I forgot to commit a line in this commit [18:51:36] :-D [18:59:21] YuviPanda: were you looking for UW? [18:59:35] oh hai chrismcmahon [18:59:50] was just helping out with the review backlog, so wondering if there is a test environment on labs or somethnig where I can test patches [19:00:19] YuviPanda: not really patches, UW on beta runs master, test2wiki runs the last deployed [19:00:33] yeah, so I guess betalabs isn't really meant for this [19:00:37] and I should setup something on labs [19:00:47] that is something EVERYONE wants but we haven't figured out yet [19:01:00] heh [19:01:17] well, just give everyone ever root on all the betalabs instances and everything will be *fine* :P [19:01:21] * YuviPanda runs away before someone hits him [19:01:39] I think that's how it's supposed to be... [19:01:44] If you've access, you've got root access [19:01:49] everyone root on betalabs? [19:02:23] ja [19:02:29] same for any labs project [19:03:51] Reedy: while you're around, does this mean anything to you? https://bugzilla.wikimedia.org/show_bug.cgi?id=47065 [19:05:00] bleugh, internet explorer [19:05:04] yeah [19:06:00] Reedy: it seems to be just test2wiki but it makes me nervous. [19:06:33] biab [19:16:07] Did bugzilla die? [19:16:57] Bugzilla is down. Wikimedia ops knows. [19:28:36] siebrand: bugzilla up :-] [19:28:45] hashar: cool [19:51:51] qgil_, how many student places do we have for GSOC this year out of interest? [19:52:04] Thehelpfulone, Google will decide [19:52:10] Is it already anounced? [19:52:17] Thehelpfulone, first we need to line up mentors & proposals [19:52:45] So they assess the proposals? [19:53:32] Thehelpfulone, Nischayn see http://www.google-melange.com/gsoc/events/google/gsoc2013 May 6 - 8 [19:54:02] Thehelpfulone, we list and rank proposals, and also tell Google the desired slots we MUST and also COULD have [19:54:20] Thehelpfulone, then they decide where the cut goes [19:54:52] Thehelpfulone, it [19:55:04] 's a very interesting balance between freedom and control, actually [19:55:04] okay, and who's "we" - some of the things could be (to communities) more important than others [19:55:38] Thehelpfulone, the mentors involved, and ultimately the org admins aka Lydia and me [19:56:03] Thehelpfulone, don't worry (now) the list of proposals formally proposed will e public [19:56:18] heh good :) [19:56:28] Thehelpfulone, May 3 [19:57:08] Thehelpfulone, also, we are connecting all proposals to a bug report. If you have something to say make sure it is said there. [20:05:58] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58716 [20:10:27] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58713 [20:12:17] New patchset: Zfilipin; "Simplified code for going to login page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58714 [20:12:30] New patchset: Zfilipin; "Simplified code for going to login page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58714 [20:13:58] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58715 [20:14:10] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58715 [20:15:15] hashar, chrismcmahonafk: I have fixed all commits in gerrit that I have messed up earlier today :) [20:15:54] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58713 [20:16:10] New patchset: Zfilipin; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58716 [20:16:16] New patchset: Zfilipin; "Refactored pages to use only one MediaWiki URL" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58717 [20:19:01] Project browsertests-test2.wikipedia.org-linux-chrome build #275: FAILURE in 17 min: https://wmf.ci.cloudbees.com/job/browsertests-test2.wikipedia.org-linux-chrome/275/ [20:19:01] * zeljko.filipin: Updated Ruby gems [20:19:02] * zeljko.filipin: Updated watir-webdriver Ruby gem [20:19:03] * zeljko.filipin: Deleted unused page [20:19:03] * zeljko.filipin: Extracted common step into background [20:19:04] * zeljko.filipin: Whitespace improvements [20:22:55] Change merged: Ori.livneh; [mediawiki/tools/release] (master) - https://gerrit.wikimedia.org/r/57770 [20:23:05] Project browsertests-test2.wikipedia.org-linux-firefox build #258: FAILURE in 20 min: https://wmf.ci.cloudbees.com/job/browsertests-test2.wikipedia.org-linux-firefox/258/ [20:23:05] * zeljko.filipin: Updated Ruby gems [20:23:06] * zeljko.filipin: Updated watir-webdriver Ruby gem [20:23:06] * zeljko.filipin: Deleted unused page [20:23:07] * zeljko.filipin: Extracted common step into background [20:23:07] * zeljko.filipin: Whitespace improvements [20:24:27] ^demon: The puppetisation of wikibugs change, what's stopping that? https://gerrit.wikimedia.org/r/#/c/53973/3 [20:24:48] We'd like wikibugs in mediawiki-visualeditor, and we were told to make it dependent on that change. [20:25:29] it's been several months now, change created in december 2012. [20:34:34] zeljkof: you are the boss :-] [20:34:44] zeljkof: I think wmf-selenium-bot should not list the changes. [20:34:59] zeljkof: that will reduce the spam there and the jenkins link provide a list of them already. [20:35:16] hashar: stupid mistake while committing large change, I forgot to commit one line, it caused all the failures [20:35:31] hashar: let me see about the bot [20:35:33] New patchset: Reedy; "Add .gitreview and .gitignore" [mediawiki/tools/grabbers] (master) - https://gerrit.wikimedia.org/r/58808 [20:37:12] New patchset: Reedy; "Add .gitreview and .gitignore" [mediawiki/tools/grabbers] (master) - https://gerrit.wikimedia.org/r/58808 [20:37:34] Change merged: Reedy; [mediawiki/tools/grabbers] (master) - https://gerrit.wikimedia.org/r/58808 [20:37:38] ( https://gerrit.wikimedia.org/r/#/c/53973/3 @ ^demon ) [20:38:46] hashar: I am looking at IRC configuration at https://wmf.ci.cloudbees.com/configure [20:38:59] <^demon> Krinkle: Someone to review & merge I suppose. [20:39:00] but I am not sure how to disable posting the changes [20:39:02] <^demon> Nothing from my end. [20:39:31] <^demon> Oh, maybe a rebase, grr. [20:40:09] Krinkle: no one with +2 in that repo has been added as a reviewer [20:40:10] New patchset: Reedy; "Add .gitreview and .gitignore" [mediawiki/tools/Cite4Wiki] (master) - https://gerrit.wikimedia.org/r/58809 [20:40:17] New patchset: Reedy; "Add .gitreview and .gitignore" [mediawiki/tools/bundles] (master) - https://gerrit.wikimedia.org/r/58810 [20:41:14] hashar: I do not see the option what to post [20:41:24] New patchset: Reedy; "Add .gitreview and .gitignore" [mediawiki/tools/dippybird] (master) - https://gerrit.wikimedia.org/r/58811 [20:41:41] zeljkof: maybe that is not possible :-] [20:42:19] Change merged: Reedy; [mediawiki/tools/Cite4Wiki] (master) - https://gerrit.wikimedia.org/r/58809 [20:42:37] Change merged: Reedy; [mediawiki/tools/bundles] (master) - https://gerrit.wikimedia.org/r/58810 [20:42:43] Change merged: Reedy; [mediawiki/tools/dippybird] (master) - https://gerrit.wikimedia.org/r/58811 [20:43:03] zeljkof: found it, wanna hangout? :-] [20:43:45] hashar: I have disabled all failing jobs until I fix them or report bugs [20:43:48] zeljkof: in the IRC notification advanced option there is a drop list named "Channel Notification Message" [20:43:56] hashar: sure, just to get my headphones [20:44:01] k :-] [20:44:38] lol, tools/upload is only accessible by one guy.. [20:52:12] zeljkof: :-] [20:52:28] hashar: thanks for the help [20:52:33] yw [20:52:55] it is not really important anyway, you can fix your template tomorrow! [20:59:56] ^demon: ori-l: Ah, that figures. [21:06:05] hashar: oh yes, tomorrow it is :) it is too late now [21:06:28] zeljkof: yup. have a good night :-] [21:06:38] hashar: good night :) [21:32:11] New patchset: Cmcmahon; "Simplified checking for page text" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58718 [21:35:59] kaldari: #mediawiki please [21:37:01] Alternatively, YuviPanda has a question that I think you're more capable of answering than I [21:37:16] ah [21:37:52] YuviPanda: I don't usually hang out in #mediawiki because it's too noisy, usually in wikimedia-dev and wikimedia-tech [21:38:00] ah, okay [21:38:09] kaldari: if you prefer to keep it here it's okay [21:38:22] kaldari: so I'm testing out UW, specifically the flickr stuff [21:38:33] kaldari: current master doesn't work for me. [21:38:39] kaldari: it tries to make a request to http://api.flickr.com/services/rest/\?\&nojsoncallback\=1\&method\=flickr.photos.licenses.getInfo\&api_key\=0ad1ef75f0079c0d33feaacdc9b00eed\&format\=json [21:38:48] but is marked as 'cancelled' i my network tab in chrome dev tools [21:38:50] and nothing happens [21:38:54] curl works fine on that [21:38:59] is there some CORS involved here? [21:39:07] or... am I missing some configuration? [21:39:21] YuviPanda: one sec, finishing up some deployment clean-up, gimme 5 minutes... [21:39:28] kaldari: sure! [21:40:47] New patchset: Cmcmahon; "Added new MediaWiki sites" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58719 [21:45:51] YuviPanda: no CORS involved, just regular AJAX get request. You may want to try setting the api URL to the secure one in case you're using HTTPS Everywhere or something.... 'flickrApiUrl' => 'https://secure.flickr.com/services/rest/?' [21:46:01] ah, hmm [21:46:02] let me try [21:46:04] i am using it [21:46:24] ori-l: I wonder if we can drop all the old clicktracking tables.. [21:46:28] the URL you posted above isn't the secure one though [21:47:01] kaldari: it isn't. I just copy pasted from the extension page. [21:47:08] Reedy: I'm pretty sure you could [21:47:16] didn't know that httpseverywhere can affect ajax URLs [21:47:26] ori-l: Reedy wouldn't that take some data with it? [21:47:35] "data" [21:48:13] I'll change the version on the Extension documentation page to be the secure one :) [21:48:21] kaldari: :) [21:48:58] ^demon: really our parser tests must die [21:49:10] ^demon: it seems to invalidate the l10n cache before each test :( [21:49:12] 12.86M rows on enwiki [21:49:42] kaldari: hmm, no change [21:49:44] it's still 'cancelled' [21:50:05] is it actually using the secure URL in the request now? [21:50:25] hmm, yes. [21:50:26] oh wait [21:50:37] kaldari: ah, works :) [21:50:41] yeehaw! [21:50:46] kaldari: I had set it to secure. but forgot to make it https :) [21:56:49] kaldari: also UW fails at flickr if TitleBlacklist isn't present. JS errors out and I get nothing. [21:57:06] If that is intended, I'll note that in the extensions page? [21:57:31] YuviPanda: make sure your core and UploadWizard are up to date, they recently moved TitleBlacklist out of core [21:57:42] kaldari: yeah, i'm on latest master of both [21:57:58] hmm, that's sctrange [21:57:59] kaldari: hence the notice about UW now requiring TitleBlacklist extension [21:58:00] stranghe [21:58:06] kaldari: as in, I don't have TitleBlacklist [21:58:07] er,g [21:58:10] and UW doesn't work without it [21:58:15] well, the flickr part at least [21:58:19] that was fixed in UW recently [21:58:49] although maybe something was overlooked [21:59:01] YuviPanda: you should file a bug on it [21:59:10] probably the Flickr part wasn't fixed [21:59:18] ah, right [21:59:20] kaldari: I'll do that [21:59:35] thanks [21:59:38] sorry about that [22:01:05] kaldari: no worries! I'm trying to help reduce the review backlog on UW [22:02:33] kaldari: oh, I was wrong. UW itself doesn't work without titleblacklist :) [22:02:36] will file bug [22:02:46] oh [22:02:47] gah [22:02:48] I could have sworn that was fixed [22:02:48] nevermind [22:02:49] false alarm [22:02:52] false alarm [22:03:05] kaldari: I'm testing one of Nischayn|Away's patches, which re-introduced that bug :) [22:03:07] that is all [22:03:12] ah [22:03:16] master is fine [22:03:23] prolly needs a rebase [22:04:15] kaldari: yeah, you're right. [22:04:20] sorry about the false alarm [22:04:41] NP, thanks for working on the UW backlog! [22:04:49] New review: Cmcmahon; "delete page no longer used" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58713 [22:04:49] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58713 [22:05:43] kaldari: :) [22:05:48] New patchset: Cmcmahon; "Simplified code for going to login page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58714 [22:06:08] New review: Cmcmahon; "delete unused page" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58714 [22:06:09] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58714 [22:06:42] New patchset: Cmcmahon; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58715 [22:07:11] New review: Cmcmahon; "page no longer used" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58715 [22:07:11] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58715 [22:07:30] New patchset: Cmcmahon; "Deleted unnecessary page" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58716 [22:07:42] New review: Cmcmahon; "page no longer used" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58716 [22:07:42] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58716 [22:08:53] New patchset: Cmcmahon; "Refactored pages to use only one MediaWiki URL" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58717 [22:09:03] New review: Cmcmahon; "manage URLs properly" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58717 [22:09:03] YuviPanda, yeah thanks for reviewing nischay's patches... I was going to but have been busy with other things [22:09:04] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58717 [22:09:40] New patchset: Cmcmahon; "Simplified checking for page text" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58718 [22:10:06] New review: Cmcmahon; "use URL correctly" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58718 [22:10:07] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58718 [22:10:17] New patchset: Cmcmahon; "Added new MediaWiki sites" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58719 [22:10:31] New review: Cmcmahon; "new test target URLs" [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58719 [22:10:39] New review: Cmcmahon; "new test target URLs" [qa/browsertests] (master); V: 2 C: 2; - https://gerrit.wikimedia.org/r/58719 [22:10:40] Change merged: Cmcmahon; [qa/browsertests] (master) - https://gerrit.wikimedia.org/r/58719 [22:12:39] Krenair: :) I also went around and merged a few other patches that were super simple but getting no love [22:12:44] think I'll do this for a while [22:13:23] TimStarling: hello [22:13:29] hi [22:13:39] ah [22:13:46] I guess it is time for me to sleep so [22:13:51] have a good day Tim ;-] [22:14:14] hashar: do you have an alarm that greps for 'Tim-away is now known as TimStarling'? [22:15:30] hop but I should :-] [22:16:09] bed for real. Have a nice day [22:38:31] Isarra: your favicon is being sync'd now [22:38:49] Whoo! [22:39:01] Isarra: :) [22:39:15] Isarra: and should be live as soon as bits cache clears [22:39:29] Aye. [22:39:33] It is. [22:39:40] It looks awful. o__o [22:39:50] do you need me to revert? [22:40:02] No, it looked awful before too. >.< [22:40:23] Seriously, man, thanks. [22:40:28] Text is just horrible. [22:41:35] Isarra: What ugly thing did mwalker just replace? [22:41:47] The wikipedia favicon. [22:42:34] \o/ [22:43:32] * marktraceur looks at it funny [22:43:35] Is it different? [22:43:55] It's buried under a lot of cache, so probably not. [22:43:57] Hrm. [22:44:23] https://brionv.com/misc/wikipedia-favicon-hidpi.png <- yep it shows nice and high-res now [22:44:24] Look at all these gwickes. [22:44:33] now we just need to redo all the damn logos on the pages [22:45:54] Browsers support multiple sizes of favicon, but how would a site serve multiple sizes of image within the page content? [22:46:21] Isarra: ahh well there's several ways [22:46:31] one is to use CSS images and @media queries. this way lies madness [22:46:41] Ye-es... [22:46:45] another is to use JavaScript as a temporary hack to implement the 'srcset' attribute [22:46:49] which is what we do for content images [22:46:59] but that doesn't work on the logo cause we use a weird CSS thing for it [22:47:12] Ah... [22:47:22] we'll think of something… the hard part really is actually rebuilding all the fucking images [22:47:29] Since the logo is a background image instead of an image? But there are other background images... [22:47:39] should probably create a makefile that generates SVGs and flattens them at the appropriate res [22:48:24] Doesn't sound like something that would necessarily ship well, since this appears to be a problem with core that would affect anyone using it... [22:48:56] Using mediawiki, I mean. [22:49:01] we'd probably add a second $wgLogoHiDPI param or something hacky [22:49:10] so if you don't set it you just use your existing low-res logo [22:49:32] or we could switch to svg. fuck IE 6/7/8! [22:49:37] Heh. [22:49:44] Aye... [22:55:30] +2 on fucking MSIE [22:55:47] or [22:55:56] we could do the logo in WebGL and make it spin [22:56:02] only work in firefox and chrome \o/ [22:56:56] brion: NaCL :P [22:57:23] NaCL is kinda neat… but I worry about the verification model. it sounds like java for x86, and java's full of holes ;) [22:57:37] kaldari: ping [22:57:40] kaldari: 'Refused to display 'http://localhost/w/api.php' in a frame because it set 'X-Frame-Options' to 'DENY'. ' [22:57:46] when it tries to start uploading [22:58:09] kaldari: and again I get POSTs to localhost/w/api.php that are marked 'cancelled' [22:58:21] oh yeah, you need to change a core config var for that [22:58:31] one minute, in the middle of something else... [22:58:39] kaldari: okay! [22:58:42] * YuviPanda waits [22:58:58] I think there might be note about it on the extension page on MediaiWiki [22:59:03] ah [22:59:05] * YuviPanda looks [22:59:40] what's loading API requests in an iframe? that sounds dreadful [23:00:07] brion: UW [23:00:38] brion: I'm wandering out of our mobile apps silo, and it is fun :) [23:00:42] ugh. doesn't it only do that on like IE 5 that didn't have XHR? [23:00:51] or does it do that for anything without FileReader? :( [23:01:10] brion: possibly. the extension page didn't really have anything... [23:01:20] yeah it's…. lightly documented [23:01:22] YuviPanda: $wgApiFrameOptions = 'SAMEORIGIN'; [23:01:47] kaldari: eek, apparently I can't read [23:01:48] sorry [23:05:42] kaldari: uh, it seems to default to using jsonfm, rather than json [23:05:48] hence causing problems with the returned output [23:06:00] that's strange [23:06:32] ------WebKitFormBoundaryaZHXvPJBFfmqYmkU [23:06:33] Content-Disposition: form-data; name="format" [23:06:33] jsonfm [23:07:02] kaldari: hmm, i found [23:07:11] // we use JSON in HTML because according to mdale, some browsers cannot handle just JSON [23:07:11] _this.addFormInputIfMissing( 'format', 'jsonfm' ); [23:07:22] really? [23:07:29] interesting [23:07:32] ? [23:07:39] kaldari: yeah, that's in mw.ApiUploadHandler.js [23:08:00] kaldari: Gblame tells me it's been around since 2010 - Nov [23:08:07] so I presume that's not causing problems [23:08:10] yea... [23:08:10] mdale: nah, just digging around UW [23:08:16] guess we nede to make sure we're always specifiing the format then [23:08:31] oh right.. I understand the context now [23:08:53] some older browsers IE ( probably ) don't like direct XHR JSON payloads [23:09:06] I usually do, but it could have been overlooked somewhere [23:10:12] nah, I found the problem. [23:10:16] in this instance, at least [23:10:25] it's refusing to upload on my local wiki because the file already exists in commons :P [23:10:33] and... instacommons doesn't le tme re-upload [23:10:35] because duplicate [23:10:36] hmm [23:11:06] yeah, that's a feature, sort of ;) [23:11:19] heh [23:11:55] would be good to have some way to let you override though [23:12:13] special:upload will let you override [23:12:18] but upload wizard just says "NO" [23:12:21] it's very annoying [23:12:21] hehe [23:12:27] I can test by turning off Instacommons [23:14:40] we should call is SlowCommons [23:15:12] if we could make it batch requests it'd be way faster [23:15:16] marktraceur: kaldari I updated https://gerrit.wikimedia.org/r/#/c/42775/ and also gave it V+2 [23:15:19] but that's awkward thanks to the way the parser works [23:15:24] So I can see [23:15:25] since I tested it and it worked [23:15:33] it just needs to support 404 handling [23:15:56] if it did, and people didn't use the default shit php session handler, it would probably be fine [23:16:39] ugh [23:16:44] we should replace the default session handler, yeah [23:16:47] with something non-blocking [23:20:04] I don't think IE 5 supports X-Frame-Options ;) [23:28:03] jdlrobson: CentralNotice is now mobile aware; it won't serve anything without a banner being manually created and then targeted; but the code is sitting there scheming evily [23:28:16] \o/ [23:28:36] the targeting works now? [23:29:10] mwalker: you should write up how you guys got to being mobile aware someplace :) I'm sure other extensions would like to too, at some point :) [23:29:15] mwalker: can we create banners that only target a device? [23:29:19] there's talk of wikibase doing it, for example [23:29:21] brion: is https://gerrit.wikimedia.org/r/#/c/18229/10 ok now? [23:29:35] Merge: that already exists..http://www.mediawiki.org/w/index.php?title=ResourceLoader/Writing_a_MobileFrontend_Friendly_ResourceLoader_module :D [23:29:47] jdlrobson: whoops, nice. [23:29:56] Merge: i made it early this morning so no need for whoops [23:30:00] i predicted the future ;-) [23:30:17] jdlrobson: well, I forwarded it to wikitech-l and then forgot about it :P [23:30:27] * YuviPanda should sleep [23:30:30] jdlrobson: hence the whoops [23:30:46] haha didn't realize you had annoyingly changed your name again :P [23:30:53] if i knew it was you i wouldn't have been so nice ;-) [23:31:09] jdlrobson: hey, I'm the merge totem for e3 and ori-l [23:31:14] or... something like that [23:31:18] yes [23:31:20] thank you [23:31:24] Merge: before you go to bed any ideas how i could plot a lot of long,lat points on a map in mobile web? [23:31:34] jdlrobson: leaflet? [23:31:38] (preferably deferring to a native app) [23:31:52] jdlrobson: ah, so if you want to plot a *single* point, you can do that [23:31:57] * jdlrobson checks up on web intents status [23:32:09] jdlrobson: there is a 'geo:' thing that iOS and Android both read [23:32:09] i want multiple points Merge i think that needs a kml file though :( [23:32:10] (I think) [23:32:40] jdlrobson: possibly, yes. And then you'll just have to fight the different versions :P [23:32:48] jdlrobson: I don't know how you'll do that on iOS, for example [23:32:54] jdlrobson: and I'm not even sure if you can do it with kml on Android [23:33:33] jdlrobson: why not leaflet? [23:33:53] Головна сторінка [23:34:04] Anybody know what that means in Ukrainian ? [23:34:17] xyzram: ask Maryana maybe? [23:34:22] hmm, she's not in this channel [23:34:33] xyzram: but is in #wikimedia-mobile :) [23:34:37] and the office :) [23:34:39] Cork_: tbh i think browsing a web based map on a mobile phone with a 2G/3G connection is broken. Native apps do that so much better. You spend most of your time waiting for tiles to load and watching titles not load [23:34:40] AaronSchulz: looking [23:34:48] OK, thanks. [23:34:57] (it could and should fall back to leaflet though) [23:34:59] jdlrobson: I don't have the UI yet to do the targeting (everything gets targeted to desktop) but that's coming in the next release; I'm about halfway through rewriting the CN UI to use HTMLForms and be all nice and polished [23:35:03] looks good [23:35:19] mwalker: sweet. I'm planning to try out a banner on beta soon so will keep you posted about that [23:35:29] xyzram: did you try google translate [23:35:54] mwalker: actually just realized - central notice will not run on mobile until next tuesday when the new method is deployed :) [23:36:09] awesome; ya, let me know, I'll need to manually add some entries to the DB [23:36:14] mwalker: thanks for fixing the bug i introduced last monday ;-) [23:36:21] *last tuesday [23:36:30] no problem :p [23:36:34] I'm sure our users in poland are happier [23:37:03] mwalker: if all goes to plan you should start seeing different device values appearing as of next tuesday [23:38:09] kk; I have no method of automatically adding them yet so it doesn't particularly matter [23:38:29] bsitu: just did -- it means "Home" ! [23:42:53] YuviPanda: having fun? [23:43:13] mwalker: oh, I was just being the status totem for ori-l's big e3 merge [23:43:44] hehe; a live history played out through irc nicks [23:43:45] I like it [23:43:55] :) [23:44:13] Now I shall go to sleep. 5:15 AM is wonderful time to sleep :) [23:44:21] yes; sleep is good [23:44:35] you're not quite at my definition of morning yet [23:44:54] so it's still a reasonable time to head that way! [23:45:40] mwalker: :) [23:46:05] mwalker: I started reviewing random extensions today, and have found myself unable to tear away [23:46:09] * YuviPanda reviews some UW patches [23:46:27] universal... uh... waterworks! [23:52:59] marktraceur: re self-merges, I guess I shouldn't +2 https://gerrit.wikimedia.org/r/#/c/42775/ [23:53:14] because I had a (small) non-rebase patchset there [23:53:18] is that correct? [23:53:19] mwalker: UW *causes* universal waterworks [23:53:30] haha [23:53:41] YuviPanda: Not to worry, you've caught me at a time when I can merge stuff [23:53:50] aha! [23:54:05] let me see if i can get one thing past [23:54:59] YuviPanda: Meaning I shouldn't merge this patch? [23:55:04] oh no [23:55:05] as in [23:55:07] another patch [23:55:09] feel free to merge this one :)