[00:00:02] ok, night all [00:01:21] 14(WFM) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Sam Reed (reedy)) [00:06:20] saper: I think my internet connection failed while cloning core.git: [00:06:32] Write failed: Broken pipe172928/356196), 29.37 MiB | 6 KiB/s [00:06:34] fatal: The remote end hung up unexpectedly [00:06:36] fatal: early EOF [00:06:38] fatal: index-pack failed [00:06:45] should I just re-run the git clone command? [00:07:02] or will I need to do some cleanup? [00:11:05] * Dantman kicks Chrome [00:12:22] ? [00:14:05] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Seunghyo Chun) [00:16:21] Muahahaha... mediaqueries + transitions [00:19:39] helderwiki: isn't your disk full? [00:20:55] saper: not yet [00:22:19] helderwiki: you can try "cd core" and check "git remote -v && git branch -vv" [00:22:32] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Bawolff) [00:23:33] saper: bash: cd: core: Arquivo ou diretório não encontrado [00:23:35] (aka file not found) [00:23:52] 03(NEW) Thumbnail not fully rendered - 10https://bugzilla.wikimedia.org/35576 normal; MediaWiki: Images and files; (M8R-udfkkf) [00:24:17] git remote -v && git branch -vv [00:24:19] fatal: Not a git repository (or any parent up to mount parent ) [00:24:21] Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set). [00:25:49] helderwiki: sorry, rm -rf core and git clone again [00:26:02] ah [00:26:08] file not found? [00:26:10] sorry then [00:26:21] where was the clone supposed to go? [00:26:25] Cloning into core... [00:26:29] pwd [00:26:35] try [00:26:46] find . -name .git -print -prune [00:27:10] "find . -name .git -print -prune" returned nothing [00:27:23] Hey, I have a question. Is there anyplace I can get a sample LocalSettings.php? My web host just nuked mine, and the mw-config directory doesn't work... [00:27:48] saper: for the record, when I used git clone I as on ~/mwtest [00:28:08] *I was [00:29:39] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Seunghyo Chun) [00:30:35] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Seunghyo Chun) [00:37:09] can anyone tell me when/if the function addExtensionField disapeared? I have an extension Im trying to install which wants this and claims its in version 1.18.2 but it definitely isnt in my fresh installation. [00:37:36] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Bawolff) [00:38:01] Phlunk3: let me check... [00:39:01] Phlunk3: which extension? which version of the extension? [00:40:49] its a non-supported extension, using the 1.2 alpha installation for OpenID and just wanting to mod it, however everything I read about this function says it should exist in 1.18, just thinking maybe its not around yet lol. [00:42:06] is freakolowsky already in gerrit? [00:45:22] added to REL1_18 here https://svn.wikimedia.org/viewvc/mediawiki/branches/REL1_18/phase3/includes/installer/DatabaseUpdater.php?r1=79536&r2=79537& [00:46:25] and it was included in 1.18.1 [00:47:15] Thanks Saper, can be it was removed in 1.18.2 or the download I used was just wrong lol. Got the extension working now regardless :) [00:47:44] nope [00:47:49] I checked wrong thing [00:47:55] addExtensionTable is @since 1.19 [00:48:59] eh [00:49:05] addExtensionField I mean [00:49:17] but you can fake it easily [00:50:37] Phlunk3: for alpha extension, I'd recommend alpha mediawiki; probably the best is current git master [00:51:45] Is there anywhere I can find a skeleton localsettings.php? My web host nuked mine... [00:52:25] 03(mod) Cannot input specific text into editor "inetd -f /etc/inetd.conf" - 10https://bugzilla.wikimedia.org/35575 +comment (10Seunghyo Chun) [00:52:43] matthewrbowker: you can run your installer again easily [00:52:51] saper: It won't let me... [00:53:13] using mysql? [00:53:17] Yes. [00:53:30] It asks for an "upgrade key" then won't let me do anything... [00:53:55] I should mention this is 1.17.0 ... XD [00:54:49] it shouldn't do that if your LocalSettings.php is missing [00:55:11] if you recreated it, you can move it away [00:55:17] TimStarling: It's there, just a 0 byte file. Will I be able to re-install if I remove it? [00:55:24] yes [00:56:33] TimStarling: I think I found a way to fetch all changes pending to do offline review [00:56:58] TimStarling: Thanks ;) [00:57:01] * :) [00:57:21] 03bsitu * 10/trunk/extensions/PageTriage/PageTriage.php: register the hook for article delete to pagetriage [01:04:57] saper: http://blog.redwerks.org/ ;) Open up in say Chrome (FF has a min-width for me) and resize till it's really narrow. [01:06:11] checking [01:06:44] sleek, man! [01:07:01] I like narrow browser windows even on a wide screen [01:07:37] I'm debating whether to separate into max-width and max-height [01:07:50] hmmm... [01:07:54] meh, skip that [01:08:24] saper: ;) The real fun part was transitions. I already had fluid width before. [01:08:27] I downloaded some framework to do this [01:08:33] So who loves gerrit!?! [01:10:02] -moz-transition-property this stuff? [01:10:08] johnduhart: openstack :) [01:10:26] 03raindrift * 10/trunk/extensions/PageTriage/ (4 files in 2 dirs): [01:10:26] made the controls to get the filter dropdown menu. [01:10:26] still need to make the menu itself [01:17:18] New review: Eloquence; "(no comment)" [mediawiki/extensions/UploadWizard] (master) C: 1; - https://gerrit.wikimedia.org/r/3891 [01:37:37] 03(mod) Have Firefox-like Beta and Nightly builds of Wikipedia App in Android Market - 10https://bugzilla.wikimedia.org/35399 (10Yuvi Panda) [01:37:38] 03(NEW) Need Icon for market beta version of Wikipedia Android app - 10https://bugzilla.wikimedia.org/35577 normal; Wikipedia App: Generic; (yuvipanda) [01:53:29] 03(mod) Need Icon for market beta version of Wikipedia Android app - 10https://bugzilla.wikimedia.org/35577 (10Yuvi Panda) [01:55:08] 03(mod) Have Firefox-like Beta builds of Wikipedia App in Android Market - 10https://bugzilla.wikimedia.org/35399 summary; +comment (10Yuvi Panda) [02:24:43] 03(mod) Have Firefox-like Beta builds of Wikipedia App in Android Market - 10https://bugzilla.wikimedia.org/35399 +comment (10Yuvi Panda) [02:51:05] ugh, I just joined a project with a 15GB database of a MW1.16 install [02:51:11] and Im tasked with upgrading it [02:54:13] * TimStarling tries to remember when Wikipedia had such a tiny database [03:09:17] 03(mod) Autolink to new Gerrit / Git changesets and SHA-1 commits - 10https://bugzilla.wikimedia.org/35144 +comment (10Chad H.) [03:14:25] 03(mod) Setup external git mirrors/clones - 10https://bugzilla.wikimedia.org/35429 +comment (10Chad H.) [03:22:19] 03krimpet * 10/USERINFO/krimpet: Updated my name, email, url [03:23:59] All right, one more question... Is the 1.17 database scheme compatable with 1.18? [03:24:23] (AKA can I upgrade and still use the same database with no problems?) [03:29:23] no, you have to run maintenance/update.php [03:31:33] TimStarling: OK, thank you :) [03:37:20] If im upgrading, and the new install has compress reversions, are all the old reversions going to be compressed? and if now, how do I force it? [03:37:37] im going from mw 1.17 to 1.18 [03:40:56] JRWR: no the old revisions won't be compressed [03:41:27] you need to run maintenance/storage/compressOld.php periodically [03:41:38] as well as after upgrade [03:42:32] pity compressOld.php doesn't do HistoryBlobStub [03:46:30] * JRWR is dealing with a 15GB database... [03:47:12] yeah, so you say [03:47:51] if it's mostly text then compressOld.php should reduce its size by 85% or so [03:49:02] 03(mod) Progress Icon for Listen In - 10https://bugzilla.wikimedia.org/35573 +comment (10Tony Cheng) [03:59:24] 03(mod) Setup external git mirrors/clones - 10https://bugzilla.wikimedia.org/35429 +comment (10Krinkle) [04:03:05] 03(mod) Setup external git mirrors/clones - 10https://bugzilla.wikimedia.org/35429 (10Krinkle) [04:03:05] 03(mod) Implement a way to bring Github pull requests from mediawiki-core forks on Github into Gerrit - 10https://bugzilla.wikimedia.org/35497 summary; +comment (10Krinkle) [04:03:19] 03(mod) Setup external git mirrors/clones - 10https://bugzilla.wikimedia.org/35429 (10Krinkle) [04:03:24] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10Krinkle) [04:10:12] I am upgrading from 1.18 to 1.18.2 on my wiki farm and I've done a lot of work in the MediaWiki name space (interface messages) is the upgrade going to overwrite the system messages I've changed ? [04:11:12] :P I guess that depends on how the upgrade is done [04:12:24] normally I just move LocalSettings.php and extensions and image directory to the new directory [04:14:28] Mlpearc: no, there's no way the system messages can be overwritten [04:14:58] Thank you sir :) [04:18:35] * Mlpearc this is like asking a question about a Guitar and having Les Paul answer you :) [04:50:55] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3895 [04:50:57] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3895 [05:00:25] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3874 [05:00:27] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3874 [05:02:34] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3879 [05:02:37] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3879 [05:04:59] New review: Aaron Schulz; "Has https://bugzilla.wikimedia.org/show_bug.cgi?id=35572 been ironed out?" [mediawiki/core] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3841 [05:32:26] question about 1.18.2: was a bug fixed that allowed non-autoconfirmed users to edit user talk pages then move them? [05:32:38] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3889 [05:32:40] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3889 [05:35:21] because I have a 4-day autoconfirm delay, and a spammer just create an account yesterday, edited his talk page, and moved it to another page, within 3 minutes of account creation [05:35:53] oh, never mind, it appears emailconfirmed users are getting edit permissions [05:36:53] how can I require both emailconfirmed and autoconfirmed for edits? [05:47:13] danielko: create another auropromote group [05:48:02] danielko: see http://www.mediawiki.org/wiki/Manual:$wgAutopromote [05:52:20] 03a_engels * 10/trunk/wikistats/squids/SquidReportArchive.pl: Phil's request [05:53:02] $wgAutopromote['autoemailconfirmed'] = array(APCOND_INGROUPS, 'autoconfirmed', 'emailconfirmed'); <-- does this look right? [05:59:19] New patchset: Santhosh; "Fix Bug 32748 - Printer friendly version of article must encode URL in unicode." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3897 [06:00:09] varnent: hey [06:00:24] nischayn22: greetings [06:01:08] varnent: are there specific slots for SMW projects in gsoc ? [06:01:58] 03(mod) Printer friendly version of article must encode URL in unicode. - 10https://bugzilla.wikimedia.org/32748 +comment (10Santhosh Thottingal) [06:02:11] nischayn22: no - they applied - and we're hoping they get in next year or into another summer coding session - and folks submitting SMW ideas get equal chance as any other WMF submission - but there aren't designated slots for SMW per se [06:03:36] varnent: but as SMW will never be used in any WMF projects, does such a project have less chances than other WMF ones ? [06:04:49] nischayn22: not necessarily - no [06:05:02] nischayn22: SMW has value to the WM community - which WMF has recognized many times [06:05:38] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3886 [06:05:41] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3886 [06:06:27] varnent:could you just elaborate how it is beneficial for WM ? I also see a new project being developed for WMF called wikidata [06:08:05] nischayn22: I personally stay out of that conversation - so I won't speak to that personally - but there are conversations on MW and elsewhere discussing it in-depth [06:08:49] nischayn22: broadly speaking - tools that are used heavily by the broader WM community has been recognized as having value to the MW community [06:09:31] varnent: ok, thanks [06:10:16] nischayn22: you're welcome [06:16:54] 03(mod) "File links" heading on File pages should include pagelinks to this file ([[:File:Example.jpg]]) - 10https://bugzilla.wikimedia.org/30126 +comment (10Martin Guy) [06:18:18] of note is that the wikidata team includes the primary developer of SMW btw... [06:21:08] nischayn22: thedj brings up a good point to know [06:22:22] varnent: thedj I know that, that's why I am concerned if SMW developers will have time to mentor a gsoc project [06:23:06] nischayn22: my understanding is since they wanted to be a part of GSOC - there's a commitment and interest to mentoring some projects [06:24:00] varnent: actually they are busy with wikidata or something and so the channel and mailing lists are very inactive [06:24:30] nischayn22: that may be - but I know they have an interest in GSOC [06:24:59] nischayn22: they've made that known :) [06:25:04] varnent: any idea when they start to code for wikidata ? [06:25:33] nischayn22: I don't - but there are folks in and out of this room who are involved and have more direct knowledge [06:25:47] I hope it doesn't overlap much with GSOC too much [06:28:08] nischayn22: probably depends on the developer - some are excellent at multi-tasking projects [06:28:44] varnent: yes that is true and wikidata coding will start in April [06:29:28] New review: Nikerabbit; "messages.inc was not updated" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3886 [06:33:21] I'm getting autoconfirmed waaaay before wgAutoConfirmAge [06:34:16] $wgAutoConfirmAge = 345600; # 60*60*24*4; # autoconfirms after 4 days [06:34:49] created an account 2 hours ago, it's autoconfirmed already, didn't change the promote condition for autoconfirmed [06:37:09] could someone help me out.. i am not able to install commit hook 'commit-msg' under my hooks directory [06:40:10] uh, never mind, I was looking at the wrong account... [06:42:55] chughakshay16: how are you trying it and whats the error? [06:43:49] mutante, when i try to tun this git-review -s.. [06:44:40] mutante, its telling me ' could not open a connection to your authentication agent' [06:44:47] *run [06:45:28] mutante, i am following the steps given at this page [06:45:44] mutante, https://www.mediawiki.org/wiki/Git/Workflow [06:46:12] chughakshay16: oh, that sounds different from adding the hook though? [06:46:39] mutante, i m just stuck with this last step.. please help me out [06:47:07] mutante, i dont understand why its not working . [06:47:08] chughakshay16: well about the hook i was about to show you this: https://labsconsole.wikimedia.org/wiki/Git#Git_configuration_required_to_commit , see the command line that starts with "curl .." there [06:47:30] mutante, i havent reached to that point yet :| [06:47:41] chughakshay16: but now this sounds different, like your ssh-agent is stopped. do you see your key with "ssh-add -l" ? [06:48:07] mutante, no i dont [06:48:24] chughakshay16: try adding it wit "ssh-add " [06:48:37] mutante, okay [06:50:28] mutante, no its throwing the same message again ' could not connect to your authentication agent' [06:53:02] mutante, it could be because my key is not in OpenSSH format.. [06:53:11] mutante, its a .pub file [06:53:40] chughakshay16: ssh-agent /bin/bash ; ssh-add .. [06:54:40] chughakshay16: check if it's running, ps aux | grep ssh-agent [06:56:04] mutante, what should i do with ssh-agent /bin/bash [06:56:08] mutante, ? [06:57:15] chughakshay16: that was in case it is not running, to fork a new shell and start it, other option: restart your X [06:58:08] mutante, to be honest i dont understand what you are saying.. [06:58:13] mutante, :| [06:59:25] chughakshay16: then just try to log off and login again first [06:59:50] chughakshay16: on your local workstation, Gnome, or whatever you use [07:01:46] mutante, i did this eval 'ssh-agent' [07:02:10] https://www.google.com/search?q=could+not+connect+to+your+authentication+agent [07:04:18] chughakshay16: i dont think its related to your key format, even though that might be an additional problem, but first you should get different errors from "could not connect.." [07:06:56] chughakshay16: so if even a reboot does not change it, would have to ask why the OS/desktop/terminal you are using is not starting it [07:07:54] mutante, i am going through the search results for 'could not connect to authentication-agent' [07:14:49] 03(NEW) IPs hidden without any log or apparent reason - 10https://bugzilla.wikimedia.org/35578 normal; MediaWiki: Revision deletion; (snowolf) [07:15:53] chughakshay16: and about the key format: doesnt matter if the filename ends in .pub ,just the contents matter, it should just start with "ssh-rsa" (or "ssh-dsa" and be a all on a single line, if it starts with something like "-- BEGIN SSH2 PUBLIC KEY".. then you should be able to convert it "ssh-keygen -i -f old_key > new_key" or similar. bbl [07:16:35] mutante, no it starts with rsa [07:16:54] mutante, would reboot my system,, i guess thats the only option left [07:17:40] chughakshay16: did you ever see "ssh-agent" in the process list though? yea, try it once i guess [07:18:23] if it fails please provide more details about what you are using, and/or try the support channel for your distro here on freenode as well [07:18:23] mutante, i tried eval 'ssh-agent' [07:18:59] mutante, its displaying me the pid for that agent [07:20:56] mutante, i m quite confused about all this .. i just did all the steps stated in that git/workflow document.. and still couldnt connect to my authentication agent [07:21:11] mutante, and now i see for google results they are even more confusing .. [07:21:16] chughakshay16: hmmmm.. if you have any other box to test, i'd go back to trying a regular ssh login there, using ssh-agent, and only after you see that working go back to the git instructions [07:21:24] mutante, each one of them has something different to say [07:22:02] mutante, this is the first time i m setting it up,,, have never done it before [07:23:36] chughakshay16: ok, try to reduce complexity by first setting up ssh-agent and your key and login anywhere, then worry about git again [07:24:44] mutante, so let me put it this way how do you setup ssh-agent ?? [07:25:23] mutante, whats this thing ssh-agent ? have never heard of it before.. [07:25:33] mutante, svn was so much simple [07:25:38] mutante, [07:25:43] mutante, :( [07:29:14] chughakshay16: depends on your distribution, if you have Debian/Ubuntu with Gnome, it is already running, and basically you just use ssh-add to load your key. it is there to keep your key loaded in the background so you don't have to repeat typing your key passphrase all the time [07:30:18] hello :-) [07:30:26] mutante, okay so when i run ssh @gerrit.wikimedia.org -p 29418... why is it showing me a welcome message if the ssh-agent is not up and running ?? [07:31:42] How would i go about creating a link to the current page? [07:32:29] say i have a template, in that template i want to have a edit link that points to the edit page on the template, no matter where i include the template. [07:34:07] chughakshay16: because you don't need ssh-agent to connect. it is just less convenient. try ssh -i ... ..sorry i need to go for a while now [07:34:35] mutante, okay [07:35:40] 03(mod) IPs hidden without any log or apparent reason - 10https://bugzilla.wikimedia.org/35578 (10orlodrim) [07:38:38] Why do people even bother using a separate key for gerrit? [07:43:18] 03(mod) IPs hidden without any log or apparent reason - 10https://bugzilla.wikimedia.org/35578 normal->04CRIT; +comment (10Dan Collins) [08:00:38] New patchset: Amire80; "Removing Bhojpuri from beta and adding it for language code bh." [mediawiki/extensions/Narayam] (master) - https://gerrit.wikimedia.org/r/3900 [08:07:23] hashar: I wanted to change access rights of Project mediawiki/extensions/Narayam in gerrit, can you help? [08:07:49] santhosh: sure. What is wrong with the default ones ? [08:08:07] I wanted to add owner as l10n-team along with Administrators [08:09:48] santhosh: done! [08:10:01] Narayam is now configured just like Translate [08:10:06] welcome as a owner! [08:10:10] hashar: thanks! [08:10:29] New patchset: Amire80; "RTL fix for special characters insertion links." [mediawiki/extensions/WikiEditor] (master) - https://gerrit.wikimedia.org/r/3901 [08:11:14] New review: Santhosh; "(no comment)" [mediawiki/extensions/Narayam] (master); V: 0 C: 2; - https://gerrit.wikimedia.org/r/3900 [08:17:29] New patchset: Amire80; "Moved character 25cc (dotted circle) before the diacritic." [mediawiki/extensions/WikiEditor] (master) - https://gerrit.wikimedia.org/r/3902 [08:18:45] hashar: hi. [08:19:04] https://gerrit.wikimedia.org/r/#change,3902 is a demonstration of the problem with the git review workflow. [08:19:32] by itself it's a stable commit so there's no problem with it. [08:20:16] however, i want to make another change in the same part of the code. if i submit it as a separate change, it will be harder to merge. [08:21:13] it would make more sense to submit it as a follow up commit in the same branch, but Roan says that in Gerrit it's an "anti-pattern". And i don't understand this. [08:22:11] now instead of committing the follow up change, i have to wait until this one is reviewed and merged. [08:30:49] hashar: ^ [08:31:02] :) [08:31:42] do you want to modify this patch or just submit a follow up ? [08:32:27] if you want to modify that patch: git review -d 3902; ; git add [08:32:42] then edit message: git commit --amend , notice how the Change-Id field is already there (do not modify it) [08:32:48] git-review -f (push change) [08:32:54] that will alter 3902 and submit a new patchset [08:33:16] if you want to publish a follow up, you can create a local branch based on that commit [08:33:42] Fetch the change objects: git fetch gerrit refs/changes/02/3902/1 [08:34:12] find out the sha1 for that patch set in gerrit interface (that is e6640f53ba … ) [08:34:29] create a local branch out of it : git checkout -b my followup e6640f53ba [08:34:57] (though you can probably just do: git checkout -b myfollowup refs/changes/02/3902/1 [08:35:01] do some hacking [08:35:02] git add [08:35:04] git commit [08:35:18] that will create a whole new commit which will have as parent e6640f53ba [08:35:52] when submitting it, a new change id will be generated and the commit will be detected by gerrit as a follow up of change 3902 [08:36:01] aharoni: does it answers your question? :-D [08:36:36] hashar: theoretically, yes. [08:36:52] i want to create a separate follow up commit. [08:37:11] so just create a new change based on tip of gerrit/master [08:37:18] git checkout -b myfollowup gerrit/master [08:37:28] i think that i already to do it in a separate branch and git review complained about multiple changes (with the "yes" and all that). [08:37:30] the commit will be based on whatever latest change is currently merged in gerrit [08:37:32] but i'll try it your way now. [08:37:52] oh there is also the multiple changes yes :-/ [08:38:06] the thing is something like: A -> B -> C [08:38:13] giving A is merged, B is pending [08:38:34] if you create C based on B and use git-review, I think it will try to push both B and C [08:38:43] although B has probably not been changed [08:38:56] I think that is because git-review only know that A is merged [08:39:03] and does not know the status of B, so it tries to push it [08:39:15] yes - that's what happened. Roan says that it's an "anti-pattern", and i'm still not convinced that there's anything wrong with it. [08:39:19] the menu (complaining about multiple changes) will show you the sha1 of B and C [08:39:39] if the sha1 of B is the same as the associated gerrit change, that will not trigger a new patchset [08:40:20] 03ning * 10/trunk/extensions/WikiObjectModel/includes/ (4 files in 2 dirs): bug fix, table cell contains html tag, fullurl magic word as link [08:40:37] but i guess that i really should mention in the commit message that it depends on a different patch set. [08:41:04] and that sounds just like what we did with the SVN review system, so where's the progress? [08:41:34] doing that as two (or more) changes in a simple git branch would make a lot more sense. [08:43:59] aharoni: I am not sure why you want a separate patch anyway [08:44:10] can't you just improve the existing one ? [08:44:26] these are two separate changes. [08:44:35] small commits are supposed to be good. [08:44:44] easier to review, if nothing else. [08:45:01] easier for finding bugs later. [08:45:27] the two changes are related, but they do different things. [08:46:20] big commits that introduce several different features are supposed to be an anti-pattern. [08:46:32] oh, and it's supposed to be easier to merge, too. [08:47:53] so consider this particular case: the patch set that i already submitted improves the display of the currently defined buttons in the special characters toolbar. [08:48:12] the one that i want to submit adds new characters. [08:48:48] the code lines are the same, and though the features related, they are functionally different. [08:48:50] so that is indeed two different features [08:49:19] is there any problem if you have one merged before the other ? [08:49:28] quite possibly. [08:50:02] and for me it's easier to test the second one locally after the first one was submitted. [08:50:18] so you want to enforce a dependency in Gerrit by using the method I gave above (checkout dependency, create a commit on top of it, submit) [08:50:26] ohhh [08:50:33] so you can just keep the patch in your local branch [08:50:35] ;) [08:50:42] i can create my own local branch for testing, merge the two changes there and then test. [08:50:46] there is probably no need to submit it to gerrit yet [08:50:51] since you will certainly improve it again [08:51:05] yeah or merge both locally [08:51:09] that would works to [08:51:30] plus if you might find an issue in the first patch while testing the second locally [08:51:34] yes, but after i merge both locally, it's like submitting the first one for review again. [08:51:53] and i only want to submit it once. [08:52:42] if this would be Github, i would create a branch, make two commits, and the issue a push request. [08:53:25] the reviewer would examine each commit, and then merge everything (if not problems would be found) [08:53:55] Gerrit makes this very complicated. [08:54:11] we have been talking about feature branches [08:54:19] have them reviewed and then merged [08:54:53] but you can surely use a local topic branch, push all your commits, that triggers as much changes all under the same topic branch in gerrit [08:55:10] then we can merge then one by one back in master but still will have to follow their dependencies [08:55:11] 03(mod) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 +comment (10reza1615) [08:56:09] 03netbrain * 10/trunk/extensions/SideBarMenu/ (SideBarMenu.hooks.php SideBarMenu.i18n.php SideBarMenu.php): added edit parameter. Wheter to display a edit link for sidebarmenu. [08:56:10] 03netbrain * 10/trunk/extensions/SideBarMenu/ (SideBarMenu.hooks.php includes/MenuParser.php): somehow dotted notation has returned! re-replaced dotts for dashes in i18n. [08:58:05] New patchset: IAlex; "Use local context to get messages" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3905 [08:58:47] 03(mod) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 +comment (10Amir E. Aharoni) [08:59:55] hashar: this sounds like a huge complication of something that should be simple :( [08:59:58] New review: Hashar; "(no comment)" [mediawiki/core] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/3905 [09:00:24] aharoni: well you did a feature and have it locally in some branch [09:00:31] aharoni: you could just checkout that branch [09:00:38] aharoni: create a second commit and git-review :- [09:00:51] not sure where is the difficulty :) [09:01:44] hashar: getDescription() is passed to OutputPage::setSubtitle() which passes it to OutputPage::addSubtitle which accepts Message objects [09:02:42] hashar: the difficulty is for the reviewer, that sees the same change twice (if i understand correctly) [09:03:54] ialex: well I did not investigate the code that much :-D [09:04:10] ialex: just remembered that some i18n people told me to always use ->parse() and ->text() [09:04:28] aharoni: the first commit will be a change. the second an other one [09:04:40] aharoni: you could try on tests/mediawiki/core maybe [09:04:45] hashar: correct, but some methods in OutputPage accepts those objects and will convert them to string using the appropriate method [09:06:13] okkkk [09:07:13] ialex: so now we have to amend includes/Actions.php [09:07:21] cause getDescription is documented to return a string [09:07:23] hashar: will do [09:07:26] return wfMsgHtml( strtolower( $this->getName() ) ); [09:07:48] so I guess we should never return a message [09:08:02] but always an wfMsgHTML equivalent [09:08:58] and I would have loved to have extensions to just have to use methods to set that kind of messages :-( [09:09:06] but that is a different ranting [09:09:12] 03netbrain * 10/trunk/extensions/SideBarMenu/ (SideBarMenu.hooks.php SideBarMenu.php): [09:09:13] renamed parameter edit to editlink [09:09:13] fixed small issue where parameter editlink was always interpreted as true [09:10:40] hashar, to sum things up: what would you suggest to try? should i create a branch for the second commit or should i do it in the same branch? [09:10:49] New patchset: IAlex; "Use local context to get messages" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3905 [09:10:55] gerrit-wm: there you are [09:11:07] arch, that was for hashar of course :) [09:14:26] RoanKattouw: hi! [09:15:01] MaxSem: hi [09:15:12] hi [09:15:15] aharoni: I would base the second commit on top of the first one [09:15:23] aharoni: this way it enforces dependency in Gerrit [09:15:56] hashar: by "base", do you mean "another commit in the same branch"? [09:15:59] MaxSem: I have submitted my GSoC Proposal, it would be awesome if you could read it once and give me valuable suggestions :) [09:16:17] MaxSem: http://www.mediawiki.org/wiki/User:Bagariavivek/GSoC_Application [09:17:04] "Typical working hours: 9 am to 3am" - you're supposed to work on this 40 hours per week [09:17:21] oh, am to am:) [09:17:38] MaxSem: :D [09:18:05] then my concern is that you should sleep more:P [09:18:17] MaxSem: i meant i can work any time between 9am to 3am [09:18:30] MaxSem: not that i will be working for so long :P [09:19:19] aharoni: I mean the first commit [09:19:24] aharoni: that will make it a parent [09:19:53] aharoni: try out using tests/mediawiki/core . I will assist you [09:20:04] ialex: looking. [09:21:39] New review: Hashar; "Looks fine to me. Would prefer Aaron / i18n team to also have a look at it." [mediawiki/core] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/3905 [09:22:20] hashar: i now realized that this particular commit affects separate lines, even though they are very close. so i'll try to do it in a separate branch. (simply because i started it already and i'm not in the mood for more Git experiments now :) ) [09:22:33] hehe [09:22:50] but i'm sure that i will have a similar situation some time soon. [09:23:03] or we can have first commit to be reviewed and merged. That will fix your issue ;-D [09:23:33] whenever you feel in a good mood to experiment with git, please contact me [09:23:42] New patchset: Amire80; "Adding pretty quotation marks for Hebrew." [mediawiki/extensions/WikiEditor] (master) - https://gerrit.wikimedia.org/r/3907 [09:23:45] and we will have some fun with tests/mediawiki/core [09:23:52] and find out a workflow that works [09:30:40] Is it possible to restrict viewing of all pages for public (unauthenticated) based on one IP? [09:31:01] !private [09:31:02] I don't know anything about "private". You might try: !botnoob !ppp !secrets [09:31:10] !secrets [09:31:10] MediaWiki was not designed with read-restrictions in mind, and may not provide air-tight protection against unauthorized access. We will not be held responsible should private information, such as a password or bank details, be leaked, leading to loss of funds or one's job. See also !lockdown [09:31:15] ahh [09:31:24] !lockdown [09:31:24] Lockdown is an extension for preventing read or write access by namespace and limiting access to special pages, see < http://mediawiki.org/wiki/Extension:Lockdown >. For per-page access protection, see !ppp. For general information on preventing access to your wiki, see !access. [09:31:43] you could probably do it with autopromote [09:31:54] !access [09:31:54] For information on customizing user access, see . For common examples of restricting access using both rights and extensions, see . [09:31:58] but i don't see the point, just give the user accounts and make them login [09:32:49] The wiki should be readable only from a school's IP, but editable from anywhere by a handful of editors [09:33:42] Dantman : hi, [09:33:47] Long-term sollution is to integrate it with OpenLDAP, but for now, I do not have the OpenLDAP system [09:33:50] Dantman : quick question [09:34:44] Dantman : in bugzilla if i see two same bugs asking for the same thing [09:34:59] Dantman : then where should i post my patch ? [09:36:09] MaxSem: I have to leave, please write your comments on the disscusion of the page, Thanks. [09:36:47] aharoni : if in bugzilla i see one bug marked as a duplicate of some other bug.. then in which one should i post the patch ? [09:37:18] chughakshay16: if 456 is a duplicate of 123, post the patch to 123. [09:38:08] aharoni : and what if its not marked as duplicate but still the bug description is same in two different bugs..? [09:38:59] aharoni : for ex see bug # 29073 and #7184 [09:39:46] aharoni : these both bugs are asking for the same thing, but none of them is marked as a duplicate of other [09:40:49] chughakshay16: Looks like you're right. Mark 29073 as a duplicate of 7184. [09:41:01] usually the newer is marked as a duplicate of the older. [09:41:19] and the patch must be submitted for 7184. [09:41:57] aharoni : okay aharoni , thanks [09:42:17] hi all! when I click "set up the wiki" link, server generates an error "HTTP 500 (Internal Server Error)". wtf? [09:43:19] 03(mod) Optionally exclude subpages from Special:Allpages and Special:Prefixindex - 10https://bugzilla.wikimedia.org/7184 +comment (10akshay chugh) [09:43:19] 14(DUP) Special:NewPages option to hide subpages - 10https://bugzilla.wikimedia.org/29073 +comment (10akshay chugh) [09:44:29] 03(NEW) UploadWizard localization bug for farsi language - 10https://bugzilla.wikimedia.org/35579 normal; MediaWiki extensions: UploadWizard; (reza.energy) [09:44:42] 03(mod) UploadWizard localization bug for farsi language - 10https://bugzilla.wikimedia.org/35579 +i18n (10reza1615) [09:45:00] 03(mod) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 +comment (10reza1615) [09:48:59] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 summary; +comment (10Amir E. Aharoni) [09:49:40] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 (10Amir E. Aharoni) [09:52:00] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10Amir E. Aharoni) [09:52:59] 03(mod) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 +comment (10Amir E. Aharoni) [10:00:05] 03(NEW) ProfilerSimpleUDP does not check for function_exists( 'socket_create' ) - 10https://bugzilla.wikimedia.org/35580 normal; MediaWiki: General/Unknown; (questpc) [10:05:50] 03(mod) Optionally exclude subpages from Special:Allpages and Special:Prefixindex - 10https://bugzilla.wikimedia.org/7184 +comment (10akshay chugh) [10:07:28] hi.long time.have u remember the bug no. which we solved. [10:21:38] 03(mod) MTC 635 not correctly rendering show/hide on smartphone layout - 10https://bugzilla.wikimedia.org/30063 +comment (10Jon) [10:22:29] New patchset: Santhosh; "Add support for Newar(new) language." [mediawiki/extensions/Narayam] (master) - https://gerrit.wikimedia.org/r/3908 [10:23:31] 03(WONTFIX) "Find" function for Wikipedia Mobile - 10https://bugzilla.wikimedia.org/20146 (10Jon) [10:24:12] New patchset: Jdlrobson; "re-enabled toggling on the footer" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3909 [10:24:13] New patchset: Jdlrobson; "reveal footer/section when open" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3910 [10:29:46] 03(mod) Devanagari transliteration tool - 10https://bugzilla.wikimedia.org/25326 (10Santhosh Thottingal) [10:32:24] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +need-review +patch; +comment (10Amir E. Aharoni) [10:34:11] New patchset: Jdlrobson; "enable login form on beta" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3911 [10:34:27] 03(FIXED) User auth on mobile site - 10https://bugzilla.wikimedia.org/31160 +comment (10Jon) [10:34:47] 03(mod) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 +comment (10Roan Kattouw) [10:39:47] New review: Catrope; "This change has major performance implications (slow SQL query), see https://bugzilla.wikimedia.org/..." [mediawiki/core] (master); V: 0 C: -2; - https://gerrit.wikimedia.org/r/3825 [10:43:57] 03(mod) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 +comment (10Immae) [10:49:22] 03(mod) Vector EditWarning breaks page caching. Edits get erased when navigating back and forth - 10https://bugzilla.wikimedia.org/22680 +comment (10Roan Kattouw) [10:49:34] 03(WONTFIX) MobileFrontend extension outputting invalid HTML in footer - 10https://bugzilla.wikimedia.org/30406 +comment (10Jon) [10:49:49] 03(mod) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 +comment (10Immae) [10:52:29] New patchset: Daniel Kinzler; "Introducing support for tags in wikitext as described in bug 33886." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3912 [10:54:35] 03(mod) to support for microdata and rdfa, allow tags so external links can have ref/rel attributes - 10https://bugzilla.wikimedia.org/33886 +comment (10Daniel Kinzler) [10:57:26] * Daniel_WMDE hopes this finally goes in this time around [10:58:16] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10reza1615) [11:02:20] 03(mod) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 +comment (10Roan Kattouw) [11:03:26] 03(mod) $wgDebugToolbar obscures footer links - 10https://bugzilla.wikimedia.org/32878 +comment (10Jon) [11:03:51] New review: Krinkle; "(no comment)" [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3912 [11:05:44] 14(INVALID) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 +comment (10Immae) [11:06:26] 03(mod) Let LocalisationUpdate fetch from Git - 10https://bugzilla.wikimedia.org/35469 +comment (10Roan Kattouw) [11:06:38] Change abandoned: Immae; "Performance problem (see bugzilla report)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3825 [11:11:39] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10Amir E. Aharoni) [11:12:14] 03(mod) Optionally exclude subpages from Special:Allpages and Special:Prefixindex - 10https://bugzilla.wikimedia.org/7184 +comment (10Krinkle) [11:14:21] 03(mod) Creation of a new mailing list for Kazakh Wikipedia (kk) - 10https://bugzilla.wikimedia.org/35518 summary (10Bolatbek) [11:16:22] 04(REOPENED) MobileFrontend extension doesn't load MediaWiki:Common.css - 10https://bugzilla.wikimedia.org/34325 +comment (10Krinkle) [11:16:56] 03netbrain * 10/trunk/extensions/Maps/includes/ (3 files in 3 dirs): added polygons parameter, can draw polygons on map [11:17:00] 03netbrain * 10/trunk/extensions/Maps/ (6 files in 4 dirs): added polygons parameter, can draw polygons on map [11:18:15] 03(mod) find another way to store titles of homepage sections - 10https://bugzilla.wikimedia.org/32583 +comment (10Krinkle) [11:21:42] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 -need-review -patch +upstream; +comment (10Krinkle) [11:22:34] New code comment: Netbrain; Something happened with git-svn merge so for some reason it was split into two commits; [11:26:14] 03(mod) Prevent creation of new (unattached?) SUL accounts with already unified names - 10https://bugzilla.wikimedia.org/29234 +comment (10Axpde) [11:29:16] 03(mod) Replace $wgPasswordSenderName with a system message defaulting to "{{SITENAME}} mail" to clarify and allow translation - 10https://bugzilla.wikimedia.org/32770 +comment (10Krinkle) [11:38:32] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10reza1615) [11:38:54] 03(WONTFIX) WebFonts is not working on IE6 with Windows 2003 Server - 10https://bugzilla.wikimedia.org/33247 +comment (10Santhosh Thottingal) [11:42:29] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10reza1615) [11:48:24] 03(NEW) Closure of nz.wikimedia.org - 10https://bugzilla.wikimedia.org/35581 normal; Wikimedia: Site requests; (mfwarburg) [11:52:43] 03(mod) Closure of nz.wikimedia.org - 10https://bugzilla.wikimedia.org/35581 +shell (10Robin Pepermans (SPQRobin)) [11:57:43] 03(mod) Redirect to canonicalize URLs (aka "squid cache of [[foo_(bar%29)]] not purged") - 10https://bugzilla.wikimedia.org/27935 summary (10Krinkle) [11:57:55] 03(mod) Redirect to canonicalize URLs (aka "squid cache of [[foo_(bar%29]] not purged") - 10https://bugzilla.wikimedia.org/27935 summary (10Krinkle) [12:03:31] ugh [12:03:41] how/where do i reply to comments in gerrit? [12:03:51] 03(mod) Language Selector dropdown on Special:WebFonts page doesnt load using WebFont - 10https://bugzilla.wikimedia.org/33639 +comment (10Santhosh Thottingal) [12:04:10] Daniel_WMDE: Review and leave a score of 0 [12:04:10] Daniel_WMDE: you hit Review [12:04:10] i can't even *see* the comment. i just see that there is one... [12:04:16] o_O [12:04:18] Oh, it says there are inline comments? [12:04:20] Daniel_WMDE: click Expand All? [12:04:41] there's no expand anything... [12:04:54] https://gerrit.wikimedia.org/r/#change,3912 [12:05:04] krinkle added a comment to my changes to parserTests.txt [12:05:20] i can't figure out how to show it on the gerrit page [12:05:50] click Diff All ? [12:05:58] Click "Unified" next to the file in question [12:06:05] https://gerrit.wikimedia.org/r/#patch,unified,3912,1,tests/parser/parserTests.txt [12:06:12] Then scroll down to line 9229 [12:06:34] RoanKattouw: ah, grat, thanks! [12:06:37] *great [12:07:28] Daniel_WMDE: Click on the comment and it'll expand and show a Reply button [12:07:53] After you've written your reply, it'll be saved as a draft, and you have to submit a 0-score review to submit it for real [12:07:56] (Yes, this is weird) [12:08:02] o_O [12:08:09] that's *very* strange [12:08:22] Welcome to the wonderful UI of Gerrit [12:08:49] New review: Daniel Kinzler; "(no comment)" [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3912 [12:09:18] RoanKattouw: well, it kind of makes sense if you want to group several comments into one review [12:09:22] it's sucks for replying, though [12:09:24] Yes [12:09:47] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10Amir E. Aharoni) [12:09:49] also, i want a big red "unsaved drafts" somewhere prominent [12:10:01] (and the review button is tiny and well hidden) [12:10:06] 03(mod) jquery.tablesorter should preserve original order of equal cells when reversing sort order twice - 10https://bugzilla.wikimedia.org/35526 summary; +comment (10Krinkle) [12:11:00] I believe the "N drafts" thing is in red [12:14:32] New review: Krinkle; "(no comment)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3842 [12:21:57] 03netbrain * 10/trunk/extensions/Maps/includes/ (6 files in 4 dirs): added parameter metadata separator [12:28:26] so, HILARIOUS recruiter spam just now [12:28:32] subject: Web Application Engineer [12:28:37]

Hi Sumana,

[12:28:41]

Are you interested in a new job opportunity? We checked out some of your git repos and we found a job opportunity that fits your skills. Twitter in San Francisco is hiring web application engineers.

[12:28:53] (yes, I'm reading it in plaintext, and yes, it included those

tags) [12:28:59] New review: Daniel Kinzler; "(no comment)" [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3912 [12:29:31] My GitHub account has 1, count 'em, 1 pull request ever - https://github.com/brainwane [12:30:08] https://github.com/GICodeWarrior/patch-tester/pull/1/files adding stuff to a README [12:30:56] there are also the 5 test commits I've made to example extensions, during the Gerrit tutorials https://gerrit.wikimedia.org/r/#q,owner:sumanah%2540wikimedia.org,n,z [12:31:24] If this qualifies me to be a web application engineer for Twitter, Twitter is doomed [12:31:38] New patchset: Hashar; "Refactor Profiler to get the actual and initial time:" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3824 [12:32:08] 03(mod) the Gerrit message "I would prefer that you didn't submit this" is very discouraging - 10https://bugzilla.wikimedia.org/35406 (10Chad H.) [12:32:15] "repos" plural? lol [12:32:48] Yeah, the only other git repos are on my laptop [12:32:58] (that could conceivably be considered mine) [12:33:15] I have a Gitorious account I've never done anything with, so I suppose that's a null repo [12:33:40] New review: Hashar; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3824 [12:33:42] Change merged: Hashar; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3824 [12:34:29] RoanKattouw: I'm leaving you for my glamorous new job as a web application engineer at Twitter! Where I presume I will just copyedit README files. But with Git! [12:34:56] I'm sure copyediting README files qualifies as "web application engineering" [12:34:56] 03(mod) Wikis ready for closing (tracking) - 10https://bugzilla.wikimedia.org/28985 (10Sam Reed (reedy)) [12:34:56] 03(mod) Closure of nz.wikimedia.org - 10https://bugzilla.wikimedia.org/35581 (10Sam Reed (reedy)) [12:35:27] I got a better recruiting e-mail the other day [12:35:41] In name of a "well-funded startup", SF-based, PHP web development [12:35:45] if you squint. if you squint so hard you are actually closing your eyes and scrunching up your face [12:36:09] sumanah: are there any folks that should refrain from commenting on the student wiki applications? staff or folks that will be involved with reviewing them or anything? [12:36:28] That sounds like the opposite of my imagined workflow [12:36:29] varnent: everyone should feel free to comment on the student applications [12:36:33] Staff and reviewers SHOULD comment [12:36:40] New review: Krinkle; "(no comment)" [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3912 [12:36:49] varnent: the point of this period is for those students to get as much feedback as possible! [12:36:53] open open open [12:36:56] sumanah: kk - just wanted to double check that before I posted it in this reply to the list - lol [12:37:06] varnent: thanks [12:37:11] "come one - come all - oh just kidding..not you folks" :) [12:37:17] 03(mod) the Gerrit message "I would prefer that you didn't submit this" is very discouraging - 10https://bugzilla.wikimedia.org/35406 +comment (10Chad H.) [12:37:41] New review: MaxSem; "(no comment)" [mediawiki/core] (master) C: 1; - https://gerrit.wikimedia.org/r/3896 [12:37:46] RoanKattouw: I failed in my mission yesterday to clarify Gerrit project ownership. however good news: Erik clarified "developer access" which makes my job easier [12:38:13] What happened yesterday? [12:38:42] RoanKattouw: https://www.mediawiki.org/wiki/Developer_access [12:39:12] Yesterday I meant to clarify the process around Gerrit project ownership and simply didn't make/have time for it [12:40:08] New patchset: Hashar; "(bug 27757) API method to retrieve tokens" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3890 [12:40:17] 14(WFM) ProfilerSimpleUDP does not check for function_exists( 'socket_create' ) - 10https://bugzilla.wikimedia.org/35580 +comment (10Sam Reed (reedy)) [12:40:27] RoanKattouw: "well-funded startup" huh? we're like the opposite :-) [12:40:36] Change abandoned: Hashar; "I have merged this change with its parent : https://gerrit.wikimedia.org/r/3890" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3896 [12:40:51] sumanah: should we add all these proposals to one category for easier tracking/locating? [12:41:01] varnent: sounds good, go ahead [12:41:01] 04(REOPENED) [Regression] Special:Log/patrol doesn't show "(automatic)" for autopatrols anymore - 10https://bugzilla.wikimedia.org/35565 +comment (10Trijnstel) [12:41:06] will do [12:41:07] New review: Hashar; "Patchset 2 is a merge (squash) of https://gerrit.wikimedia.org/r/3896" [mediawiki/core] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3890 [12:41:14] sumanah: I know :) it was the kind of recruiter e-mail I would've responded to if I had been in any way unhappy with my current job [12:41:51] * sumanah is glad you're not [12:42:22] so, ^demon - https://bugzilla.wikimedia.org/showdependencytree.cgi?id=22596&hide_resolved=1 and http://etherpad.wikimedia.org/BugTriage-2012-03 [12:42:35] is now a good time to talk about them? [12:43:36] <^demon> Yeah now's fine. [12:43:52] so - priorities right? [12:45:08] Uh… The Labs ToS really should be fixed… It's still marked as a draft which implies it can change, but as a legal binder it can't unless it has a statement clearly defining that before the person agrees... which it doesn't [12:45:10] sumanah: https://www.mediawiki.org/wiki/Category:Summer_of_Code_2012_applications [12:45:14] <^demon> sumanah: Looks about right so far. [12:46:30] <^demon> I think the upstream suggestion to switch to H2 or PG for Gerrit is kind of silly when it can be done perfectly fine within MySql. [12:46:48] New review: MaxSem; "(no comment)" [mediawiki/core] (master) C: 1; - https://gerrit.wikimedia.org/r/3905 [12:46:52] <^demon> It's only because gerrit was brain-dead and did latin1 collation. [12:47:03] You can just change the collation on the tables, right? [12:47:07] <^demon> Or didn't cheat and make everything binary ;-) [12:47:20] You don't have to worry about screwing up any existing UTF8 data because that's already broken anyway [12:47:28] ^demon: hashar and Tim-away had a question re how to deal with the language bot thing. Evidently the auto-approve bot for l10n isn't working -- Nikerabbit can you speak more on this? [12:47:33] <^demon> Should be able to. I was gonna test that on gerrit-dev. [12:47:43] <^demon> I know it's not working yet. [12:47:47] so the question is, what's better as a fix for right now? making the bot work, or a straight push to master, skipping Gerrit? [12:47:52] I'm for the former [12:47:58] thanks varnent [12:48:20] <^demon> No no, the auto-approve thing needs fixing properly. [12:48:29] <^demon> Skipping gerrit opens a can of worms I'd not like to do. [12:48:48] how do I merge head commit to 1.19? [12:49:25] ^demon: ok, I'm with you [12:49:53] Also ^demon -- Gerrit has 2.3 out -- release candidate. Final version will come out in April. [12:49:53] Antoine thinks we should look at it, cherry-pick a few changes next week (when Ryan comes back from vacation), then upgrade when 2.3 really comes out [12:49:53] Antoine working on this [12:50:12] <^demon> I've already got 2.3 running on the dev install. [12:50:22] <^demon> Did that early yesterday morning. [12:50:48] hexmode: btw you may want to see this conversation right now to get updates for your triage report [12:50:57] <^demon> Cherry picking changes in creates an maintenance nightmare. [12:51:04] ^demon: any particular reason not to upgrade to the release candidate now? [12:51:12] ^demon: 2.3-rc0 or the stable branch ? [12:51:14] ^demon: hi [12:51:24] <^demon> I'm already running 2.3-rc0 on gerrit-dev.wmflabs. [12:51:38] <^demon> If it's stable enough, we can talk to Ryan about upgrading. Could use some testing first. [12:52:09] Ryan won't be available till Monday. [12:52:29] <^demon> Well we can do stuff with the test install without him. [12:52:34] Any particular stuff that needs testing, ^demon ? [12:52:37] <^demon> And I wouldn't want to rush in an upgrade today/tomorrow anyway [12:52:43] We could ask Chris McMahon for some help on that. [12:52:53] <^demon> I want to make sure the new submodule registration thing works as advertised. [12:53:04] <^demon> Make sure our current workflow is still working (no regressions) [12:53:41] <^demon> The latter anyone can do actually. [12:54:28] and gerrit-dev is also running the nice skin that Krinkle & RoanKattouw and the OpenStack people worked on, right? [12:54:52] 03(NEW) Unable to revert edits - 10https://bugzilla.wikimedia.org/35582 critical; Huggle: Application; (1994constant) [12:55:05] <^demon> Yeah. That won't necessarily be a part of the upgrade...I don't believe it's finished yet RoanKattouw? [12:55:14] Ask Krinkle [12:55:32] I've submitted it into the puppet repo but -1ed it because Krinkle's CSS has a big warning on top saying "UNSTABLE" [12:55:36] ^demon: look at the full history of gerrit. There is a few changes made post rc0 that we want [12:55:48] ^demon: as an example refs/for/draft/ being renamed in refs/for/drafts/ [12:55:49] <^demon> Link? [12:56:02] RoanKattouw: indeed [12:56:06] should not go live [12:56:09] Also, Krinkle said that it's quite conceivable that the CSS will break completely with the upgrade [12:56:13] <^demon> hashar: I thought they were doing refs/drafts/Foobar [12:56:20] <^demon> They really need to pick a name and stick with it :) [12:56:38] <^demon> RoanKattouw: Well that's mostly not-true since we've upgraded and the CSS seems fine :) [12:56:57] sumanah: I think I got the other emails lingering from yesterday that you hadn't already replied to - will be able to reply to them quickly today - let me know if I missed any emails or apps for that category (I think I found them all) [12:57:01] ^demon: git log --graph --oneline v2.3-rc0..stable-2.3 --no-merges [12:57:11] <^demon> I suppose I should clone it :p [12:57:39] thank you varnent [12:57:43] 03(mod) Unable to revert edits - 10https://bugzilla.wikimedia.org/35582 +comment (10Ankit Maity) [12:58:59] <^demon> The heck? [12:59:03] <^demon> Never seen this one before [12:59:09] ^demon: and the totally kicking ass feature : $ git shortlog v2.3-rc0..stable-2.3 [12:59:13] :--) [12:59:32] now I need to find out the commit I talked about [12:59:34] <^demon> :( http://p.defau.lt/?fLeCRy_FR3XvnUMNM2KrBA [12:59:54] wtf [13:00:21] <^demon> git server borked on their end? [13:01:00] one last thing, we can get to it later, Chad -- diff stuff. diff between 2 patchsets, and diffs not showing up on merge commits [13:01:03] It's still hard to do a diff between 2 patchsets; that only happens if the 2nd patchset is rebased. We can use the script Tim posted http://tstarling.com/gerrit-differ.php . And for merges, one should download the branch locally -- the problem of diffs not showing up on merge commits is high priority, Tim suggested yesterday. He meant to check IRC logs to track down a possible bug # (bugzilla or code.google.com). [13:01:07] (notes from yesterday) [13:01:57] ^demon: looks like [13:02:13] ^demon: anyway 2.3-rc0 to 2.3 is just pending a few doc change and some trivial fixes [13:02:21] ^demon: we could ask them to release it :-D [13:02:59] Hi, how do I add a fb like button all the pages of my wiki? [13:03:17] 14(DUP) No way to mark merged commits as needing fixing - 10https://bugzilla.wikimedia.org/35535 +comment (10Chad H.) [13:03:19] 03(mod) Free-form tagging in gerrit - 10https://bugzilla.wikimedia.org/35534 +comment (10Chad H.) [13:04:00] !facebook [13:04:05] I don't know anything about "facebook". [13:04:05] stupid bots [13:04:13] SocialProfile extension? [13:04:43] https://www.mediawiki.org/wiki/Extension:SocialProfile oh no. [13:04:47] no Like buttons [13:04:54] we have a Like extension [13:05:00] They have everything except facebook [13:05:02] that comes with linked in / twitter / whatever [13:05:58] But, the code has to be copied to every single page for the button to appear [13:06:07] which is manual work ;( [13:06:19] https://www.mediawiki.org/wiki/Extension:Social_Sidebar ? [13:06:30] <^demon> hashar: If they're close to releasing 2.3, we can use the time to test rc0 and then upgrade when it goes final perhaps. [13:06:39] <^demon> Although "sometime in April" is a bit vague for my taste. [13:06:42] ^demon: hop just use 2.3 final [13:06:44] wizor: looking at https://www.mediawiki.org/wiki/Extension:FBConnect , https://www.mediawiki.org/wiki/Extension:Social_Sidebar , https://www.mediawiki.org/wiki/Extension:Facebook , https://www.mediawiki.org/wiki/Extension:Widgets which mentions some Facebook stuff [13:06:53] wizor: do any of those suit you? [13:07:04] ^demon: rc0 has at least one nasty bug :-D On the other hand, we could run from stable-2.3 (aka their release branch) [13:07:07] sumanah: let me check [13:07:10] Thanks BTW [13:07:25] ^demon: they also have an hackaton sometime at the end of may / beginning of june [13:07:34] <^demon> I saw from the list. [13:07:37] you're welcome :-) wizor I just searched mediawiki.org for "facebook" and then tried a search for "like button" [13:07:55] <^demon> hashar: Does stable-2.3 have the .war generated, or would we have to do that ourselves? [13:08:01] <^demon> That's a big ugh if we do :( [13:08:07] wizor: have a look at https://www.mediawiki.org/wiki/Extension:AddThis [13:08:54] there is also the widget extension [13:09:01] will let you get widget such as http://www.mediawikiwidgets.org/Facebook_Like_Box [13:09:19] varnent: maybe for SEO purposes you should add the phrase "like button" to https://www.mediawiki.org/wiki/Extension:AddThis :-) [13:09:46] sumanah: good call - I'm about to Pinterest that thing [13:10:05] ^demon: I could make one :-D [13:10:12] <^demon> I know we *can* [13:10:14] ^demon: something like: mvn release [13:10:15] <^demon> That's not the question [13:10:25] so "we have to do that ourselves" [13:10:44] <^demon> From an ops/maintenance standpoint, that's annoying as hell when we could just be patient and wait a bit ;-) [13:11:11] well given Gerrit.war was provided by puppet, it is all about pushing the war to operations/puppet.git *Grins* [13:11:25] anyway, I agree we should wait [13:11:29] we have more urgent things to do [13:12:36] * varnent is actually surprised how quickly wikis have started using the AddThis extension [13:13:10] 03(mod) Replace $wgPasswordSenderName with a system message defaulting to "{{SITENAME}} mail" to clarify and allow translation - 10https://bugzilla.wikimedia.org/32770 +comment (10Strainu) [13:13:25] 03daniel * 10/branches/Wikidata/phase3/includes/actions/RawAction.php: trigger http error when non-text content is requested using action=raw [13:13:59] New patchset: Hashar; "PostgreSQL schema:" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3699 [13:14:24] New review: Hashar; "I have updated commit message per https://www.mediawiki.org/wiki/Git/Commit_message_guidelines" [mediawiki/core] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3699 [13:14:35] 03(mod) usercontribs API: sort results by timestamp,user or user,timestamp - 10https://bugzilla.wikimedia.org/35349 (10Bryan Tong Minh) [13:14:40] New review: Krinkle; "Currently tokens for 'rollback' and 'patrol' are not available here." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [13:14:56] 03daniel * 10/branches/Wikidata/phase3/includes/EditPage.php: tabs [13:15:13] 03daniel * 10/branches/Wikidata/phase3/includes/ContentHandler.php: reminder [13:16:13] ^demon: so, re diff between 2 patchsets, and diffs not showing up on merge commits - any thoughts? [13:16:23] (my infodump starting "It's still hard to do a diff between 2 patchsets...") [13:16:24] New review: Demon; ""PostgreSQL schema" isn't a very descriptive 1-line summary." [mediawiki/core] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3699 [13:17:52] <^demon> sumanah: The former is possible, although awkward. The latter is a known bug and filed upstream. [13:18:06] and both have workaround in command line :-D [13:19:07] hashar: what's the command line workaround for the latter? [13:19:25] New review: Hashar; "(no comment)" [mediawiki/core] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/3699 [13:20:14] New review: Krinkle; "(no comment)" [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [13:21:03] sumanah: git show -m ? [13:21:31] hashar: What is -m ? [13:21:41] Also, remember that merge commits have two possible diffs [13:21:48] git show probably shows both [13:21:59] RoanKattouw: ask git diff-tree to show the diff coming from merge commits :-] [13:22:24] aha [13:22:29] Yeah without -m it shows a 3-way diff [13:23:07] What you really want is to diff it to the parent that's on master [13:24:00] hmm [13:24:15] git diff ^ ? :-D [13:24:39] Well what does ^ mean for a merge commit? [13:25:21] Seems to work for me though [13:25:26] And git show -m seems to work too [13:25:42] RoanKattouw: a merge commit has a parent [13:25:48] Two parents [13:25:57] well as I understand it a merge commit take a few changes and apply them to something [13:26:03] that something could be considered the parent [13:26:10] Yes, but it works the other way too [13:26:12] * hashar looks for an octopus merge [13:26:26] Oh wait, it says [13:26:31] commit eb1f9c064c2141eb5e80c19d98815defe3bf312a [13:26:33] Merge: b914ee7 a28f1aa [13:26:41] So it knows in which direction the merge occurred [13:26:50] lol [13:27:18] I am confused roan [13:27:38] Well, whether you merge A into B, or B into A, the resulting tree is the same, right? [13:27:39] you merge stuff on top of something [13:27:54] And the resulting merge commit has two parents [13:27:59] RoanKattouw: the end result is surely the same [13:28:04] but the topology is different to me [13:28:22] (the end result will have trees with differents sha1 anyway) [13:28:24] well should [13:28:45] New review: MrBlueSky; "Can I amend this change to update messages.inc or won't that work because its already merged?" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3886 [13:28:46] so yeah,technically you could merge master on your topic branch [13:28:58] I did merge master into a topic branch earlier [13:28:59] and get a second merge topic that merge your topic branch on master [13:29:03] https://gerrit.wikimedia.org/r/#change,3854 [13:29:08] and end up with two identical branches code wise [13:29:27] but do not quote me on that. Just expectations [13:29:35] sumanah, when do you intend to approve mentors on Melange? I suppose students have started submitting their proposals, so there's something to discuss [13:31:48] 03(FIXED) [Regression] Special:Log/patrol doesn't show "(automatic)" for autopatrols anymore - 10https://bugzilla.wikimedia.org/35565 +comment (10Max Semenik) [13:32:28] Oh, that reminds me. I still need a (proposed) mentor for https://www.mediawiki.org/wiki/User:Jarry1250/GSoC_2012_application [13:32:43] Just gonna keep spamming it til someone agree :P [13:32:47] * agrees [13:33:45] hi MaxSem [13:34:53] MaxSem: I have to check what Melange will let me do for some edge cases -- it's a little different from last year [13:35:18] Jarry1250: maybe you have some prospects you can reach out to personally? [13:36:12] Picking on people? Yeah, I'll start doing that soon enough :) [13:36:18] Jarry1250: Have you asked Brion? This seems like the kind of thing that's up his alley [13:36:43] He was #1 on my list of people to ask, yes :) [13:37:38] I mentioned it vaguely a few days ago, he said he was going to flick through a few proposals to get a feel for this year's offering. [13:37:50] no. [13:38:01] I suggest someone different :-) [13:38:19] sumanah: can you load the site? it seems offline in europe [13:38:32] Daniel_WMDE: en.wp? [13:38:36] Daniel_WMDE: Melange? [13:38:38] something else? [13:38:45] meta.wikimedia.org [13:38:48] de.wikipedia.org [13:38:52] Daniel_WMDE: enwiki WFM [13:38:55] mediawikiwiki is down too [13:38:56] Daniel_WMDE: one of the bits is being sucky [13:38:56] actually, it seems to be bits. so everything [13:39:08] sumanah: Well, I'll work out who else I can ask shortly. [13:39:21] Cool. [13:39:21] p858snake|l: one or all? [13:39:22] New patchset: Jdlrobson; "don't assume existence of #content" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3913 [13:39:36] ok, seems like people are slowly waking up :) [13:39:38] Who broke the interwebs?! o_O [13:39:45] Daniel_WMDE: I know one was earlier [13:40:01] Meta loads for me. [13:40:04] well, if it's just one, everyone is hitting it [13:40:10] I was just trying to edit. What have I done O_o [13:40:13] no bananas here. at all. [13:40:18] ok, sounds like a networking issue [13:40:30] go into #wikimedia-tech and start mentioning your IP addresses please [13:41:09] no that isn;'t needed [13:41:14] This is not a networking issue [13:41:18] oh never mind then [13:41:19] esams bits is fried [13:41:25] I was going on what Tim said recently [13:41:30] but I misunderstood [13:42:22] He said something about no identifiable cause [13:42:27] http://ganglia.wikimedia.org/latest/?c=Bits%20caches%20esams&m=load_one&r=hour&s=by%20name&hc=4&mc=2 [13:45:36] Jarry1250: https://www.mediawiki.org/wiki/Summer_of_Code_2012#Mentor_signup has a list [13:49:57] sumanah: you were right, sometimes we have to wait, got a reply from Jeroen now :) [13:51:08] nischayn22: yeah -- you may want to straightforwardly ask them what the expectations should be, so you know when to start worrying and when to relax :-) [13:52:40] sumanah: he and Markus are in a project right now and so they are busy, Jeroen replied now from the airport, so I think he is interested in my proposal :) [13:53:09] nischayn22: yes, that's good [13:53:13] my suggestion still stands [13:53:22] it's setting up foundations for better communication in the future [13:53:55] sumanah: I asked him already, but now he has time only to reply to a mail of mine [13:54:08] nischayn22: then maybe he will respond later. [13:54:48] sumanah: Is the project schedule very strict ? or can I modify it later after I have talks with Jeroen and Markus [13:55:10] nischayn22: until April 6th, it's totally changeable. [13:55:39] nischayn22: after that, it's partially up to me and up to the mentors, on a case-by-case basis [13:55:44] I suggest you SCOPE SMALL [13:55:46] sumanah: I can't change it after that even on MW.org ? [13:55:49] ok [13:55:51] nischayn22: look at https://www.mediawiki.org/wiki/Summer_of_Code_2012/management#GSoC_management_philosophy [13:56:09] sumanah: seen that already, very strict [13:56:23] my scope is small indeed [13:58:13] nischayn22: well, of course we will have to respond to changing conditions. This is why I think it's a good idea to start with a very conservative schedule. Then during the consideration period in early April, you and the mentors and I would talk more about it and possibly modify the schedule and deliverables. Any change in the student's project schedule after that is possible, but I am not going to be ok with drastic changes unless people run [13:58:14] into big problems or opportunities that they could not have anticipated [13:58:54] Daniel_WMDE: is the site better for you now? [13:59:26] sumanah: yes, thanks you [13:59:32] thank Roan :-) [13:59:41] Daniel_WMDE: how is the Wikidata rampup going? [13:59:47] I've seen you're as busy as a bee! [14:00:02] sumanah: next two weeks will be craszy [14:00:07] Just had a unique experience: a popup window announcing "500 Mediawiki configuration error" while using hotcat on a farm wiki. [14:00:18] a bunch of new people coming in, need intros... [14:00:23] Amgine: ! that seems a mite unusual [14:00:31] Daniel_WMDE: coming in to Wikidata? the workers? [14:00:52] yes [14:00:53] sumanah: I'll be asking you about material and suggestions for mediawiki 101 sessions tomorrow [14:00:56] 03(mod) Farsi translations are incorrect in the jQuery datepicker - 10https://bugzilla.wikimedia.org/35579 +comment (10Amir E. Aharoni) [14:01:21] Daniel_WMDE: https://www.mediawiki.org/wiki/Category:Tutorials has all I have, I think [14:01:22] we have 8 new people starting on monday [14:01:28] 4 or 5 of them have never worked with mediawiki [14:01:31] sumanah: got it, I will act accordingly, going for dinner now:) [14:01:42] Daniel_WMDE: it would be ROCKIN' if you could record your sessions via audio or even video [14:01:48] good luck nischayn22 [14:01:49] Sumanah: I think it's part of HotCat gadget. (not finding anything like it anywhere else) [14:02:08] hm. [14:02:22] sumanah: i'll suggest it to abraham. the problem is: we will be at the new temp office on monday. before the office actually moves there on wednesday and thursday. [14:02:36] so, minimal infrastructure, lots of overhead, noise, running around, etc [14:02:47] it'll be hard even top find a camera :/ [14:02:49] Daniel_WMDE: ah, so, recipe for extraordinarily crappy audio. [14:02:58] crappy everythign, really :) [14:03:07] it's going to be awesome anyway. has to be [14:03:08] Daniel_WMDE: you could point the folks at the https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker/Workshop now [14:04:07] thanks! [14:04:12] hm, that needs a note about git... [14:04:43] it sure does. [14:04:46] https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker/Extension_Writing_Tutorial is a little better [14:05:09] oh god, that needs updating too, for Git [14:06:14] I'll ask awjr to fix [14:08:29] 03(mod) the Gerrit message "I would prefer that you didn't submit this" is very discouraging - 10https://bugzilla.wikimedia.org/35406 +comment (10Chad H.) [14:08:59] ^demon: have you ever read "Bartleby the Scrivener"? [14:09:20] <^demon> Nope. [14:11:53] Chad: Old Melville story. It has a guy who keeps saying "I would prefer not to" -- all the recent talk about the "I'd prefer" wording in Gerrit reminds me of it. http://www.bartleby.com/129/index.html in case you want some old-school fiction [14:12:27] * ^demon downloads to tablet [14:12:32] <^demon> Lecture will probably be boring :) [14:12:40] <^demon> hashar: That you ^ [14:12:42] :-D [14:12:49] <^demon> mw-jenkinsbot [14:12:57] people are bored when reading me ? :D [14:13:05] yeah restarting jenkins [14:13:28] <^demon> Mmk, just making sure. [14:13:41] I need a way to have Jenkins to self update somehow [14:15:11] <^demon> Alrighty, I'll be back in ~2h [14:15:26] <^demon> sumanah: We'll look at fixing that "I'd prefer..." bug this afternoon when I'm back [14:15:28] 03(mod) Keep home-random menu expanded after clicking random - 10https://bugzilla.wikimedia.org/30842 +comment (10Jon) [14:16:28] New patchset: Amire80; "Add support for Newari (new) language." [mediawiki/extensions/Narayam] (master) - https://gerrit.wikimedia.org/r/3908 [14:18:24] New review: Amire80; "Tested on my local wiki. Amended the comments a little." [mediawiki/extensions/Narayam] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/3908 [14:18:52] New review: Amire80; "Tested on my local wiki. Amended the comments a little." [mediawiki/extensions/Narayam] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3908 [14:18:55] Change merged: Amire80; [mediawiki/extensions/Narayam] (master) - https://gerrit.wikimedia.org/r/3908 [14:25:58] MaxSem: hi [14:26:06] hi [14:26:11] MaxSem: did you read my application? [14:26:21] MaxSem: sorry if i am disturbing you [14:28:41] potter, while I think that your proposal is a step in a right direction, there are far more things can be done with watchlists [14:28:51] for example, shared watchists [14:29:16] if you're about to ign them, don't stop in the middleredes [14:29:33] OTOH, a GSoC proposal shouldn't be overambitious [14:29:40] * varnent likes the shared watch list idea - or maybe "topic" watchlists? [14:29:52] varnent: Someone wrote an extension for that [14:30:02] Except that the ext also does 7 other things and it's spaghetti [14:30:04] yay - okay - good :) [14:30:11] lol - that's less good [14:30:31] * varnent hopes in two years we can distribute shirts that say "there's an extension for that" [14:30:55] MaxSem: there is an extension for sharing watchlists [14:31:16] and it has problems^^^ [14:31:37] MaxSem: oh.. [14:32:14] MaxSem: what other ideas do you think can be implemented [14:32:47] MaxSem & RoanKattouw: this is a project for down the road - but someone requested we start the list - is that extension a candidate for recovery? https://www.mediawiki.org/wiki/Project:WikiProject_Extensions/Projects#Recovery [14:33:45] Maybe [14:34:54] MaxSem: varnent: Is the size of the project i proposed, apt for GSoC ? [14:35:15] potter: can you resend me the link - sorry [14:35:28] varnent: sure [14:35:30] varnent: http://www.mediawiki.org/wiki/User:Bagariavivek/GSoC_Application [14:36:21] 03(NEW) GitInfo::isSHA1 returns true for non-sha1 values - 10https://bugzilla.wikimedia.org/35583 trivial; MediaWiki: General/Unknown; (alexis.moinet) [14:38:32] New patchset: Reedy; "Bug 35583 - GitInfo::isSHA1 returns true for non-sha1 values" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3914 [14:39:01] 03(FIXED) GitInfo::isSHA1 returns true for non-sha1 values - 10https://bugzilla.wikimedia.org/35583 +patch +reviewed; +comment (10Sam Reed (reedy)) [14:41:31] potter: it looks like you're tackling one big issue - which seems appropriate to me [14:42:20] potter: I can't speak to how many devo hours will be required - but I could see one more WatchList addition being viable - but I'd defer to others on that [14:43:04] varnent: thanks. Any suggeston? Any improvements [14:43:20] New review: Kaldari; "@Krinkle: Patrol tokens are also a complicated case since getPatrolToken requires a RecentChange obj..." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [14:43:35] sumanah : hi, [14:43:55] potter: not off the top of my head - just to assess with others about the scope - as you're already doing :) [14:44:02] New patchset: Reedy; "(bug 33536) Make a list of print on demand publishers configurable for Extension:Collection" [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3440 [14:44:15] varnent: :D [14:44:30] potter: I think you've done a nice job of explaining the problem, your proposed solution and your plan of action [14:45:08] varnent: thanks again [14:45:25] potter: maybe check out that extension mentioned earlier - see if it can be absorbed in as well maybe - just a thought [14:45:43] varnent: http://www.mediawiki.org/wiki/Extension:CollaborativeWatchlist ? [14:46:37] potter: yeah - I haven't checked it out too much - but I get the impression it could use some help - might be able to integrate from it what you want [14:47:02] varnent: i will look into the codebase of this extension [14:47:27] potter: worth a gander - might not be worth absorbing..but might save some duplication of work later [14:48:09] varnent: I have submitted my proposal on melange, is it fine to mail the wikitech-l. [14:48:18] varnent: spam... [14:48:35] potter: absolutely - I'd link to the wikipage you created - I don't know if the melange apps are public or not [14:49:01] varnent: they can be made public, by the proposer himself [14:49:06] potter: I don't think it's spam - but feel free to blame me if someone says something :) [14:49:18] varnent: :D :D [14:50:45] 03(mod) Don't surface fix me templates on Mobile Frontend - 10https://bugzilla.wikimedia.org/30421 +comment (10Max Semenik) [14:53:01] New review: Krinkle; "@Kaldari: That function does need it but that's only a disguise because of the API structure (that f..." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [14:53:30] New review: Krinkle; "Basically getPatrolToken doesn't need an $rc object, just like all the others don't need $pageid." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [14:54:41] New review: Reedy; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3905 [14:54:44] Change merged: Reedy; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3905 [15:00:40] New patchset: Catrope; "Followup r113740" [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3442 [15:01:59] New review: Reedy; "(no comment)" [mediawiki/extensions/Collection] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3442 [15:03:19] New patchset: Catrope; "Documentation stuffs" [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3443 [15:04:20] New patchset: Catrope; "So apparently Collection still had a PHP4 constructor" [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3445 [15:04:20] New patchset: Catrope; "Reports of undefined index notices related to display title" [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3444 [15:04:30] New review: Reedy; "(no comment)" [mediawiki/extensions/Collection] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3443 [15:04:40] Reedy: That should fix all of them [15:05:08] Hello, im trying to upgrade from 1.15.3 to 1.18.2, but i get Database returned error "1146: Table 'civilbilar.user' doesn't exist (localhost)" but there is no prefix! databasename is civilbilar [15:05:13] New patchset: Jdlrobson; "first version of ajax loader" [mediawiki/extensions/MobileFrontend] (history) - https://gerrit.wikimedia.org/r/3894 [15:06:15] New review: Jdlrobson; "This is currently just for transparency purposes. Please do not merge into core." [mediawiki/extensions/MobileFrontend] (history) C: -1; - https://gerrit.wikimedia.org/r/3894 [15:09:49] New review: Kaldari; "Yeah, but getPatrolToken actually uses the RecentChange object. For the other tokens I can just pass..." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [15:12:46] New review: Krinkle; "Yeah, it should return the token $wgUser->useRCPatrol() OR $wgUser->useNPPatrol() is true. And logic..." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [15:19:24] any one? [15:19:45] Where do you get the error? [15:20:09] php update.php [15:20:18] or in the installer mw-config [15:21:21] you get it in both? [15:21:29] yeah [15:21:35] 03(mod) Autolink to new Gerrit / Git changesets and SHA-1 commits - 10https://bugzilla.wikimedia.org/35144 +comment (10Mark A. Hershberger) [15:22:10] 03(NEW) Continuous integration (tracking) - 10https://bugzilla.wikimedia.org/35584 normal; Wikimedia: Testing Infrastructure; (hashar) [15:22:27] 03(mod) Continuous integration (tracking) - 10https://bugzilla.wikimedia.org/35584 (10Antoine "hashar" Musso) [15:22:27] 03(mod) Tracking bug (tracking) - 10https://bugzilla.wikimedia.org/2007 (10Antoine "hashar" Musso) [15:22:43] Reedy "There are MediaWiki tables in this database. To upgrade them to MediaWiki 1.18.2, click Continue." then the error [15:23:23] I seem to recall there being something like that in the web installer (maybe), but via update.php seems weird [15:23:34] It works fine on my 1.18 dev wiki (just verified) [15:25:20] 03(mod) Implements continuous integration (tracking) - 10https://bugzilla.wikimedia.org/35584 summary; +comment (10Antoine "hashar" Musso) [15:26:19] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 summary (10Krinkle) [15:26:54] Reedy [15:26:55] Database returned error "1146: Table 'civilbilar.watchlist' doesn't exist (localhost)" [15:27:14] Krakke: It looks like you don't have all the tables then [15:27:15] You haven't overwritten your LocalSettings.php or anything? [15:27:27] no i got all files backed up [15:27:33] He mentioned user before... [15:27:37] And he went from 1.15 [15:28:17] its using like prefix, "civilbilar" that is the wikiname, user, database ... [15:28:51] 03(NEW) Jenkins job: universal lint check - 10https://bugzilla.wikimedia.org/35585 normal; Wikimedia: Testing Infrastructure; (hashar) [15:29:06] 03(mod) Jenkins job: universal lint check - 10https://bugzilla.wikimedia.org/35585 (10Antoine "hashar" Musso) [15:29:07] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Antoine "hashar" Musso) [15:30:26] 03(NEW) Remove all feedback forms but option 1 - 10https://bugzilla.wikimedia.org/35586 normal; MediaWiki extensions: ArticleFeedbackv5; (yoni) [15:31:18] 03(ASSIGNED) Remove all feedback forms but option 1 - 10https://bugzilla.wikimedia.org/35586 normal->15enhancement; +aftv5-1.5 (10Yoni Shostak) [15:31:39] http://dpaste.com/hold/723552/ [15:31:55] all text [15:32:07] awjr : hi arthur , [15:32:43] hi chugakshay16 how's it going [15:32:59] awjr, need some help . [15:33:06] 03(FIXED) Long execution time for extension init - 10https://bugzilla.wikimedia.org/29361 +comment (10Max Semenik) [15:33:23] New review: Reedy; "(no comment)" [mediawiki/extensions/Collection] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3445 [15:33:24] hi sumanah ... another student is interested in "Create a way to have books for Wikisource / wikibooks"... what procedure should I take [15:33:36] sumanah, ? [15:33:41] Hi Raylton [15:33:48] who is it? [15:33:55] 03(NEW) Make AFTv5 url parameters work all the time - 10https://bugzilla.wikimedia.org/35587 normal; MediaWiki extensions: ArticleFeedbackv5; (reha) [15:34:00] awjr, i am not able to install 'commit-msg' hook .. [15:34:01] New review: Reedy; "(no comment)" [mediawiki/extensions/Collection] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3444 [15:34:03] 03(NEW) Jenkins job: PHP code style analyzer - 10https://bugzilla.wikimedia.org/35588 normal; Wikimedia: Testing Infrastructure; (hashar) [15:34:09] sumanah, Ashwini Kumar [15:34:17] 03(mod) Jenkins job: PHP code style analyzer - 10https://bugzilla.wikimedia.org/35588 (10Antoine "hashar" Musso) [15:34:18] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Antoine "hashar" Musso) [15:34:21] awjr, could u please help me out. [15:34:33] New review: Reedy; "(no comment)" [mediawiki/extensions/Collection] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3440 [15:34:35] chughakshay16: what's the issue? maybe someone else here can help [15:34:36] Change merged: Reedy; [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3445 [15:34:37] Change merged: Catrope; [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3442 [15:34:39] Change merged: Reedy; [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3443 [15:34:40] chughakshay16: what error do you get? [15:34:40] Change merged: Reedy; [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3444 [15:34:41] Change merged: Reedy; [mediawiki/extensions/Collection] (master) - https://gerrit.wikimedia.org/r/3440 [15:34:43] lolol [15:34:47] hehe [15:34:50] chughakshay16 im not sure what that is [15:34:51] I love gerrit dependencies :-D [15:34:54] Raylton: well, has Ashwini done any coding work yet, or written a good proposal? [15:35:23] Raylton: Historically, sometimes we receive multiple proposals for very similar tasks for Google Summer of Code, but we only accept one (factors going into our decision will include which proposal is better, which student has participated more in the Wikimedia community, which student has better technical skills, and which student has better communication skills). [15:35:40] Reedy did you se the http://dpaste.com/hold/723552/ ? [15:35:45] 03(NEW) Retire AFTv5 "bucket" url parameter - 10https://bugzilla.wikimedia.org/35589 normal; MediaWiki extensions: ArticleFeedbackv5; (reha) [15:35:53] is mediawiki.org not loading for other folks? [15:36:09] Raylton: Therefore, Ashwini and Aashish should either choose to strive to make their applications the best they can be, or one of them should choose a different topic. [15:36:10] ...logging table already exists. [15:36:10] ...user table does not exist, skipping new field patch [15:36:10] ...watchlist table does not exist, skipping new field patch [15:36:11] awjr , https://www.mediawiki.org/wiki/Git/Workflow in this document in the last step of git-review.. that i am kind of stuck with [15:36:49] awjr, when i run this command git-review -s [15:37:01] New patchset: Reha Sterbin; "Make it so that the debug parameter, if given, overrides the config setting and matches the use of debug elsewhere; make acceptance of url paramenters not dependent on debug setting" [mediawiki/extensions/ArticleFeedbackv5] (master) - https://gerrit.wikimedia.org/r/3916 [15:37:15] chugakshay16 what is the error you get? [15:37:23] and are you using windows? [15:37:25] awjr, its not installing a 'commit-msg' hook under my hooks directory [15:37:32] awjr, yes windows 78 [15:37:38] * windows7 [15:37:41] sumanah: google is probably going to host a Gerrit hackaton :-] [15:37:47] hashar: oh good, when? [15:37:54] * RoanKattouw looks up [15:38:05] sumanah: not sure, end of may / beginning of june IIRC [15:38:09] .... [15:38:22] New patchset: Reha Sterbin; "Removed buckets 2, 3, 4, and 5 from the js" [mediawiki/extensions/ArticleFeedbackv5] (master) - https://gerrit.wikimedia.org/r/3917 [15:38:28] hashar: link to discussion? [15:38:30] lol [15:38:37] hashar: that's of course a difficult time for us. [15:38:41] sumanah, so this time I have to provide the same support Ashwini I provided the Aashishi to its proposal, right? [15:38:45] lol, I googled "Gerrit Hackathon 2012" and mediawiki.org is the second result [15:38:45] Raylton: no. [15:38:52] chughakshay16 i'm not familiar with the process for getting git-review set up on windows - although i've heard a few people have struggled with it [15:39:36] chughakshay16: are you running git-review -s in your cloned repository? [15:39:43] awjr, i ma getting an error ' Problems encountered installing commit-msg hook , Permission denied(publickey) ' [15:39:51] awjr, yes [15:39:55] sumanah, why? [15:40:14] chughakshay16: Did you add your public key in the Gerrit web UI? [15:40:17] Raylton: well, what kind of support did you give Aashish? and was it all in public on mailing lists/IRC? [15:40:21] Oooh wait [15:40:23] if so you can just point Ashwini to those logs [15:40:26] I remember Aaron's problem with this [15:40:39] awjr, yes i did [15:40:44] His command-line ssh client didn't use the Plink agent to get the key [15:41:09] chughakshay16 are you a registered committer and have you added your public key to gerrit? [15:41:13] New patchset: Reedy; "Change the default setting to match $wgScript if not explicitely set. See http://thread.gmane.org/gmane.org.wikimedia.mediawiki/39432 for the kind of problems produced by leaving it to ''." [mediawiki/extensions/CentralNotice] (master) - https://gerrit.wikimedia.org/r/3717 [15:41:32] chughakshay16: Try downloading the hook file manually. It's at https://gerrit.wikimedia.org/r/tools/hooks/commit-msg and you need to put it in the .git/hooks directory [15:41:36] awjr : i am not a registered commiter.. i added my public key couple of days back [15:41:38] awjr: yes, chughakshay16 has a Gerrit/Labs account [15:41:48] I'm assuming that's what you mean? [15:41:50] sumanah: Gerrit hackaton is May 7-11 on Google Campus apparently http://groups.google.com/group/repo-discuss/browse_thread/thread/5efcefc59f8fcbca# [15:41:55] oh good [15:42:00] sumanah: but he just said he's not a committer [15:42:08] ? [15:42:27] sumanah, by e-mail ... I showed the bugs on bugzilla .. community Wishlist , My own ideas of good solutions etc. [15:42:27] awjr: I think chughakshay16 may have misunderstood your question. [15:42:39] awjr, i made my account couple of days back.. does that mean I am a commiter ? [15:42:39] Raylton: can you just forward all those emails to ashwini? [15:42:44] awjr, :) [15:44:12] sumanah: I will drop a mail to Ryan he might be interested [15:44:16] Raylton: I'll email Ashwini & Aashish to explain what happens next -- basically one of them should probably choose a different topic. [15:44:38] chughakshay16, sumanah, doesn't he need more than a labs account to be able to commit to the mediawiki repo, like to have been accepted as a committer? [15:44:56] Nope, shouldn't do [15:45:25] hashar: ok, thanks [15:45:31] New review: Reedy; "(no comment)" [mediawiki/extensions/CentralNotice] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/3717 [15:45:54] awjr: https://www.mediawiki.org/wiki/Developer_access has an explanation -- a Labs account is also a Gerrit account. [15:46:09] Akshay has developer access. [15:46:09] sumanah, if I forward e-mail I sent to Aashish first have to ask him permission [15:46:17] Raylton: ok, never mind, don't do that yet. [15:46:35] New review: Reedy; "(no comment)" [mediawiki/extensions/AbuseFilter] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3843 [15:46:38] Change merged: Reedy; [mediawiki/extensions/AbuseFilter] (master) - https://gerrit.wikimedia.org/r/3843 [15:46:45] sumanah: i guess i not familiar with the new set up since the move to git :) [15:46:57] New patchset: Reha Sterbin; "Renamed the bucket parameter to aftv5_form" [mediawiki/extensions/ArticleFeedbackv5] (master) - https://gerrit.wikimedia.org/r/3918 [15:47:25] awjr, same here..(well i have no idea at all) :| [15:47:47] chughakshay16, have you added your public key in gerrit? if so, try what RoanKattouw suggested above [15:48:06] awjr, yes arthur.. [15:48:21] RoanKattouw chughakshay16: Try downloading the hook file manually. It's at https://gerrit.wikimedia.org/r/tools/hooks/commit-msg and you need to put it in the .git/hooks directory [15:48:59] RoanKattouw, one more thing even if my ssh-agent isnt connected , i should be able to make commits to the code .. right ? [15:49:23] You're being very vague [15:49:43] What do you mean with your ssh agent being "connected"? [15:49:45] sumanah, you will send you an email to them? [15:49:49] 03(ASSIGNED) Make AFTv5 url parameters work all the time - 10https://bugzilla.wikimedia.org/35587 +comment (10Reha Sterbin) [15:49:52] chughakshay16: i think i remember two other devs using windows who had difficulties getting git-review set up. they're not online right now, but they probably will be later. they might be able to help you out more - AaronSchulz and MaxSem [15:49:57] And by "commit" do you mean make commits or *submit* commits? [15:50:11] * AncientEntity is MaxSem [15:50:15] RoanKattouw, in that workflow document.. for git-review setup i am supposed to go through four steps... [15:50:32] 03(mod) Make AFTv5 url parameters work all the time - 10https://bugzilla.wikimedia.org/35587 +comment (10Reha Sterbin) [15:50:43] lol AncientEntity … i can't keep track of you [15:51:13] RoanKattouw, even the second step of adding ssh-add was giving me an error... [15:51:19] Raylton: I have now done so. [15:51:20] AncientEntity, chughakshay16 is having trouble getting git-review set up under windows [15:51:37] RoanKattouw, https://www.mediawiki.org/wiki/Git/Workflow [15:51:45] Raylton: just sent to wikitech-l and cc'd them. [15:51:46] 03(mod) Remove all feedback forms but option 1 - 10https://bugzilla.wikimedia.org/35586 +comment (10Reha Sterbin) [15:52:03] Oh [15:52:05] Thanks for the heads-up Raylton [15:52:11] Yeah ssh-add is Unix/Linux/Mac-specific, it doesn't work on Windows [15:52:32] ssh-add works in git bash [15:52:35] Oh, OK [15:52:51] Anyway -- git will need to use your key when cloning, updating and submitting things [15:52:57] Basically whenever it contacts the Gerrit server [15:53:03] 03(ASSIGNED) Retire AFTv5 "bucket" url parameter - 10https://bugzilla.wikimedia.org/35589 +comment (10Reha Sterbin) [15:53:14] AncientEntity, i did it from git bash only [15:53:16] Local operations that do not contact the Gerrit server do not need to use the key [15:53:26] sumanah, Now I saw the email ... thank you [15:53:57] You're welcome, Raylton! [15:54:25] RoanKattouw, AncientEntity please help me out with this.. i have spent half of my day to set this up.. so that i could submit bug patches.. and its still not done.. [15:54:31] to make ssh work: go to \etc and create file ssh_config with the following content: [15:54:47] Hostname gerrit.wikimedia.org [15:54:47] Port 29418 [15:54:47] User [15:55:11] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3914 [15:55:14] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3914 [15:55:28] AncientEntity, okay [15:56:19] chughakshay16, note that I currently resort to installing the commit hook manually anyway [15:56:25] AncientEntity, arent we supposed to put this file under ~/.ssh directory ? [15:56:53] I'm telling you what helped me [15:57:09] ~/.ssh is POSIX [15:57:13] Windows is not [15:57:26] lol [15:57:48] AncientEntity, okay even if i install that manually.. i am still gettng error for not being able to connected to authentication agent [15:58:09] *not being connected to authentication agent [15:58:14] eval `ssh-agent` [15:58:24] git/workflow mentions this [15:58:36] AncientEntity, that was the second step/.. [15:58:42] AncientEntity, yeah did that... [15:59:06] AncientEntity, third step is to add to the key to the authentication agent .. thats what i am getting error with [15:59:39] with ssh-add path/to/key? [16:00:10] AncientEntity, yes... this is the command which throws me error [16:00:17] What's the error message? [16:00:28] chughakshay16: i have to finish getting ready to head into the office so im going to sign off - but it seems like you're now in good hands :) [16:00:28] what do you see when starting ssh-agent? [16:01:09] SSH_AUTH_SOCK=/tmp/ssh-tMTlgH9008/agent.9008; export SSH_AUTH_SOCK; [16:01:09] SSH_AGENT_PID=12148; export SSH_AGENT_PID; [16:01:09] echo Agent pid 12148; [16:01:36] awjr, okay.. and thanks [16:01:49] AncientEntity, SSH_AUTH_SOCK=/tmp/ssh-tMTlgH9008/agent.9008; export SSH_AUTH_SOCK; [16:01:49] SSH_AGENT_PID=12148; export SSH_AGENT_PID; [16:01:49] echo Agent pid 12148; [16:02:18] disable windows firewall and UAC [16:03:29] chughakshay16: Did you type eval `ssh-agent` correctly? With the right kind of quotes? It's ` not ' [16:03:34] (backticks) [16:03:53] eval 'ssh-agent'...i used normal quotes [16:03:59] hahaha [16:04:09] that's the problem [16:04:24] AncientEntity, okay let me try it our [16:08:40] New patchset: Kaldari; "fix for bug Bug 27757, applying John Du Harts patch for an API method for retrieving tokens" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3890 [16:08:44] AncientEntity, i am able to add the key to the agent now.. but i am stilll getting error for git-review -s [16:08:55] which one? [16:08:58] AncientEntity, this is the error -Problems encountered installing commit-msg hook [16:08:59] Permission denied (publickey). [16:09:11] RoanKattouw: the interesting thing about "eval `ssh-agent`" is that if there is a system-wide ssh agent, it actually overrides it with a ssh agent for given console session [16:09:14] are you able to ssh into gerrit? [16:09:44] vvv: Yeah, ssh-agent isn't very nicely implemented that way [16:09:46] AncientEntity, yeah that step is going fine.. its printing me a welcome message [16:09:54] 03(mod) Change table.metadata to metadata - 10https://bugzilla.wikimedia.org/31510 +comment (10Max Semenik) [16:12:07] AncientEntity, should i try the next alternative which is given in that document in case of not able to download the commit-msg hook [16:12:11] AncientEntity, -P 29418 -v @gerrit.wikimedia.org:hooks/commit-msg [16:12:46] yes, try installing manually [16:13:08] AncientEntity, then i would be able to submit my patches..right ? [16:13:12] AncientEntity, :) [16:13:25] tryyyyyy [16:13:43] AncientEntity, okay :) [16:15:12] AncientEntity, what should be the name of this file ? [16:15:32] commit-msg [16:16:08] AncientEntity, commit-msg.sample is already present in my hooks folder.. [16:16:11] AncientEntity, ? [16:16:20] so what? [16:16:54] AncientEntity, so should i override this one over the previous one.. ? [16:17:04] pffff [16:17:13] we've been seeing a lot of spam accounts registered with the same account name format (NameName###) that are getting past recaptcha, is there a better extension to stop these? [16:17:19] AncientEntity, please bear with me .. :| [16:17:21] how do you think, why these files have different names? [16:18:11] i know there are a list of spam extensions on the wiki, but i don't know which ones would work [16:18:23] saiyr - yeah, reCAPTCHA barely works any more. QuestyCaptcha is a good one. [16:18:51] It doesn't help in digitizing old books, but what can you do. :) [16:19:07] yeah, so sad :( [16:19:15] AncientEntity, saved it .. what now ? am i done with the setup... ? ... :| [16:19:34] try committing [16:19:53] AncientEntity, okay [16:20:02] New review: Reedy; "(no comment)" [mediawiki/core] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/3790 [16:20:32] anyway, thanks o/ [16:20:55] please don't ping me unless you're asking a question, okays and smilies don't require to be seen immediately [16:24:10] New patchset: Reedy; "(bug 33214) catch exceptions in ApiFeedContributions" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3328 [16:26:28] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Antoine "hashar" Musso) [16:26:29] 03(mod) Setup PHP Test and Code Coverage Analysis - 10https://bugzilla.wikimedia.org/31518 (10Antoine "hashar" Musso) [16:28:47] 03(mod) Progress Icon for Listen In - 10https://bugzilla.wikimedia.org/35573 +comment (10Patrick Hayes) [16:30:15] New review: Reedy; "Bar fixing the help urls per Max, this could be submitted as is, and the extra token functions added..." [mediawiki/core] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3890 [16:33:31] AncientEntity, now when i run this - git pull master , its throwing me fatal: 'master' does not appear to be a git [16:33:31] fatal: The remote end hung up unexpectedly [16:33:54] chughakshay16: try doing git pull origin master [16:34:01] i had the same problem, it worked for me [16:34:12] drecodea_, okay [16:38:08] drecodeam, thanks a lot ! [16:38:35] chughakshay16: happy to help [16:39:45] 03(NEW) Disparity between submit clicktracking events and AFTv5 records created - 10https://bugzilla.wikimedia.org/35590 normal; MediaWiki extensions: ArticleFeedbackv5; (reha) [16:40:42] 03(mod) Disparity between submit clicktracking events and AFTv5 records created - 10https://bugzilla.wikimedia.org/35590 +comment (10Reha Sterbin) [16:42:34] New review: Kaldari; "I've already written the new code for the patrol token support, but apparently hashar's merge has me..." [mediawiki/core] (master) C: 0; - https://gerrit.wikimedia.org/r/3890 [16:44:40] New patchset: Amire80; "(bug 34470) Apply explicit margin-left/right according to directionality." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3920 [16:46:09] 14(WFM) TestSwarm routinely halts in browsers due to temporary errors - 10https://bugzilla.wikimedia.org/30888 +comment (10Krinkle) [16:47:49] 03(ASSIGNED) margin for edit section button on RTL wikis with LTR user interface is on the wrong side - 10https://bugzilla.wikimedia.org/34470 trivial->minor; +comment (10Amir E. Aharoni) [16:47:59] 03(mod) Monitor PHPUnit test coverage in Jenkins - 10https://bugzilla.wikimedia.org/31518 summary (10Krinkle) [16:48:01] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Krinkle) [16:48:50] 03(mod) margin for edit section button on RTL wikis with LTR user interface is on the wrong side - 10https://bugzilla.wikimedia.org/34470 minor->normal; +comment (10Amir E. Aharoni) [16:49:28] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Krinkle) [16:49:30] 03(mod) [TestSwarm] Submit jobs for Special:JavaScriptTest as well - 10https://bugzilla.wikimedia.org/33499 (10Krinkle) [16:50:08] 03(mod) Jenkins job: PHP code style analyzer - 10https://bugzilla.wikimedia.org/35588 +comment (10Krinkle) [16:50:09] 03(mod) Implementing continuous integration workflow (Gerrit/Jenkins/PHPunit/TestSwarm) - 10https://bugzilla.wikimedia.org/35584 (10Krinkle) [16:53:12] 03(WONTFIX) Add 1.18wmf1 to Jenkins - 10https://bugzilla.wikimedia.org/32334 +comment (10Krinkle) [16:55:33] 03jeroendedauw * 10/trunk/extensions/EducationProgram/maintenance/importWEPFromDB.php: fix bug in import script [16:58:25] 03(mod) Upgrade Jenkins from 1.431 to 1.4xx - 10https://bugzilla.wikimedia.org/31877 summary (10Krinkle) [16:59:23] 03(mod) Update PHPUnit from 3.5.x. to 3.6.x on gallium - 10https://bugzilla.wikimedia.org/33714 +ops; summary (10Krinkle) [17:00:50] New patchset: Kaldari; "follow-up to Change I58bc5847 for bug 27757 - adding support for patrol tokens (and making the rc param in getPatrolToken optional)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3921 [17:00:52] New patchset: Kaldari; "fix for bug Bug 27757, applying John Du Harts patch for an API method for retrieving tokens" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3890 [17:07:50] 03(NEW) About Screen gets overlaid on top of article - 10https://bugzilla.wikimedia.org/35591 normal; Wiktionary App: Wiktionary; (pfhayes) [17:08:24] 03(mod) About Screen gets overlaid on top of article - 10https://bugzilla.wikimedia.org/35591 (10Patrick Hayes) [17:12:01] New patchset: Jdlrobson; "add mf-feedback.js file for feedback page" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3924 [17:15:51] I just updated 1.16 to current 1.18.x, and my wiki appears to have lost its frames/outer navigation pages. Anyone know what might be up? [17:19:29] 03(ASSIGNED) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 (10Amir E. Aharoni) [17:21:04] New code comment: Bsitu; yeah, we can use page id as the primary key. I just have a preference that tables have their own pr; [17:21:48] git: how can I comment on a comment, i.e. I want to comment on https://gerrit.wikimedia.org/r/#change,3391 [17:22:16] Wikinaut: you click Review [17:23:00] Where is that ? [17:23:14] Where exactly ? [17:23:30] !git [17:23:31] MediaWiki development is using git, a distributed source control manager, starting on March 21st, 2012; details: https://www.mediawiki.org/wiki/Git_happens Instructions for using it: https://www.mediawiki.org/wiki/Git/Workflow To get an account: https://www.mediawiki.org/wiki/Project:Labsconsole_accounts [17:23:48] Wikinaut: look at the screenshots in https://www.mediawiki.org/wiki/Git/Workflow#How_to_comment_on.2C_review.2C_and_merge_code_in_Gerrit [17:23:58] Seems to have lost formatting too. Database is okay though [17:24:16] Change abandoned: Robmoen; "This was a test." [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3888 [17:24:18] sumanah: Do you mean I haev to enter [17:24:42] my comments intot he field labelled "Cover Message" ? [17:25:04] Wikinaut: Yes. [17:25:06] "Cover Message" == Comment ? [17:25:10] pls rename it then [17:25:16] I saw this [17:25:21] but was fully unsure [17:25:34] and it is not harmonised [17:25:49] because [17:25:56] on https://gerrit.wikimedia.org/r/#change,3391 it is called "Comments" [17:26:10] Wikinaut: would you mind filing a bug about this and marking it as blocking bug 22596? thanks [17:26:11] 03(NEW) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 normal; Wikimedia: Git/Gerrit; (mail) [17:26:24] awjr: hi! [17:26:41] potter: hello [17:26:51] awjr: I sent you mail regarding GSoC 2012 [17:27:14] potter: im working my way through my unread email now :) [17:27:31] sumanah: [17:27:35] awjr: :) [17:27:46] potter: sumanah told me that you are one of the potential mentor for my proposal [17:28:14] awjr: Fyi - http://www.mediawiki.org/wiki/User:Bagariavivek/GSoC_Application . [17:28:15] potter: Arthur (awjr) just mentioned to me that he probably will not be able to mentor this summer, due to conflicting commitments. [17:28:41] but i am still happy to review applications etc [17:28:47] sumanah: oh [17:28:54] !b 35592 [17:28:55] https://bugzilla.wikimedia.org/show_bug.cgi?id=35592 [17:28:59] awjr: thanks :) [17:29:04] !b 35593 [17:29:04] https://bugzilla.wikimedia.org/show_bug.cgi?id=35593 [17:29:07] potter no problem! [17:29:14] sumanah: so only kaldari and Nikerabbit are the potential mentors ? [17:29:20] New review: Robmoen; "(no comment)" [mediawiki/extensions/CentralNotice] (master) C: 1; - https://gerrit.wikimedia.org/r/3717 [17:29:23] potter: no, you silly person. :-) [17:29:29] potter: what makes you say that? [17:29:39] sumanah: i mean for my proposal :) ? [17:29:56] potter: you've looked at https://www.mediawiki.org/wiki/Summer_of_Code_2012#Mentor_signup [17:29:58] sumanah: you had given me these names [17:30:04] Bug 35593 - On Review page change test "Cover Message" to "Comment" [17:30:07] potter: sorry if i told something wrong [17:30:13] suman [17:30:17] sumanah: yeah. I saw that [17:30:17] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10T. Gries) [17:30:30] sumanah: how to say a reviewer, that something is fixed in a later commit ? [17:30:36] in svn it was simply [17:30:44] simple to add follow up [17:30:50] ...revision nnn [17:31:03] or to say fixed in r123456 [17:31:17] (i hate git) [17:31:18] potter: I think your use of "only" in "so only kaldari and Nikerabbit are the potential mentors?" was confusing. Indians use "only" in a way that tends to confuse people like me who use US English. [17:31:23] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 (10T. Gries) [17:31:39] sumanah: :D lol [17:32:00] sumanah: So whom else should i approach ? [17:32:15] Wikinaut: RoanKattouw might have a more specific suggestion for how to indicate a followup revision. [17:32:49] potter: when you look at the Watchlist-related code in MediaWiki's history, who has put work into changing it (recenty) [17:32:49] Wikinaut: Use the change number from the Gerrit URL [17:32:52] recently*? [17:33:06] sumanah: ok. [17:33:19] sumanah: I gotta leave. bye :) [17:33:20] difficult to trace now (regarding my E:RSS) [17:33:24] RoanKattouw: [17:33:28] 03(FIXED) MTC 635 not correctly rendering show/hide on smartphone layout - 10https://bugzilla.wikimedia.org/30063 +comment (10Tomasz Finc) [17:33:31] would like to skip all intermediate [17:33:35] commits [17:33:44] and then to let the final reviewed [17:33:51] E:RSS [17:33:54] Wikinaut: You want to squash them all into one big commit? [17:34:04] I think, that makes sense [17:34:08] because [17:34:09] I have enabled $wgDebugToolbar but where should I see the debug information? [17:34:16] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10T. Gries) [17:34:21] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 (10T. Gries) [17:34:24] Wikinaut: Would you like me to tell you how to do it, or to do it for you? [17:34:24] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10T. Gries) [17:34:26] I tried to have everything fixed before the 21.March [17:34:41] Roan: I have only limited time. Let me check carefully, [17:34:52] what I want to be "squashed" [17:34:56] 03(mod) On Review page change test "Cover Message" to "Comment" - 10https://bugzilla.wikimedia.org/35593 +comment (10T. Gries) [17:35:02] and then I'll sent you by mail [17:35:06] or here [17:35:22] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 (10T. Gries) [17:35:23] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10T. Gries) [17:35:27] KO [17:35:29] *OK [17:35:40] New review: Robmoen; "This is looking good. Though, the CSS change does not follow coding standards. " [mediawiki/extensions/UploadWizard] (master) C: 0; - https://gerrit.wikimedia.org/r/3808 [17:35:41] You may use both Gerrit change IDs or old SVN revision numbers to refer to commits [17:35:53] Or SHA-1sss [17:35:57] I can work with all of those [17:36:34] <^demon|away> Squashing all of those commits into 1 omnibus commit is bad IMHO. Squash ones that are related, but if they're unrelated they shouldn't go together as a gerrit change. [17:36:38] 03(mod) On Gerrit's Review screen, change text "Cover Message" to "Comment" - 10https://bugzilla.wikimedia.org/35593 summary; +comment (10Sumana Harihareswara) [17:37:56] ^demon but if one comments negatively on a single commit, this does not make sense, if that commit is alkready repaired [17:38:01] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10Chad H.) [17:38:03] 03(mod) Migrate subversion to git (tracking) - 10https://bugzilla.wikimedia.org/22596 (10Chad H.) [17:39:13] Wikinaut: Right, but if B fixes A, then A and B are related, and you should squash them [17:39:25] But you shouldn't squash C into them if C has nothing to do with A or B [17:39:47] Roan: I did a lot for RSS. And think, that ... [17:39:55] the final version should be reviewed [17:40:02] OK [17:40:13] let me make a "final" check. [17:40:18] You want me to just take them all and roll them up into one huge commit? [17:40:25] We can always split bits off later [17:40:28] * ^demon|away headdesks [17:40:38] kindergarten [17:40:49] ^demon|away: Seriously though, this is like one huge chain of followup rNNNNN fix blah [17:41:04] yes [17:41:13] it is a limited thing [17:41:17] <^demon|away> Not really. I see a whole bunch of unrelated things here. [17:41:19] * ^demon|away shrugs [17:41:21] <^demon|away> Oh well. [17:41:38] ok, I am out, and leave it up to you [17:41:41] I'll do an interactive rebase and apply judgment [17:41:52] man judgment [17:42:00] in my (svn) commits .. [17:42:00] hmm, command not found [17:42:06] ;-) [17:42:14] indeed I added a feature, then I removed that [17:42:19] for example, [17:42:27] tracking category [17:42:48] it is now "in", but without an extra parameter [17:43:03] simply using a system message for it [17:43:17] and if you put "-" into that message, it's disabled [17:43:35] there's no need for intermediate code review of that [17:43:41] <^demon|away> sumanah: `apt-get install lake-of-fire` should cover libjudgement. [17:43:42] intermediate code [17:43:45] ends [17:44:09] Ha! [17:44:24] 03(mod) On Gerrit's Review screen, change text "Cover Message" to "Comment" - 10https://bugzilla.wikimedia.org/35593 +comment (10Chad H.) [17:45:05] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10T. Gries) [17:45:17] <^demon|away> wikibugs lag? [17:45:38] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10T. Gries) [17:47:48] Roan: can you please look to https://svn.wikimedia.org/viewvc/mediawiki/trunk/extensions/RSS/ and https://svn.wikimedia.org/viewvc/mediawiki/trunk/extensions/RSS/RSS.php?view=log [17:47:52] you will see that [17:48:15] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3909 [17:48:17] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3909 [17:48:18] my last version is https://svn.wikimedia.org/viewvc/mediawiki/trunk/extensions/RSS/RSS.php?revision=114168 [17:48:33] Yes, I noticed [17:48:40] but there are reverted ones... [17:48:47] going back to (mom) [17:48:57] New patchset: Drecodeam; "Added HTML5 drag and drop support" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3808 [17:49:00] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3910 [17:49:02] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3910 [17:49:16] r111347 or so [17:50:16] I don't have an overview and deep knowledge of git, and think, all of the reverts-for-git-migrations of E:RSS should be squashed [17:50:18] into one [17:50:32] New patchset: Catrope; "Squashed commit of all RSS changes ported from SVN:" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3925 [17:50:32] and then I can comment [17:50:39] why I did this and that [17:50:59] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +upstream; +comment (10Chad H.) [17:51:00] Change abandoned: Drecodeam; "Abandoning change due to some error. Wrong file got pushed." [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3808 [17:51:02] There [17:51:09] merci bedankt danke [17:51:14] If there's unrelated stuff in there, it can be split out [17:51:15] thanky [17:51:19] yes [17:51:31] but RSS is fine now, in my view [17:51:42] Simply by creating a new commit with that change only, submitting it for review, and then amending the big one to drop that change [17:52:01] New review: preilly; "(no comment)" [mediawiki/extensions/MobileFrontend] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3911 [17:52:04] Change merged: preilly; [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3911 [17:52:46] 03(mod) About Screen gets overlaid on top of article - 10https://bugzilla.wikimedia.org/35591 +comment (10Sam Reed (reedy)) [17:53:45] New patchset: Catrope; "Squashed commit of all RSS changes ported from SVN." [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3925 [17:55:31] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10T. Gries) [17:56:11] 03(mod) Jenkins job: PHP code style analyzer - 10https://bugzilla.wikimedia.org/35588 normal->15enhancement (10Sam Reed (reedy)) [17:57:42] hey hexmode - if you're working on a triage report re the Git triage, http://etherpad.wikimedia.org/BugTriage-2012-03 has been updated with more info from Chad [17:58:30] sumanah: k [17:58:38] thanks [17:59:39] hexmode: seems like some of these issues are of interest to the general community -- is there anything more urgent on your plate, or do you think you can get that triage report out today? [18:02:04] New review: Kaldari; "Good idea. Don't know why I never thought to add this myself." [mediawiki/extensions/CentralNotice] (master) C: 1; - https://gerrit.wikimedia.org/r/3717 [18:02:08] hexmode: but really there's just a few specific items there that need publicizing ... any updates on https://www.mediawiki.org/wiki/Review_queue ? [18:02:43] New patchset: Catrope; "Fix bug that dropped additional info for save events" [mediawiki/extensions/ClickTracking] (master) - https://gerrit.wikimedia.org/r/3926 [18:03:31] sumanah: no, I've been working on the bz-patch-to-gerrit thing today [18:04:04] hexmode: ah, cool. looking forward to that [18:04:17] sumanah : just a quick question ? [18:04:20] chughakshay16: go ahead [18:04:54] sumanah : do i still need to post a patch to bugzilla to have some one confirm it or could just commit it to the code right away.. ? [18:04:54] hey gicode -- thought you might want to know that hexmode is working on getting Bugzilla patches turned into Gerrit branches automatically [18:05:20] chughakshay16: go ahead and commit it via Git right away, and then post a link to the changeset in a Bugzilla comment. [18:05:45] Change restored: Drecodeam; "(no reason)" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3808 [18:05:46] chughakshay16: If your commit addresses a bug in Bugzilla, please comment on that bug to note that the commit is in the merge queue, and link to its changeset in Gerrit. [18:06:05] New patchset: Drecodeam; "Added HTML5 drag and drop support" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3808 [18:06:06] gicode: sumanah: the ugly thing is the bz api doesn't have all the query strength that the web ui does [18:06:08] * hexmode sighs [18:06:14] hexmode: that is annoying! [18:06:19] 03(mod) ProfilerSimpleUDP does not check for function_exists( 'socket_create' ) - 10https://bugzilla.wikimedia.org/35580 +comment (10Dmitriy Sintsov) [18:06:21] * hexmode toddles off to work on the report [18:06:25] thanks hexmode [18:06:29] :) [18:06:30] New review: Reedy; "(no comment)" [mediawiki/extensions/ClickTracking] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3926 [18:06:32] Change merged: Reedy; [mediawiki/extensions/ClickTracking] (master) - https://gerrit.wikimedia.org/r/3926 [18:07:22] AaronSchulz: I believe you're on 20% today? if you could sweep through https://www.mediawiki.org/wiki/Review_queue a little bit and clarify the statuses of a few of those, that would be neat [18:07:46] 03catrope * 10/branches/wmf/1.19wmf1/extensions/ClickTracking/ClickTracking.hooks.php: 1.19wmf1: Merge https://gerrit.wikimedia.org/r/3926 [18:07:58] sumanah: ping [18:08:02] hi nit3ch [18:08:05] !gsoc [18:08:06] MediaWiki participates in the Google Summer of Code mentorship program. http://www.mediawiki.org/wiki/GSoC Please read http://en.flossmanuals.net/GSoCStudentGuide/ if you're thinking of applying to GSoC. Student applications will be accepted between March 26 and April 6. [18:08:48] suamanah : okay, i got my gerrit issue fixed. happy !! :) [18:08:54] great, glad to hear it, chughakshay16! [18:09:10] chughakshay16: if your problem's solution did not appear in the documentation on mediawiki.org then please add it. [18:09:37] sumanah:thanks for the link [18:09:50] sumanah : if u dont mind , i didnt get you completely what you said earlier :S [18:10:42] sumanah: http://www.mediawiki.org/wiki/Extension:AuthDrupal i would like to extend this project for drupal 7 and fix existing issues and features [18:10:47] New review: Kaldari; "OK, the change for Max is merged into this change and the change for adding patrol tokens is Change ..." [mediawiki/core] (master) C: 1; - https://gerrit.wikimedia.org/r/3890 [18:11:21] 03bsitu * 10/trunk/extensions/PageTriage/ (4 files in 2 dirs): alter pagetriage tables and remove redundant metadata [18:11:42] 03(mod) ProfilerSimpleUDP does not check for function_exists( 'socket_create' ) - 10https://bugzilla.wikimedia.org/35580 +comment (10Sam Reed (reedy)) [18:11:52] nit3ch: a good idea! [18:12:09] chughakshay16: you mean that you didn't understand when I said "if your problem's solution did not appear in the documentation on mediawiki.org then please add it."? [18:12:53] chughakshay16: if https://www.mediawiki.org/wiki/Git/Workflow didn't help you fix your problem, then please improve https://www.mediawiki.org/wiki/Git/Workflow [18:13:02] sumanah : no, nothing i got ur point.. when i read that again 'If your commit addresses a bug in Bugzilla, please comment on that bug to note that the commit is in the merge queue, and link to its changeset in Gerrit.' [18:13:03] nit3ch: go ahead! [18:13:23] nit3ch: am I right in inferring that you want to do this as a Google Summer of Code project? [18:13:27] 03(mod) Update PHPUnit from 3.5.x. to 3.6.x on gallium - 10https://bugzilla.wikimedia.org/33714 +comment (10Sam Reed (reedy)) [18:13:53] chughakshay16: ok. So, you are committing code because you want to solve a bug, right? and this bug is reported in a Bugzilla issue, correct? [18:14:03] sumanah : yes.. [18:14:11] sumanah:ya [18:14:19] chughakshay16: ok. So that is why I said that your commit addresses a bug in Bugzilla. [18:14:38] sumanah: i have done some contribution to drupal but nothing in mediawiki but will love to [18:14:42] chughakshay16: so, you comment on that bug. You mention that you have made a commit that is now awaiting merge (it is in the merge queue). [18:14:48] Amend the commit, add to the summary! [18:14:51] chughakshay16: and you provide a link to the changeest [18:15:10] nit3ch: you have read https://www.mediawiki.org/wiki/Summer_of_Code_2012#Student_applications ? [18:15:36] sumanah: i read it [18:15:57] nit3ch: ok. So what do you think is a good step for you to take next? [18:16:38] sumanah : thank you sumanah , for listening to my trivial questions and answering them too.. its just that i m new to git and gerrit so get stuck with small issues. [18:16:59] sumanah: write a good proposal [18:17:03] chughakshay16: you're also new to a lot of the English-language jargon around software development [18:17:26] sumanah : yeah that too.. [18:17:46] and our jargon is not as clear as one would like. [18:17:57] nit3ch, yes, that's important [18:18:14] nit3ch: you will be competing against people who have already been working on MediaWiki for several months [18:18:24] sumanah : i guess with time i would get used to it.. :| [18:18:26] nit3ch: so pointing to your previous coding work will be important [18:18:31] chughakshay16: yes [18:18:44] is there a way to embed a What_links_here list in a page? [18:18:57] balrog: {{Special:Whatlinkshere/Foobar}} might work [18:18:58] chughakshay16: http://blog.steveklabnik.com/posts/2011-12-15-marx-anarchism-and-web-standards may interest you -- about the conflict between learnability and usability [18:19:23] sumanah: okay , is there any need to mnetion about my git or svn experience [18:19:26] nit3ch: and you should also follow the steps in https://www.mediawiki.org/wiki/How_to_become_a_MediaWiki_hacker [18:19:34] looks like what I may want is Semantic MediaWiki [18:19:36] nit3ch: yes, those are good things to mention as well [18:19:49] 03(mod) replacing "hidden by Article Feedback V5" link with the actual account - 10https://bugzilla.wikimedia.org/35501 +comment (10Elizabeth M Smith) [18:20:09] nit3ch: when you write a proposal, here are the other proposals that you are competing with: https://www.mediawiki.org/wiki/Category:Summer_of_Code_2012_applications [18:20:35] nit3ch: while I am glad that you are interested in working with MediaWiki, I am curious -- since you already have Drupal experience, why are you more interested in working with MediaWiki than with Drupal? [18:20:43] 03kaldari * 10/trunk/extensions/PageTriage/tests/phpunit/SpecialPageTriageTest.php: creating test user and test articles for further unit tests to come [18:21:09] nit3ch: after all, they are also participating in GSoC https://www.google-melange.com/gsoc/org/google/gsoc2012/drupal [18:21:32] sumanah: ya i know [18:22:06] and also submitted propsal for drupal [18:22:57] 03kaldari * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ (8 files in 2 dirs): adding some graphics for the list view, changing red text to dark red text [18:23:19] mediawiki is a great community and i want to be a part of [18:23:44] that's lovely to hear. [18:24:11] i am invlove with drupal more than a year , that's why i choose a project which is a bridge between druapl and mediawiki [18:24:47] and my love to drupal is like .... , i cann't stop contributing to drupal :P [18:25:40] 03(mod) replacing "hidden by Article Feedback V5" link with the actual account - 10https://bugzilla.wikimedia.org/35501 +comment (10Yoni Shostak) [18:26:12] sumanah : blog link was really worth a read :| [18:26:35] nit3ch: :-) [18:28:00] sumanah: can you suggest me soem mentor for thsi project [18:28:04] some* [18:28:30] nit3ch: have you contacted the authors of the extension you want to improve? [18:29:29] authors name are not linked [18:30:56] nit3ch: then you might have to do a little detective work. [18:31:11] 04(REOPENED) AFT5 Feedback Page Permalinks to Hidden, Oversight requested and Oversighted Posts - 10https://bugzilla.wikimedia.org/34883 +comment (10Elizabeth M Smith) [18:32:52] sumanah: :P okay [18:40:36] 03(mod) Enable Extension:ShortUrl on or.wikipedia, ta.wikipedia... - 10https://bugzilla.wikimedia.org/1450 +comment (10Sumana Harihareswara) [18:44:06] 03kaldari * 10/trunk/extensions/PageTriage/ (3 files in 2 dirs): displaying status icons, styling the header and footer [18:45:34] ok, I have to go for a bit [18:45:39] see you online later [18:51:02] Is apache's mod_cache recommended? I wonder cos I don't find anything about on mw.org [18:52:03] mod_disk_cache to be more precise [18:52:41] It's a web cache, so it's irrelevant about the MW part [18:52:53] should be same as using squid/varnish/... [18:54:59] kk. [18:55:31] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10Chad H.) [18:57:16] New patchset: awjrichards; "Added special page to act as controller for special events, including feedback form handling" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3928 [19:02:31] 03(mod) MobileFrontend extension outputting invalid HTML in footer - 10https://bugzilla.wikimedia.org/30406 +comment (10Daniel Friesen) [19:18:57] 03(mod) multiple templates and mw templates - 10https://bugzilla.wikimedia.org/35554 +comment (10John McClure) [19:20:33] New patchset: Amire80; "(bug 25277) Moved the space before the dirmark." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3929 [19:21:41] 14(WFM) hat symbol in templatename (embed in field) - 10https://bugzilla.wikimedia.org/35500 +comment (10John McClure) [19:22:17] 03(ASSIGNED) invisible directional character on File page pollutes pasted filenames - 10https://bugzilla.wikimedia.org/25277 -need-review -patch ; +comment (10Amir E. Aharoni) [19:23:23] New patchset: MaxSem; "(bug 35425) Respect " [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3930 [19:24:41] 03(FIXED) Mobile version does not handle __NOINDEX__ - 10https://bugzilla.wikimedia.org/35425 +comment (10Max Semenik) [19:26:39] is there any way to reach Database object via $context (I'm in SpecialBlock::processForm I'd like to open/close transactions) [19:26:56] Just do it yourself [19:27:06] $dbw = wfGetDb( DB_MASTER ); [19:27:26] cheater! [19:27:48] yeah, that's cheating ;( [19:29:06] actually transactions would be handled best at the context/request level [19:29:07] 03(mod) multiple templates and mw templates - 10https://bugzilla.wikimedia.org/35554 +comment (10Yaron Koren) [19:30:10] New review: Reedy; "(no comment)" [mediawiki/extensions/CentralNotice] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3717 [19:30:13] Change merged: Reedy; [mediawiki/extensions/CentralNotice] (master) - https://gerrit.wikimedia.org/r/3717 [19:31:45] saper, we've got shitloads of code that assumes otherwise, so live with it as it is :| [19:32:29] New review: Robmoen; "(no comment)" [mediawiki/extensions/UploadWizard] (master) C: 1; - https://gerrit.wikimedia.org/r/3808 [19:33:47] MaxSem: yeah I know [19:33:52] 03(mod) MobileFrontend extension should stop special-casing main page - 10https://bugzilla.wikimedia.org/30405 +comment (10Max Semenik) [19:33:56] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3468 [19:33:57] I rewrote one Zope/Python app once [19:33:59] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3468 [19:34:06] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3470 [19:34:10] they did "BEGIN"/"COMMIT" by hand [19:34:12] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3469 [19:34:15] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3470 [19:34:16] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3469 [19:34:44] and they didn't know that Zope nicely commits and rollback transactions on all transactional data sources for them [19:34:59] so we ended up with data losses [19:35:28] like "put some data" " begin " try-insert" "failed?" "rollback" .... and "put some data" were lost [19:35:39] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3891 [19:35:42] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3891 [19:36:36] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3473 [19:37:06] so I hired one more python guy and rewrote all those parts during the weekend [19:37:40] 03(FIXED) Upload wizard: RTL support is incomplete - 10https://bugzilla.wikimedia.org/24692 +comment (10Amir E. Aharoni) [19:44:30] saper: around/ [19:44:32] ? [19:45:58] 03(NEW) Meta-oversight mailing list creation - 10https://bugzilla.wikimedia.org/35594 normal; Wikimedia: Mailing lists; (Courcelleswiki) [19:46:42] 03(mod) Tell git on gerrit.wikimedia.org not to generate spurious terminal sequences - 10https://bugzilla.wikimedia.org/35462 (10Mark A. Hershberger) [19:47:06] hexmode: yes [19:47:41] MaxSem: how long can I "have" result of wfGetDB? Is it ok to move it out the method to the constructor ($this->dbw=wfGetDB()) ? [19:47:42] saper: hashar said you were going to patch git upsteam for https://bugzilla.wikimedia.org/35462 so I've assigned it to you [19:47:47] New patchset: Catrope; "Add HTML DOM -> linear model converter" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3931 [19:48:10] saper, Y NOT? [19:48:12] hexmode: I guess this could be solved by configuration @our place that's why I posted it to bugzilla [19:48:40] but I didn't yet figure out how to set TERM for git within gerrit [19:49:05] saper: what about /etc/profile or ~/.bash_profile/ [19:49:06] ? [19:49:17] dang shift key [19:51:39] hexmode: I tried setting it before gerrit started but it didn't help [19:51:51] saper: or, since it looks like it just runs via an init script... [19:51:52] it doesn't work on a client side either [19:51:53] oh [19:52:13] I tested it on my toy gerrit instance [19:52:23] I don't know really if it's a client issue or server [19:53:02] *IF* it's server and *IF* it is something so simple like adding TERM=dump to our gerrit environent, that's why I think this bug is there [19:53:17] :) [19:53:30] in the unlikely event as hashar suspects [19:54:32] tl;dr assigning it to me is fine :) [19:54:58] hexmode: I wonder, does git have a bug tracker at all? :) [19:55:00] sumanah: thanks for more appropriate-forwarding [19:55:15] hi YuviPanda! I may not be done yet! [19:55:17] saper: they store it in git ;0 [19:55:26] sumanah: :D [19:55:39] YuviPanda: re WikiPronouncer - you've already asked mobile-l? [19:55:48] MaxSem: I guess calling wfGetDB( DB_MASTER ) is somehow expensive? [19:55:51] 03(NEW) Deprecate old MF format=json API hack - 10https://bugzilla.wikimedia.org/35595 enhancement; MediaWiki extensions: MobileFrontend; (maxsem.wiki) [19:56:02] sumanah: i'm not sure if I should. The code was utterly incomplete (the student ran into sdk issues that were not fixable) [19:56:28] hexmode: checkout fossil-scm.org, excellent vcs w/integrated wiki & bugtracker - you clone everything together [19:56:35] YuviPanda: Still. Point to the person, the project, make it discoverable by potentially interested people, on the list now and later for SEO [19:56:44] bugtracker is unfortunately poor [19:56:46] hi chrismcmahon - ready for the call in a few min? :-) [19:56:49] saper, should not be really slower than getting a slave connection [19:56:51] saper: It's not, the connection os cached [19:56:55] sumanah: okay. Do you think you can do that? [19:56:55] *is [19:57:04] YuviPanda: I'm not on mobile-l, you are :-) [19:57:46] YuviPanda: just forwarding the mobile-related bits of your report would do, right? [19:58:12] sumanah: should be. [19:58:19] then go ahead :-) [19:58:50] RoanKattouw: ok I think I need moving wfGetDB() from RecentChanges:::save to the constructor *OR* give $dbw as an optional parameter [19:59:31] RoanKattouw: I did that for LogPage already __construct gets optional $dbw [20:04:33] alolita1: discussing https://www.mediawiki.org/wiki/User_talk:Cmcmahon#Community_Testing.2FQA now [20:04:41] ^demon|away: We have *@users.mediawiki.org in git, but we can't tell gerrit to associate our imported commits with us because we can't confirm those addresses. [20:04:52] Let's try again, im trying to upgrade from 1.15.3 to 1.18.2, but i get Database returned error "1146: Table 'civilbilar.user' doesn't exist (localhost)" http://dpaste.com/hold/723552/ [20:04:57] sumanah: yup [20:05:06] walking over to conf room [20:05:09] <^demon|away> Dantman: I know. Which is why I want that to be a forwarding address :) [20:05:11] is tomasz joining in [20:06:03] ^demon|away: Btw, did you see Tim's comment on gerrit's magic port [20:06:11] <^demon|away> Nope [20:06:33] ^demon|away: get a 2nd ip, use port 22 and redirect [20:06:44] New review: Catrope; "Testing Chad's -1 change" [test/mediawiki] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/3793 [20:07:54] sumanah: where do i join in? [20:09:47] * aharoni got the git blues [20:09:55] aharoni: :) see [20:10:08] Errors running git rebase -i remotes/gerrit/master. Interactive rebase already started [20:10:22] i got that after running `git review' [20:10:26] aharoni: git rebase --abort [20:10:35] working on master? [20:10:41] no, branch [20:10:57] your branch or REL1_19? [20:11:20] 03(FIXED) the Gerrit message "I would prefer that you didn't submit this" is very discouraging - 10https://bugzilla.wikimedia.org/35406 +comment (10Chad H.) [20:11:44] eh... my branch. [20:12:03] I'd do "git review -R" OR "git push gerrit HEAD:refs/for/master" (if it's a new change) or "git push gerrit HEAD:refs/changes/XXXX" (if it's change number XXXX) [20:12:09] good [20:12:22] then try "git review -R" [20:12:54] `git rebase --abort'... kinda... helped... but now i get: [20:13:25] error: could not apply c77669f... / hint: after resolving the conflicts, mark the corrected paths / hint: with 'git add ' and run 'git rebase --continue' [20:13:45] don't rebase :) [20:13:52] but why would i have conflicts? :( [20:14:11] git sucks [20:14:22] is your commit conflicting with master or what? [20:14:24] oh, and it switched me to a branch called "(no branch)" [20:14:42] aharoni: yeah [20:14:48] you have to run. fast. [20:14:53] git reset --soft HEAD [20:15:08] I do this and "git rebase --abort" [20:15:16] seriously? i once thought that git sucks, but now i don't think that it sucks. i do think that gerrit and git review suck. [20:15:18] and never touched git-review again :) [20:15:36] Reedy the web setup page shows more info, http://dpaste.com/hold/723681/ [20:16:09] git has one architectural problem, it does not work with diffs (patches) at all and that makes it unsuitable for what gerrit wants to achieve [20:16:10] ^demon|away: To be precise: http://p.defau.lt/?lC7meVURJh2QJuijNFOKxA [20:16:20] 03(mod) Activating Upload Wizard Extension in fa.wiki - 10https://bugzilla.wikimedia.org/31070 +comment (10reza1615) [20:17:34] saper: You may take that as an architectural problem, but that almost looks like a good thing to me [20:17:44] 03(mod) Activating Upload Wizard Extension in fa.wiki - 10https://bugzilla.wikimedia.org/31070 +comment (10Sam Reed (reedy)) [20:18:33] I like Gerrit better than Phabricator partly because it works with real git commit objects rather than diffs/patches [20:19:32] 03(mod) ProfilerSimpleUDP does not check for function_exists( 'socket_create' ) - 10https://bugzilla.wikimedia.org/35580 +comment (10Dmitriy Sintsov) [20:19:51] * Dantman wishes we could have setup a custom review system that combined the advantages of gerrit's gated trunk and pull-requests [20:19:52] Dantman: no, I take it as a reasonable choice - finally a good tool to manage sets of changing files [20:20:06] Dantman: that's why it's different and this is good [20:20:10] it's not RCS [20:20:47] Dantman: I totally want to write my own code review system now [20:20:52] ;) Yeah, that's not an architectural problem... that's just more evidence that gerrit's methodology is fundamentally wrong [20:20:59] But of course I won't have time to do that for years [20:21:03] lol [20:21:21] RoanKattouw: by the way, I remember you told me about free pushes for non-master branches. Is it implemented now? [20:21:47] vvv: It's not [20:22:00] I've been meaning to start a discussion about this (and an alternative approach) on wikitech-l [20:22:11] What's alternative? [20:22:20] On http://www.mediawiki.org/wiki/Manual:$wgUseCommaCount it is said that "This feature was deprecated (or possibly removed completely) in version 1.18." but WMF wikis are using MW 1.19 and http://noc.wikimedia.org/conf/InitialiseSettings.php.txt still have wgUseCommaCount for some wikibooks projects [20:22:30] Give liberal +2 access on non-master branches [20:22:30] shoudn't them be updated? [20:22:43] (to wgArticleCountMethod ) [20:22:44] Dantman: looking now at Mercurial's MQ system, interesting but somehow I'm not sure I like it [20:23:17] RoanKattouw: One of my thoughts was to use real merges and have a system based on the same (every use has a public repo) setup; Where the review system pulls commits (instead of a hack push) from a user's public repo. With the possibility of integrating things like web hooks so that you can push things to your own repo and have them automatically become review branches. [20:23:31] helderwiki: $wgUseCommaCount is still there, but marked as deprecated [20:23:34] So it's still working [20:23:47] hmm, ok. No problem then, right? [20:23:48] as long as $wgArticleCountMethod is null [20:25:14] no reason we can't migrate it over now though [20:25:21] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10T. Gries) [20:26:52] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10Chad H.) [20:26:54] New patchset: Amire80; "Added *.kate-swp to .gitignore." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3932 [20:27:39] New patchset: awjrichards; "Revert "redesign feedback form", which was a test merge to see how merging from remote branches is handled" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3933 [20:27:39] New patchset: awjrichards; "Make javascript mf_useformat cookie setting set a domain that will cover all subdomains; Update desktop view URL to not include useformat=desktop" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3934 [20:27:40] New patchset: awjrichards; "Remove usage of wgCookiePrefix for mf_useformat cookie; make cookiepath and cookiedomain configurable in JS cookie writing code; pass cookie path and cookie domain to JS from mobile frontend body" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3935 [20:27:41] New patchset: awjrichards; "Wrapping up cookie handling fixes; passing in domain/path to JS from PHP to ensure that values will match; fixed js cookie generation to not generate two different cookies...; retrurning to use ?useformat=desktop in desktop view URL to offer a graceful fa" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3936 [20:27:42] New patchset: awjrichards; "Added a way to set values for pre-existing object properties with configuration variable () in MobileFrontend; Added relevant test method" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3937 [20:27:43] New patchset: awjrichards; "fix slide up animation for browsers which support position fixed" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3938 [20:27:43] New patchset: awjrichards; "submitting form opens first search result of previous search" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3939 [20:27:44] New patchset: awjrichards; "delay clear search so it doesn't interfere with blur event" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3940 [20:27:45] New patchset: awjrichards; "normal line height for iphone and ipad" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3941 [20:27:45] ahhhhh [20:27:46] New patchset: awjrichards; "enabling toggling on for browsers that don't support querySelectorAll and indexOf" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3942 [20:27:46] New patchset: awjrichards; "re-enabled toggling on the footer" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3943 [20:27:47] New patchset: awjrichards; "reveal footer/section when open" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3944 [20:27:48] New patchset: awjrichards; "enable login form on beta" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3945 [20:27:49] New patchset: awjrichards; "Added special page to act as controller for special events, including feedback form handling" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3928 [20:29:05] <^demon|away> RoanKattouw: I really really don't want to write CodeReview v2 :( [20:29:28] New review: Reedy; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3932 [20:29:31] Change merged: Reedy; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3932 [20:29:41] ^demon|away: I'm not talking MW-specific, I'm talking a general code review system [20:29:52] Just like Gerrit, but without all the things that are wrong with Gerrit [20:30:14] <^demon|away> Actually the things wrong with gerrit are growing on me :p [20:30:19] <^demon|away> That might not be good. [20:30:20] Not written in Java, proper support for feature branches, non-terrible UI, ... [20:30:40] <^demon|away> What do you mean proper branch support? [20:30:47] Before or after pyppet? ;) [20:30:51] Well, currently to do follow-ups in Gerrit you have to --amend [20:31:04] <^demon|away> No you don't. There's about 3 different ways you can do it [20:31:06] The reason that's better than committing a real followup is mostly because of centralization of comments [20:31:06] <^demon|away> Amend is one [20:31:11] heh [20:31:19] "Not written in Java" [20:31:26] 03kaldari * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ (8 files in 2 dirs): updating graphics [20:31:28] Ruby [20:31:37] <^demon|away> No Ruby. [20:31:56] Followup tracking in general is a good thing, also for things that have already been merged [20:32:03] The ops people might not like ruby. But tbh it does have more value going for it than java imho. [20:32:09] Could be done fairly trivially with proper feature branch support [20:32:15] <^demon|away> I don't understand the java hate. [20:32:20] what was the reason we didn't just pick Github? i committed some simple jQuery fixes in Github today and it was such a relief after the Gerrit madness. [20:32:35] Also, imagine how cool it would be if you could have a jQuery frontend to git rebase -i :D [20:32:40] <^demon|away> aharoni: Because we're not hosting our code on a 3rd party website. [20:32:47] aharoni: Ahm, not open source, don't want to rely on 3rd party service, etc. [20:32:50] <^demon|away> Been there, done that, got burned. [20:32:50] aharoni: Proprietary control and gated trunk [20:32:58] Oh, yeah, gated trunk [20:33:03] with review [20:33:12] Ryan Lane hates pull-requests [20:33:27] <^demon|away> pull requests are nice from a user standpoint but they totally suck for review. [20:33:30] I guess Java would be an appropriate language to write *part* of this system in [20:33:41] Like, the backend that wraps around git [20:33:45] <^demon|away> Just fork gerrit ;-) [20:33:49] (Seriously, it should wrap around git, not reimplement all of git) [20:33:52] *shudder( [20:33:56] <^demon|away> Like they forked Reitveld [20:34:07] And then rewrote it in a different language? [20:34:17] <^demon|away> Well that was v2.0 :p [20:34:18] I think I would feel more comfortable rewriting the backend in C++ to be honest [20:34:23] I would say Python but I don't speak it [20:34:32] RoanKattouw: why? [20:34:35] The frontend should be a real web app [20:34:52] How much of it in C++? [20:34:54] <^demon|away> I guess all that javascript's not friendly to text browsers. [20:34:59] That too [20:35:00] :) [20:35:06] 03(mod) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 +comment (10T. Gries) [20:35:13] I guess there would be three parts here [20:35:14] <^demon|away> Oh, and the backend is mostly jgit, which is an upstream thing from gerrit [20:35:19] As there are three parts to Gerrit [20:35:41] The git wrapper (Gerrit's isn't a wrapper, but a full implementation of SSH and git) [20:35:56] in C++ probably, could be Java [20:36:20] The backend of the web interface + the API that AJAX requests use, I'd write that in PHP probably [20:36:22] My view is it should be possible to have a good review system that doesn't need a real wrapper around ssh/git. [20:36:23] <^demon|away> gerrit is an ssh daemon wrapped around jgit with a JSON/RPC interface for the UI :) [20:36:25] huh? i thought that github's source is open. [20:36:25] And the frontend in JS [20:36:26] Someone created an account for your e-mail address on Labs (https://labsconsole.wikimedia.org/w/index.php) named "Pokeswap", with password "******". [20:36:29] You should log in and change your password now. [20:36:31] <^demon|away> aharoni: Not at all. [20:36:35] where do i go to sign in [20:36:45] aharoni: No, that's Gitorious. GitHub is completely proprietary. [20:37:03] Note that in Gerrit, the entire UI is built from JavaScript, and that JavaScript isn't even written in JavaScript, it's written in Java [20:37:21] And the web server is Jetty, which is a pain in the ass [20:37:33] how do i get gerrit access now that i have a labs account? [20:37:42] <^demon|away> Login to gerrit with that username/password. [20:38:15] RoanKattouw: why not just write everything in PHP? [20:38:23] I really dislike how Gerrit reimplements half an operating system, and a lot of that is because they wrote it in Java [20:38:26] <^demon|away> Are there any good git libraries in php? [20:38:29] vvv: A git wrapper in PHP? [20:38:31] <^demon|away> Or would we have to DIY there? [20:38:47] Doing socket I/O on a network port? [20:38:54] Hmmmm [20:39:04] That's a valid reason to use a low-level language [20:39:09] See, there's Gerrit and Phabricator [20:39:10] the labs one? [20:39:14] <^demon|away> Yes [20:39:33] The Gerrit people didn't think much further than the very back end of the backend (git wrapper), and they did a great job at that part [20:39:40] RoanKattouw: since when Java became a low-level language? [20:40:04] However, because they wrote that part in Java (and it makes sense to write that in a system-level (-ish) language), they wrote the rest in Java (even the frontend!) and that sucks [20:40:07] Phabricator is the opposite [20:40:19] how do i set ssh KEY/find one [20:40:23] They wrote a good frontend in PHP, but their backend (even the CLI scripts) is also in PHP and it's not as good [20:40:44] pokeswap: you need to create one [20:41:16] RoanKattouw: so what we need is a C backend and PHP frontend? [20:41:37] Sort of [20:41:49] C backend, PHP middle....end?, JS frontend [20:41:53] ok [20:41:58] <^demon|away> That's already sounding nasty. [20:42:00] RoanKattouw: JS front-end? [20:42:13] <^demon|away> JS front-end won't work on w3m. [20:42:14] You mean HTML5+CSS+JS front-end? [20:42:15] Ell [20:42:18] Yeah, sorry [20:42:22] 03raindrift * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ (2 files): fixed floaty bar to have a z-index so it shows up above list view always, and reduced width to account for border [20:42:36] I say this to distinguish it from Gerrit's let's-write-our-JS-in-Java-cause-we-know-Java insanity [20:42:50] But yeah PHP for the web-facing stuff [20:42:54] That's really the frontend in this case [20:43:04] it is telling be no file found [20:43:08] I can tolerate PHP under only one condition. [20:43:17] <^demon|away> hiphop? [20:43:19] RoanKattouw: I personally prefer C+Python+PHP hybrid [20:43:27] vvv: Where would the py come in? [20:43:31] ;) We use a number of the classes we have in core as part of the backend. [20:43:41] I could be convinced to write parts in Python. I just don't speak it so I wouldn't pick it on my own [20:43:41] PHP on it's own is insanity. [20:43:53] <^demon|away> Dantman: So we've now come full circle...we're back to CodeReview v2 [20:44:14] ;) Why do you think I avoid PHP when starting a new project? [20:44:25] <^demon|away> We could write it using the ZendFramework instead ;-) [20:44:35] RoanKattouw: well, I usually prefer C for sensitive calculations and other critical stuff, Python for logically involved stuff, PHP for front-end and something else for desktop GUI [20:44:36] ^demon|away: That's option A). Option B) would be a framework. [20:44:46] Here we'll apparently either have py+php or c+php [20:45:03] New patchset: saper; "Demonstration of dbw concept" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3946 [20:45:14] <^demon|away> vvv: We could do the low-level stuff in C/C++ as a php extension. [20:45:22] * RoanKattouw beats ^demon|away [20:45:27] RoanKattouw: is it okay to go along the lines of https://gerrit.wikimedia.org/r/3946 ? [20:45:27] <^demon|away> Similar to pecl's svn extension [20:45:30] <^demon|away> RoanKattouw: What? [20:45:35] Oh, hmm [20:45:38] The *really* low-level stuff [20:45:40] <^demon|away> I was being serious :) [20:45:41] ^demon|away: well, we already do that [20:45:41] <^demon|away> Yeah [20:45:42] Yeah, that could work [20:46:16] <^demon|away> So then at the php level, you'd have stuff like git_diff(), etc. [20:46:28] saper: Looks good but why do you want it? [20:46:35] Classes might be nicer. [20:46:54] Dantman: You could build PHP classes around that of course [20:47:09] RoanKattouw: I need this to wrap SpecialBlockpage in a DB transaction to fix bug 35572 [20:47:14] ^demon|away: The reason I wouldn't do it this way on my own: writing PHP extensions is fucking horrible [20:47:16] ^demon|away: you can make it OOP [20:47:19] ** Gerrit needs a custom ssh/git backend because it abuses git and creates fake constraints that don't exist in git. IMHO it should be possible to create a good review system that sticks to standard git constraints. [20:47:26] Oh, right, extensions can use OOP [20:47:31] RoanKattouw: limilar will be done for LogPage (I can see that newer code already does this, like RevisionDeleteUser) [20:47:31] RoanKattouw: is it? It's not *that* horrible [20:47:33] <^demon|away> Yeah you can. [20:47:34] Mhmm... [20:47:38] Spl has some. [20:47:40] <^demon|away> It's really not that bad. [20:48:00] <^demon|away> Tim's got plenty of experience doing it too :) [20:48:04] New review: Reedy; "(no comment)" [mediawiki/extensions/ProofreadPage] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3464 [20:48:04] True [20:48:06] Change merged: Demon; [mediawiki/extensions/ProofreadPage] (master) - https://gerrit.wikimedia.org/r/3467 [20:48:07] Change merged: Demon; [mediawiki/extensions/ProofreadPage] (master) - https://gerrit.wikimedia.org/r/3465 [20:48:09] Change merged: Demon; [mediawiki/extensions/ProofreadPage] (master) - https://gerrit.wikimedia.org/r/3466 [20:48:10] Change merged: Reedy; [mediawiki/extensions/ProofreadPage] (master) - https://gerrit.wikimedia.org/r/3464 [20:48:12] Did you ever have to call any other function in the PHP codebase? [20:48:19] Were any of them documented? :) [20:48:27] 03(mod) Change Gerrit database schema to support utf8 - 10https://bugzilla.wikimedia.org/35455 +comment (10Mark A. Hershberger) [20:48:32] <^demon|away> Dunno, haven't dug far into it :) [20:49:03] <^demon|away> I imagine it's about as bad as most open source is. [20:49:12] Dantman: Hmm, yeah, with feature branch support you can probably build something that doesn't constrain git as much as gerrit does [20:49:25] ^demon|away: you're saying it like closed-source C code is any better [20:49:25] ^demon|away: It's worse. Nothing is documented. [20:49:37] RoanKattouw: no! They have a nice code search feature [20:49:45] If you have enough C++ and the PHP boilerplate is big enough you can probably just write the C and then have a PHP wrapper. [20:49:45] created but no file [20:50:13] Dantman: Still, though, I like the fact that Gerrit is also a Git repo manager with fine-grained permission support, I want that too and for that you need to wrap around git to possibly deny things [20:50:21] The whole refs/for/master fake branch thing is a bit evil yes [20:50:22] <^demon|away> I mean there's already git libraries in c/c++, so the extension should just be some glue between php and those. [20:50:24] ie: All C++ function names, then just one or so C++ files that expose the C++ as a PHP interface. [20:50:42] Yeah, if you're just exposing functions it's probably fine [20:50:47] Change abandoned: saper; "more to come" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3946 [20:50:49] To be fair the only PHP ext I've ever worked on is wmerrors [20:51:04] Which is an error handler that needs to call various functions from elsewhere in the PHP codebase [20:51:52] RoanKattouw: What permission support do you really want? I really prefer giving people complete access to their own public repo. [20:52:12] Dantman: Well I prefer not giving anyone except the review system itself push access to master :) [20:53:34] RoanKattouw: My thought is that absolutely no-one would have push access to the real repo. The review system would handle pulling things for review from other repos. The review system would do merges. etc... No-one would ever push anything to that repo. You'd only pull from it yourself. [20:53:38] <^demon|away> Ooh, another feature of 2.3 you'd have to write into any hypothetical review tool... [20:53:43] <^demon|away> Automatic submodule updates. [20:53:53] Dantman: You're saying it'd be a different remote altogether? [20:54:02] ^demon|away: Enlighten me [20:54:11] RoanKattouw: Mhmm... sounds about right [20:54:24] <^demon|away> http://code.google.com/p/gerrit/issues/detail?id=377 [20:54:49] <^demon|away> From the rel-notes: "Allow superprojects to subscribe to submodules updates" [20:55:07] <^demon|away> Basically, mediawiki/extensions.git is automagically updated when one of its submodules is pushed to. [20:55:16] hello can u help me? how i can hide a main page title without edit common.css? [20:55:17] ^demon|away: Also, *stab* in the direction of update-submodules.sh . It doesn't work with anonymous checkouts at all [20:55:18] 03(ASSIGNED) Gerrit: Preferences: change "reviewed" preference when visiting changes to unchecked - 10https://bugzilla.wikimedia.org/35592 (10T. Gries) [20:55:31] * Dantman never really wanted us to use submodules for extensions [20:55:36] <^demon|away> RoanKattouw: Yeah I know, but 2.3 fixes this so none of that will be necessary :) [20:55:58] <^demon|away> Dantman: It's the easiest way to clone all extensions and keep them in check. [20:56:02] <^demon|away> Other than writing your own scripts. [20:56:32] It's practially nothing but a script wrapper itself. One that can't be tweaked to do what we want' [20:56:40] ^demon|away: Does that mean that in 2.3 if I git clone extensions.git I will automagically get all extensions, and git pull will Just Work? [20:56:46] Anyone?) [20:56:53] <^demon|away> RoanKattouw: git submodule update, but yes [20:57:02] OK but still [20:57:05] <^demon|away> Yes. [20:57:08] Iniquity: The proper way to do that is by editing css. [20:57:13] If it automatically puts the submodules in place when I clone, I'd be very happy with that already [20:58:03] <^demon|away> Anyway, I'll be testing that functionality on gerrit-de [20:58:06] <^demon|away> v [20:58:17] Dantman, but this code will be load all pages in my wiki [20:58:20] it is bad [20:58:47] Iniquity: Css is always loaded on all pages... there is nothing bad about that... for page targeting we have body.page-Main_Page [20:59:10] problem is size [20:59:15] New patchset: Reedy; "follow-up r111350 . check if optional parameter isset and is_numeric, otherwise limit to the built-in default (30000)" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3391 [20:59:30] Iniquity: It's a few pithy bytes... we load more unused css than that in a skin. [20:59:34] New review: Reedy; "(no comment)" [mediawiki/extensions/RSS] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3391 [20:59:41] every page will increase on 88 bytes [20:59:45] New review: Reedy; "(no comment)" [mediawiki/extensions/RSS] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3391 [20:59:59] it is no ergonomic [21:00:02] New review: Reedy; "(no comment)" [mediawiki/extensions/RSS] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3391 [21:00:02] Iniquity: Why are you trying to hide hundreds of pages individually? [21:01:02] Dantman, Only one, mane page. [21:01:27] Because all wikiprojects have problem with this. [21:02:21] ...I still don't see what's actually wrong besides priorities. [21:02:21] If i have article (83b) , i dont want to have code witch will be bigger than article. [21:02:49] Give up. You're already loading css that's larger than that. [21:02:56] *unnecessary code [21:03:08] Your priorities are screwed up. [21:04:07] i dont think so)) [21:04:13] New patchset: Reedy; "fix for bug30377 : add a new parameter to limit the number of characters when rendering the channel item " [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3390 [21:04:24] css is not embedded into the page. It's part of a completely separate request. And to top it off it's cached. [21:04:42] New review: Reedy; "(no comment)" [mediawiki/extensions/RSS] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3390 [21:04:45] <^demon|away> Iniquity: I think Dantman could be a bit nicer, but he's got a point. You're micro-optimizing for something that's really not that big of a deal. With caching, you're talking about a very minimal change here. [21:07:27] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3933 [21:07:29] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3933 [21:07:36] We have a code that should be used only on 1 page. This code is used in numerous projects. It is strange that he is still entered in the commons [21:08:05] Iniquity: That's basically how css works. [21:08:41] It gets loaded by the browser from one place. And cached once. Then everything uses the same cached stylesheet. [21:08:45] But we have MediaWiki:Mainpage-nstab [21:08:53] That's not css. [21:09:06] That's part of the html content separate for each page. [21:09:15] oh, yes sorry) [21:10:26] Question: Can it possible to make another "append" css for mp? [21:10:26] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3928 [21:10:51] New review: Reedy; "(no comment)" [mediawiki/extensions/RSS] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3398 [21:11:05] Iniquity: You asking about adding css from an article itself? [21:11:19] 03raindrift * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ (3 files): [21:11:19] added gradients and shadow for the menu [21:11:19] moved a property into its correct file [21:11:29] RoanKattouw: as you made that squashed commit for RSS, can we abandon the rest? It's really a dependency clusterfuck now [21:11:29] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3928 [21:11:33] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3945 [21:11:37] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3944 [21:11:41] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3943 [21:11:45] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3942 [21:11:49] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3941 [21:11:53] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3940 [21:11:57] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3939 [21:12:01] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3938 [21:12:05] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3937 [21:12:07] Dantman, not necessarily. [21:12:08] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3936 [21:12:12] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3935 [21:12:15] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3934 [21:12:17] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3928 [21:12:18] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3945 [21:12:19] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3944 [21:12:21] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3943 [21:12:22] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3942 [21:12:23] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3941 [21:12:24] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3940 [21:12:26] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3939 [21:12:27] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3938 [21:12:28] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3937 [21:12:29] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3936 [21:12:30] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3935 [21:12:31] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3934 [21:12:36] * Reedy looks at awjr [21:12:53] * awjr waves at Reedy [21:13:01] experimenting with git/gerrit scripting :) [21:13:05] heh :) [21:13:36] Reedy: Yes please. I meant to and then forgot halfway through discovering you can't mass-abandon things in Gerrit [21:13:46] You can't mass ANYTHING in gerrit [21:13:57] Reedy, RoanKattouw but you can from the command line! [21:14:00] Another feature for my hypothetical CR system! [21:14:08] Yup. [21:14:16] Dantman, may be i can edit file on server? [21:14:19] (although it's kind of a no-brainer) [21:14:19] 03kaldari * 10/trunk/extensions/PageTriage/ (5 files in 2 dirs): adding triage buttons [21:14:28] When you list stuff, add checkboxes. You just know you're gonna need them later [21:14:30] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3402 [21:14:33] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3403 [21:14:36] Iniquity: what were you trying to do in this case? [21:14:37] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3388 [21:14:41] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3389 [21:14:44] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3394 [21:14:49] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3396 [21:14:52] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3397 [21:14:58] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3399 [21:15:00] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3400 [21:15:04] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3404 [21:15:07] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3405 [21:15:11] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3406 [21:15:16] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3407 [21:15:17] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3401 [21:15:20] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3392 [21:15:25] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3393 [21:15:28] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3395 [21:15:29] RoanKattouw: ;) We can also make sure it understands notifications enough to group some types of notifications like irc messages into a single line. [21:15:31] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3390 [21:15:31] Reedy, RoanKattouw tomorrow during my 20% time i'm going to try to slap together a quick script to take different gerrit actions based on the output of gerrit query, which should make mass-changes trivial [21:15:33] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3391 [21:15:36] Change abandoned: Reedy; "Replaced by squash commit I2d972431" [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3398 [21:15:37] * Reedy whistles [21:16:00] Dantman, delete text from commons and entered it to main page. [21:16:38] 03(mod) Optionally exclude subpages from Special:Allpages and Special:Prefixindex - 10https://bugzilla.wikimedia.org/7184 +comment (10Max Semenik) [21:17:41] Iniquity: That still sounds like the same unnecessary micro-optimization. [21:17:49] switch (true) {}/// [21:17:51] I see.. [21:18:03] we should really just build an entire cli for gerrit [21:18:30] Dantman, macro-optimization, micro-optimization it is optimization [21:19:30] Iniquity: No... micro optimization is optimization with neglegable benefits which often creates piles of unnecessary extra work to gain almost no benefit to performance. [21:19:45] New patchset: Reedy; "Squashed commit of all RSS changes ported from SVN." [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3925 [21:20:45] Dantman, so, why mediawiki create MediaWiki:Mainpage-nstab? It was micro-optimization [21:21:00] Iniquity: No... that has absolutely nothing to do with micro-optimization. [21:21:08] oO [21:21:29] Why? [21:21:42] Reedy: ahah! that RSS squash must be a huge one :-D [21:21:53] That outputs html into the page. It's part of i18n. [21:22:00] Reedy: I started reviewing that mess and eventually gave up :-( [21:22:24] We had code in common, and all looked good. [21:22:32] 03(mod) A button on the page to toggle hidden categories - 10https://bugzilla.wikimedia.org/21119 -need-review -patch ; +comment (10Max Semenik) [21:22:38] have a good night everyone [21:22:47] New patchset: Reedy; "Squashed commit of all RSS changes ported from SVN." [mediawiki/extensions/RSS] (master) - https://gerrit.wikimedia.org/r/3925 [21:23:20] That's part of a feature letting you override the i18n text used on the mainpage for a namespace tab, not optimization. [21:24:26] en.wp wanted the namespace tab text on the main page to be "Main Page" instead of "Page". So that was added. [21:26:10] And what's the difference? [21:26:21] CSS is not i18n. CSS is not HTML. [21:26:23] it is micro [21:26:30] No it's not. [21:26:40] Ok) [21:26:44] It has absolutely nothing to do with optimization. [21:26:48] So, can i do it? [21:26:51] or not? [21:27:31] Moving css from Common.css to a page... no. [21:27:44] 03(mod) Links to file redirect pages do not have the mw-redirect class - 10https://bugzilla.wikimedia.org/23204 +comment (10Max Semenik) [21:27:45] Did someone fixed bug 9526? [21:27:45] it is pity( [21:27:47] I wasn't able to reproduce it on https://en.wikipedia.org/?oldid=484602370 [21:27:56] !b 9526 [21:27:56] https://bugzilla.wikimedia.org/show_bug.cgi?id=9526 [21:27:59] thanks, chao! [21:29:10] 03kaldari * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ext.pageTriage.listItem.css: fixing the wrapping of the snippets [21:29:24] New review: Reedy; "(no comment)" [mediawiki/extensions/WikiLove] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3838 [21:29:26] Reedy: if that matters, I tested on Google Chrome 18 [21:29:36] RoanKattouw: I know ops hates ruby, but we should see how much of this intersects with Gitorious. Writing a review system and using Gitorious' already written source code to handle repository management would be a plus. [21:29:57] s/be a plus/avoid needing to reimplement a lot of basics from scratch/ [21:30:01] Oh great [21:30:17] C, PHP, Python, OCaml, and now Ruby [21:30:24] lol [21:30:35] 03kaldari * 10/trunk/extensions/PageTriage/SpecialPageTriage.php: making triage buttons actually link to articles [21:31:14] vvv: ;) You're forgetting the fact we already have Java and Ruby. [21:31:24] I hate Ruby from a dev perspective *and* an ops perspective [21:31:24] Wait, where? [21:31:36] I have to say, I'm quite impressed that gitorious runs Ruby in production [21:31:37] [01:30:43] ChanServ [ChanServ@services.]: You have been devoiced on #mediawiki by c (o_O) [21:31:41] vvv: Gerrit = Java; Puppet = Ruby; [21:31:57] Change abandoned: Reedy; "Abandoning in favour of I9c0fd70c" [mediawiki/extensions/Narayam] (master) - https://gerrit.wikimedia.org/r/3753 [21:32:16] We need Tcl ASAP [21:33:33] And by the way, we should become more gnuified [21:33:44] And use GNU scripting solution as much as possible [21:33:50] (which is Scheme) [21:34:03] Lisp! [21:34:25] Everybody loves Lisp [21:34:39] We have too few purely functional PLs running on our servers [21:35:49] helderwiki: comment on the bug as such. You can close it was resolved worksforme and someone can reopen it if necessary [21:35:56] New review: awjrichards; "These changes are fine for now - I dig the makefile. The 'mobile_action' stuff will need to change s..." [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3924 [21:36:20] Reedy: ok, thanks. were you able to reproduce it? [21:36:39] I'm writing a giant email about QA for wikitech-l --- what a glamorous life I lead. [21:37:06] heh [21:37:19] I need to write one about direct push vs liberal +2 for feature branches [21:38:22] New patchset: Reedy; "recommiting unreviewed r113321 change from SVN" [mediawiki/extensions/WikiLove] (master) - https://gerrit.wikimedia.org/r/3831 [21:38:26] RoanKattouw: You think we should write all the repo management from scratch? [21:38:48] Hmm probably not [21:39:13] You could borrow liberally from Gerrit or gitorious [21:39:40] New patchset: awjrichards; "add mf-feedback.js file for feedback page" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3924 [21:39:44] New review: Reedy; "(no comment)" [mediawiki/extensions/WikiLove] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3831 [21:39:48] Change merged: Reedy; [mediawiki/extensions/WikiLove] (master) - https://gerrit.wikimedia.org/r/3831 [21:39:59] New review: awjrichards; "(no comment)" [mediawiki/extensions/MobileFrontend] (contact-us-redesign); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3924 [21:40:01] Change merged: awjrichards; [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3924 [21:40:11] RoanKattouw: Well... Gitorious is better than gerrit in this regard... so that would practically mean running Gitorious. [21:41:14] Hi. [21:41:46] 03(mod) CSS properties in wikitext can be abused to spam and make it difficult to remove them - 10https://bugzilla.wikimedia.org/9526 (10mybugs.mail) [21:41:55] Maybe write some drafts... [21:42:09] Try to find out all the things we'd want to do. [21:42:24] Find out what implements them. And how hard it would be to write from scratch. [21:42:54] Then finally figure out whether it's easy enough to write from scratch or we would want to still run something like Gitorious. [21:44:26] 03(mod) replacing "hidden by Article Feedback V5" link with the actual account - 10https://bugzilla.wikimedia.org/35501 +comment (10Fabrice Florin) [21:44:55] 03(mod) replacing "hidden by Article Feedback V5" link with the actual account - 10https://bugzilla.wikimedia.org/35501 15enhancement->normal (10Fabrice Florin) [21:47:04] New patchset: Trevor Parscal; "Replaced es with ce for Sandbox and main demo" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3947 [21:47:32] Hi, I am fighting right now to get a botched upgrade back on track - is this a good place to ask some questions? :-/ [21:48:04] RoanKattouw: thank you re http://opensourcebridge.org/proposals/877 [21:48:16] alolita1: ping ping ping http://opensourcebridge.org/events/2012/proposals :-) [21:48:48] sumanah: thanks for the poke!!! will complete this now :-) [21:48:50] hexmode: around? [21:49:01] i am [21:49:22] good [21:49:29] I think I have found this pgsql transaction bug [21:49:48] thank you alolita1 [21:49:54] ok, off to a community manager meeting [21:50:08] i updated my wiki to 1.8 a little while ago. Everything was running great, i added some extensions I used, still run great. Went to lunch, and came back to this: http://wiki.embeddedethernet.com/ Any ideas? Its like the page is not being parsed properly. I made no changes, its like a maintenence script ran and broke something [21:50:12] sumanah: yw [21:50:33] hexmode: care to test? :) [21:50:45] embeddedethernet: 1.18 you mean? [21:50:56] yes :) the latest 1.18.x [21:50:58] embeddedethernet: It's giving 404s for http://wiki.embeddedethernet.com/load.php?debug=false&lang=en&modules=mediawiki.legacy.commonPrint%2Cshared%7Cskins.vector&only=styles&skin=vector&* [21:51:11] saper: sure, I can try... you have a commit? [21:51:56] embeddedethernet: Your need to update your rewrite rules. http://wiki.embeddedethernet.com/load.php is interpreted as a request for the [[load.php]] page. You need to add load.php to a list of exceptions you presumably have somewhere, including index.php and api.php [21:52:05] BTW, this is why we recommend against using the URL scheme that you chose [21:52:33] hexmode: in a moment [21:52:34] Has anyone here an idea why a media wiki user can get lost during an update? [21:52:45] I updated mediawiki from 1.16.5 to 1.18.2. [21:52:48] it might be 27283 and maybe even 35358 [21:52:56] You definitely should not lose users in an upgrade [21:53:01] The user is still in the (sqlite) db, but mediawiki tells me it does not exist. [21:53:19] wl: What's the username? [21:53:21] When I look into the DB, I still see all users, so it is not really lost... [21:53:24] Maybe it became invalid? [21:53:25] saper: url? [21:53:38] ALL users from the previous install have gone out of the window (!) [21:53:48] New review: awjrichards; "btw i ignored this for the merge of this changeset since it's going to get changed in just a minute ..." [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3924 [21:54:02] Thanks roan, i'll look in to that. I was updating from a dreamhost 1 click install managed wiki to a tarball install directly from mediawiki [21:54:04] wl, anything in `user` DB table? [21:54:16] yes. All old users are there. [21:54:35] I DID make a backup (one of the DB, and also an XML dump). [21:55:34] that is users are in the DB, but MW doesn't see them? [21:55:51] New patchset: awjrichards; "Follow up to Change-Id: Ia065e1f2, fixing static call to non-static method" [mediawiki/extensions/MobileFrontend] (contact-us-redesign) - https://gerrit.wikimedia.org/r/3948 [21:55:56] wrong DB? [21:56:15] well, the content is visible. I inherited the install :-/ [21:56:18] hexmode: committing in a moment, just smoke testng [21:56:44] New review: Platonides; "How is this different than the old code?" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3725 [21:56:57] are they visible on Special:Listusers? [21:57:26] 03(NEW) Add keyboard shortcuts - 10https://bugzilla.wikimedia.org/35596 enhancement; Analytics: ReportCard; (dvanliere) [21:57:38] No. They are not visible on this page, only the new user I had to recreate is here. This is a bureaucrat & administrator, so I could recreate them, but why? [21:58:17] Are you using a table prefix? [21:58:25] New review: Trevor Parscal; "This code needs to eventually be moved to the client because transmitting the linear model over the ..." [mediawiki/extensions/VisualEditor] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3931 [21:58:29] wl, did that installation start with 1.16? [21:58:34] 03(NEW) Add read-only mode - 10https://bugzilla.wikimedia.org/35597 normal; Analytics: ReportCard; (dvanliere) [21:58:39] 03(mod) Add read-only mode - 10https://bugzilla.wikimedia.org/35597 normal->15enhancement (10Diederik van Liere) [21:59:20] RoanKattouw, prefix doesn't quite explain why other tables such as page are intact [22:00:44] @MaxSem - yes, the 1st installed version was 1.16. I found problems with webapp-configs, so I tried to do a full backup, cleanup the old install and reinstall it. [22:00:56] well, brought it on myself, really... [22:01:08] I did not find a table prefix in the settings [22:01:19] $wgDBprefix [22:01:27] Can someone tell me what I need to do to fix "Change cannot be merged due to unsatisfiable dependencies." ? [22:02:37] no prefix in LocalSettings, no. [22:02:44] I [22:03:00] I'm running a sqlite DB, btw, not sure if this matters [22:03:24] yeah, you already mentioned this [22:03:41] lemme try a quick repro [22:04:10] New patchset: Platonides; "Unneeded globals after e2ee009e5 (aka. r114233)" [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3949 [22:05:07] New review: Aaron Schulz; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3949 [22:05:10] Change merged: Aaron Schulz; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3949 [22:07:03] wl: Is LocalSettings.php pointing to SQLite or to some other DB backend? Is it pointing to the same database? [22:08:28] is $wgSQLiteDataDir absolute or relative? [22:08:35] New review: Catrope; "Looks good except for some minor things" [mediawiki/extensions/VisualEditor] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/3947 [22:08:41] in the latter case, it shouldn't be [22:11:08] 03raindrift * 10/trunk/extensions/PageTriage/ (5 files in 2 dirs): [22:11:08] added the filter form, not hooked up to anything yet. [22:11:08] forgot to checkin the parts of the gradient in SpecialPageTriage... oops. [22:11:40] @Max: I just found something very strange here, wait a minute... [22:11:59] New patchset: Amire80; "(bug 25475) Replacing table with div and fixing directionality." [mediawiki/extensions/DismissableSiteNotice] (master) - https://gerrit.wikimedia.org/r/3951 [22:12:58] New patchset: Reedy; "adding first pass at SVG sprites for heart icon" [mediawiki/extensions/WikiLove] (master) - https://gerrit.wikimedia.org/r/3838 [22:13:15] 03(ASSIGNED) Use div instead of table for the layout of the sitenotice - 10https://bugzilla.wikimedia.org/25475 -need-review -patch ; +comment (10Amir E. Aharoni) [22:13:23] New review: Reedy; "(no comment)" [mediawiki/extensions/WikiLove] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3838 [22:13:26] Change merged: Reedy; [mediawiki/extensions/WikiLove] (master) - https://gerrit.wikimedia.org/r/3838 [22:13:40] *augh* webapp-config seemed to decide to just move the datadir around. [22:13:57] I re-imported the dump into a new sqlite DB, the users are in the old dB. [22:14:24] hmm [22:14:36] git chckout new branch, fetch revision, git rebase master, git review [22:15:50] @MaxSem: Thank you, the users are there again! That was the bug. [22:16:00] OK, and I still had to run update.php first, but ok... [22:16:14] so, what was the bug? [22:16:44] 14(DUP) Make sitenotice table 100% wide rather than 80% - 10https://bugzilla.wikimedia.org/24066 +comment (10Amir E. Aharoni) [22:16:44] 03(mod) Use div instead of table for the layout of the sitenotice - 10https://bugzilla.wikimedia.org/25475 +comment (10Amir E. Aharoni) [22:16:56] RoanKattouw: Thanks for the tip on the rewrite rule. I added load and api, and added the short url fix, and its all cleaned up now. [22:16:57] webapp-config apparently set this DB path to ANOTHER directory. A new DB was created - without content. [22:17:32] New review: Inez; "(no comment)" [mediawiki/extensions/VisualEditor] (master); V: 0 C: 2; - https://gerrit.wikimedia.org/r/3947 [22:17:33] OK, I re-imported it with importDump, but the users did not get across this dump, it seems... [22:18:34] ^demon|away: aren't drafts (refs/draft/xxxx) already available in our gerrit? (There is My > Drafts). Maybe we just need permissions set? [22:18:36] yes, users aren't saved in XML dump [22:19:24] Well, I also had a DB dump - maybe just reimporting THAT would have helped, but the main problem was the path change. [22:19:41] OK, thank you for your patience..... [22:20:15] 04(REOPENED) CSS properties in wikitext can be abused to spam and make it difficult to remove them - 10https://bugzilla.wikimedia.org/9526 +comment (10Platonides) [22:22:58] Is it possible to somehow grep all my wiki pages ? [22:23:55] diverdude, umm, search them? [22:26:02] MaxSem: yes...if i know i have written something somewhere....but i dont know where [22:26:37] there's a search bar;) [22:26:57] New patchset: Trevor Parscal; "Resolved issues in change If8d13bf7011616d222be78899b23186859d5ed70" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3953 [22:27:47] <^demon|away> saper: Drafts exist in 2.2.x, but the syntax for pushing them is awkward iirc (I can't even remember it off the top of my head). 2.3 just makes it easier to use. [22:29:02] MaxSem: and it searches all pages? [22:29:10] 03raindrift * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ext.pageTriage.listControlNav.js: menu closes when clicked outside [22:29:16] lol [22:33:30] 03(mod) Make AFTv5 url parameters work all the time - 10https://bugzilla.wikimedia.org/35587 +aftv5-1.5 (10Fabrice Florin) [22:33:44] New review: Trevor Parscal; "I7c20de33 has the fixes needed for this to be merge ready." [mediawiki/extensions/VisualEditor] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3947 [22:34:34] New patchset: Catrope; "Replaced es with ce for Sandbox and main demo" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3947 [22:34:58] ^demon|away: I thought it's just "git push gerrit HEAD:for/draft/master", not? [22:36:24] <^demon|away> saper: Maybe. Dunno. [22:36:57] Hi RoanKattouw, I have a few extensions I would like to have migrated over from svn [22:37:03] Do I put it here: http://www.mediawiki.org/wiki/Git/New_repositories/Requests [22:37:12] i am not in a hurry [22:37:15] Yes [22:37:20] ok [22:37:21] thanks [22:38:44] * AaronSchulz wtfs at bug 35578 [22:38:47] New review: Catrope; "Introduces excess whitespace, looks good otherwise" [mediawiki/extensions/VisualEditor] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/3947 [22:44:59] New patchset: Trevor Parscal; "Replaced es with ce for Sandbox and main demo" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3947 [22:46:38] New review: Catrope; "(no comment)" [mediawiki/extensions/VisualEditor] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3947 [22:46:41] Change merged: Catrope; [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3947 [22:47:12] New review: Trevor Parscal; "Fixed whitespace issue in patch set 3" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3947 [22:50:45] Change abandoned: Catrope; "Squashed into 3947" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3953 [22:51:58] 03(mod) IPs hidden without any log or apparent reason - 10https://bugzilla.wikimedia.org/35578 +comment (10Snowolf) [22:55:04] 03raindrift * 10/trunk/extensions/PageTriage/modules/ext.pageTriage.views/ext.pageTriage.listControlNav.js: made resize behavior slightly better [22:55:21] 03(NEW) Article Feedback v5 - Update Notices for Feedback Page Actions - 10https://bugzilla.wikimedia.org/35599 normal; MediaWiki extensions: ArticleFeedbackv5; (fflorin) [22:55:48] 03(mod) Article Feedback v5 - Update Notices for Feedback Page Actions - 10https://bugzilla.wikimedia.org/35599 +aftv5-1.5 (10Fabrice Florin) [22:56:29] New patchset: Trevor Parscal; "Added ve.Surface, a wrapper around an editor instance" [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3955 [22:58:46] 03(mod) Some IPs hidden without any log or apparent reason on frwiki - 10https://bugzilla.wikimedia.org/35578 summary (10Snowolf) [23:06:00] 03(mod) Don't surface fix me templates on Mobile Frontend - 10https://bugzilla.wikimedia.org/30421 +comment (10MZMcBride) [23:06:35] New review: Robmoen; "(no comment)" [mediawiki/extensions/VisualEditor] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3955 [23:07:13] 03(mod) Make git-review easier to install on Windows - 10https://bugzilla.wikimedia.org/35145 +comment (10Siddhartha Ghai) [23:07:16] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3471 [23:07:19] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3471 [23:07:24] New review: Reedy; "(no comment)" [mediawiki/extensions/UploadWizard] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3472 [23:07:26] Change merged: Robmoen; [mediawiki/extensions/VisualEditor] (master) - https://gerrit.wikimedia.org/r/3955 [23:07:27] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3473 [23:07:28] Change merged: Reedy; [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3472 [23:07:54] Reedy: how do you commit just one branch via git-review? [23:07:58] Reedy: https://bugzilla.wikimedia.org/35145 comment is for u [23:08:22] AaronSchulz: I ended up doing https://www.mediawiki.org/wiki/Git/Workflow#Committing_to_non_master before [23:08:28] is there any way to define key, value pairs that can be transcluded like templates without needing to hack up PHP files as described here?: http://www.mediawiki.org/wiki/Manual:Variable [23:08:45] Didn't someone commit gitreview for branches? :/ [23:09:00] I'm considering just defining one template per constant, but I'd ideally like to do this all in one place [23:09:06] "You have more than one commit that you are about to submit." [23:09:17] Reedy: which there was more DWIM ;) [23:09:29] oh [23:09:30] lols [23:09:59] * AaronSchulz splits things into branches for a reason [23:10:17] there must be some magic parameter flag [23:12:24] New patchset: Reedy; "Fix up all the trailing whitespace which is irritating me" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3957 [23:12:54] I have a personal wiki setup, and often when I click search or click on a link it will start downloading a .GZ file rather than taking me to the page. Anyone know what the issue is? [23:16:30] New patchset: Reedy; "Added HTML5 drag and drop support" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3808 [23:17:26] 04(REOPENED) MobileFrontend extension outputting invalid HTML in footer - 10https://bugzilla.wikimedia.org/30406 +comment (10MZMcBride) [23:20:59] New patchset: Reedy; "Fix more trailing whitespace" [mediawiki/extensions/UploadWizard] (master) - https://gerrit.wikimedia.org/r/3958 [23:23:46] So what's the viewvc equivalent now? [23:23:54] !viewvc [23:23:54] See [23:24:23] gitweb [23:24:41] !gitweb [23:24:41] I don't know anything about "gitweb". [23:24:45] * Reedy kicks mw-bot [23:28:53] New review: Robmoen; "(no comment)" [mediawiki/extensions/UploadWizard] (master) C: 1; - https://gerrit.wikimedia.org/r/3808 [23:32:05] !r 109926 [23:32:05] https://www.mediawiki.org/wiki/Special:Code/MediaWiki/`e1 [23:32:10] !r 109926 [23:32:10] https://www.mediawiki.org/wiki/Special:Code/MediaWiki/109926 [23:32:11] New patchset: Aaron Schulz; "Added user Id sanity check to setUsernameBitfields() for bug 34755." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3959 [23:37:15] New patchset: MaxSem; "action=mobileview improvements * Optionally follow redirects * Optionally return normalized page title" [mediawiki/extensions/MobileFrontend] (master) - https://gerrit.wikimedia.org/r/3960 [23:37:26] 03(NEW) Enable hook to block addition of trailing whitespace - 10https://bugzilla.wikimedia.org/35600 enhancement; Wikimedia: Git/Gerrit; (sam) [23:37:32] 03(mod) MobileFrontend extension outputting invalid HTML in footer - 10https://bugzilla.wikimedia.org/30406 +comment (10MZMcBride) [23:37:34] Reedy: ^ [23:37:42] Also, what's the veiwvc equivalent? [23:37:50] gitweb [23:37:54] wherever it is [23:38:00] It's non obvious (helpfully) [23:38:13] New review: Tim Starling; "(no comment)" [mediawiki/core] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3959 [23:38:16] Change merged: Tim Starling; [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3959 [23:38:32] https://gerrit.wikimedia.org/r/gitweb?p=mediawiki/core.git [23:38:43] !gitweb is For core: https://gerrit.wikimedia.org/r/gitweb?p=mediawiki/core.git [23:38:43] Successfully added keyword: gitweb [23:38:43] !gitweb is https://gerrit.wikimedia.org/r/gitweb?p=mediawiki/core.git [23:38:44] Keyword "gitweb" already exists. Use "!gitweb del" to remove it. [23:38:49] I win. [23:38:52] Thank you, Reedy! [23:39:42] https://gerrit.wikimedia.org/r/gitweb?p=mediawiki/extensions/WikiLove.git [23:40:43] Joan: what you were pointing at? [23:40:51] *what were you [23:41:01] Reedy: I thought you might find https://bugzilla.wikimedia.org/show_bug.cgi?id=30406#c16 interesting. [23:41:11] ah [23:41:14] hackhackhack [23:47:50] 03aaron * 10/branches/wmf/1.19wmf1/includes/revisiondelete/RevisionDeleteUser.php: MFG Change-Id: I44288d199d86045151f8184f7abaa3df886d9d99 [23:51:22] New review: Reedy; "How is it possible to actually test the functionality of this commit?" [mediawiki/extensions/UploadWizard] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/3808 [23:54:45] Change abandoned: Reedy; "Was done and merged in Ia83ee414" [mediawiki/extensions/WikiLove] (master) - https://gerrit.wikimedia.org/r/3474 [23:57:56] New review: Daniel Friesen; "$expiration is a &reference so the change I made removing the `$expiration = ...;` made a regression..." [mediawiki/core] (master) - https://gerrit.wikimedia.org/r/3725 [23:59:28] New patchset: Reedy; "Followup-To: r113793 Update RELEASE-NOTES for Extension:Nuke version 1.1.5" [mediawiki/extensions/Nuke] (master) - https://gerrit.wikimedia.org/r/3458 [23:59:58] New review: Reedy; "(no comment)" [mediawiki/extensions/Nuke] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/3458