[00:03:02] (03PS1) 10Tim Starling: Resubmit "Add support for mysqli extension" [core] - 10https://gerrit.wikimedia.org/r/90269 [00:04:36] (03PS3) 10Reedy: Vector: Fix font size regression [core] - 10https://gerrit.wikimedia.org/r/90134 (owner: 10Jdlrobson) [00:04:43] (03CR) 10Reedy: [C: 032] "WFM" [core] - 10https://gerrit.wikimedia.org/r/90134 (owner: 10Jdlrobson) [00:07:51] (03CR) 10Swalling: "Is that the cause of https://en.wikipedia.org/wiki/Wikipedia:Village_pump_(technical)#Wikipdia_font_change.3F ?" [core] - 10https://gerrit.wikimedia.org/r/79948 (owner: 10Jdlrobson) [00:11:32] (03CR) 10Legoktm: "(1 comment)" [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/88223 (owner: 10Legoktm) [00:11:38] (03PS3) 10Legoktm: Only allow Thanks for public revisions [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/88223 [00:12:45] (03Merged) 10jenkins-bot: Vector: Fix font size regression [core] - 10https://gerrit.wikimedia.org/r/90134 (owner: 10Jdlrobson) [00:19:35] (03CR) 10MarkTraceur: [C: 04-1] "Need to fallback and link to file page. Maybe rebase on top of Icfab093a44530949d1672a683fb8d4c0e7b901fc" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [00:24:54] (03CR) 10Jdlrobson: "Shouldn't be. Was only merged on Tuesday and fix merged today." [core] - 10https://gerrit.wikimedia.org/r/79948 (owner: 10Jdlrobson) [00:32:41] (03PS1) 10Reedy: Revert "Add new recentchanges field rc_source to replace rc_type" [core] - 10https://gerrit.wikimedia.org/r/90273 [00:34:49] (03CR) 10Aaron Schulz: [C: 032] "Not live yet is blocker indeed" [core] - 10https://gerrit.wikimedia.org/r/90273 (owner: 10Reedy) [00:34:50] (03CR) 10Reedy: "This should not have been merged with unconditional usage of a new field." [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson) [00:37:24] (03PS2) 10Hiong3-eng5: added check mode and corrections API add_syntrans [extensions/WikiLexicalData] - 10https://gerrit.wikimedia.org/r/90080 [00:37:46] (03Merged) 10jenkins-bot: Revert "Add new recentchanges field rc_source to replace rc_type" [core] - 10https://gerrit.wikimedia.org/r/90273 (owner: 10Reedy) [00:40:36] (03CR) 10Aude: "certainly a setting should be added that enables use of the new field." [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson) [00:41:50] (03CR) 10Aude: "or if the schema change can be deployed first before the patch is re-merged, that works too." [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson) [00:43:13] At least it should be OSC-able [00:43:23] 6.65M rows on enwiki [00:43:39] what's "OSC-able"? [00:43:56] online schema change [00:44:11] thanks [00:44:12] recentchanges is PK'd [00:44:59] So makes springle not want to stab people so much when they change tables [00:45:00] :D [00:45:20] Hi everyone [00:45:45] Reedy: this one is not so horrible [00:46:00] at least it gets pruned regularly :) [00:46:05] indeed [00:46:17] Wondering if anyone has a few minutes to help me get going on some bugs in the EP extension [00:46:19] I feel sorry for translatewiki though :( [00:46:30] with their recentchanges back to the stoneage [00:46:43] oh really? [00:47:03] IIRC yeah [00:47:18] AndyRussG: Better to ask more specific questions ;) [00:47:39] Yes I have 'em, one sec [00:47:59] Here's what I'm starting on https://bugzilla.wikimedia.org/show_bug.cgi?id=55568 [00:48:17] (03CR) 10Aaron Schulz: "The schema change should just be done first. That was just forget this time (which is bound to happen, such things should be -2ed until th" [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson) [00:48:19] It happens that there are a few DB-related bugs and I'm guessing that some of them are related [00:48:46] https://bugzilla.wikimedia.org/show_bug.cgi?id=44561 [00:48:54] https://bugzilla.wikimedia.org/show_bug.cgi?id=43782 [00:49:02] https://bugzilla.wikimedia.org/show_bug.cgi?id=44110 [00:49:10] https://bugzilla.wikimedia.org/show_bug.cgi?id=44365 [00:49:17] https://bugzilla.wikimedia.org/show_bug.cgi?id=47439 [00:49:36] https://bugzilla.wikimedia.org/show_bug.cgi?id=40433 [00:51:18] My theory is that somewhere the DB is getting corrupted and this has unpredicted effects, that are not related in any obvious way to the code where the real bugs are [00:51:50] Yay, abstraction [00:52:15] If this is right, then I think the first thing I need to do is get a look at the tables that the EP is using and check what, if anything, may be wonky about them [00:52:44] Are they replicated to labs and/or toolserver? [00:53:01] So I was hoping to get some input as to where this is the right way to go, and how to go about getting to the actual DB tables with the production data [00:53:23] If it's replicated to labs it's much easier [00:53:39] If it's not, you'll need an RT ticket signed off by your manager and have to wait a couple of days [00:54:11] Ah OK [00:54:33] Coren: Any idea if the educationprogram tables are replicated to labs? [00:55:07] Reedy: What table names? We did try to replicate everything that didn't contain PII no matter how arcane. [00:55:21] Reedy: they are [00:55:28] They're all ep_ [00:55:29] Aha [00:55:37] AndyRussG: Have you got a labs/gerrit account setup yet? [00:56:02] Yeah, the ep_ tables are there [00:56:08] great :) [00:56:09] I have a gerrit account. Haven't gotten my feet wet with labs yet [00:56:14] Coren: fantastic [00:56:23] Have you added an SSH key? [00:56:47] yes [00:57:02] I think someone gave me shell access, too [00:58:10] sshing to bastion.wmflabs.org would be your first port of call [01:00:23] Hmmm, it doesn't like me--"Permission deined (publickey)" [01:01:09] However, "ssh AndyRussG@gerrit.wikimedia.org -p 29418" works, and I can do git review stuff [01:02:06] definitely in the bastion project [01:02:40] You don't look to have a home directory though [01:04:13] When I log in to the wikitech Web interface I do get a "Lab Users" category in the toolbar on the left [01:05:13] Hmmm [01:05:22] Do you need to add your SSH key in 2 seperate places still? [01:05:37] Have you got a public key listed at https://wikitech.wikimedia.org/wiki/Special:Preferences#mw-prefsection-openstack ? [01:05:47] Yes [01:06:48] Though actually it's a new one. I deleted my old one, but on gerrit I kept my old one, and my new one didn't work on gerrit. I'll try adding my old one on the Web interface again [01:08:30] well if your key didn't work on gerrit, that might be a hint it won't work elsewhere [01:08:40] did you make sure the key was done properly? [01:09:24] I'm pretty sure the new one worked elsewhere [01:10:03] Adding the old one, strange it says "failed to iport keypair" [01:11:00] and did you have a svn account in the pregit days? [01:11:09] Nope [01:11:15] although I don't think gitreview would work in that case [01:11:27] Just deleted my new key, re-added the old one, it seems to have worked [01:13:59] It's still all "Permission denied" when I ssh to bastion, but I don't know if there's a turnaround time for the key to become active [01:15:22] "not long" [01:16:17] AndyRussG: people in #wikimedia-labs might be able to help better [01:16:32] if not, just file a bug in bugzilla and let someone else look at it [01:17:06] OK [01:18:12] Any suggestions about the general approach to these bugs? And when I'm in labs, how do I look at the tables? (Or mabe I should ask that on labs too...) [01:21:44] OK, I'm in labs [01:22:30] I've no idea... [01:22:39] aude? :D [01:23:08] AndyRussG: are you in tool labs? [01:23:26] i am not sure how / if to access from bastion or other labs projects [01:23:57] heh [01:24:39] i did "ssh andyrussg@bastion.wmflabs.org" and am in a shell. That's all I know [01:25:06] https://wikitech.wikimedia.org/wiki/User:Magnus_Manske/Migrating_from_toolserver#Replicated_databases [01:25:11] sql enwiki_p as usual [01:25:38] That works from tools-login apparently [01:25:57] Which I don't have access to [01:26:36] ask Coren [01:26:48] I just created https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Access_Request/Reedy [01:26:49] * Reedy grins [01:27:16] Reedy: ha ha [01:27:36] Reedy: Granted. [01:27:39] (03PS1) 10Tim Starling: Fix mysqli support [core] - 10https://gerrit.wikimedia.org/r/90278 [01:27:47] $ sql enwiki_p [01:27:47] sql: command not found [01:27:52] And yes, to get access to the replicas you need to go on tools. :-) [01:27:57] AndyRussG: My point exactly [01:28:00] AndyRussG: ^^ [01:28:08] AndyRussG: I'll go give you access to tools now. What's your wikitech username? [01:28:09] https://wikitech.wikimedia.org/wiki/Special:FormEdit/Tools_Access_Request [01:28:48] I'm being asked for a password [01:29:03] (03CR) 10Tim Starling: "Bug fixes in dependent commit" [core] - 10https://gerrit.wikimedia.org/r/90269 (owner: 10Tim Starling) [01:30:26] Reedy: ? Where and when? [01:30:43] sql enwiki_p [01:30:43] Enter password: [01:30:43] ERROR 1045 (28000): Access denied for user 'reedy'@'10.4.0.220' (using password: YES) [01:30:43] reedy@tools-login:~$ [01:30:53] hm. we should remove pam_password from sudo's pam config [01:31:08] https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Access_Request/AndyRussG [01:31:08] then if the sudo policy is denied, they just get denied [01:31:17] Reedy: It takes a few minutes before your SQL credentials get in. [01:31:45] AndyRussG: You're in. [01:31:55] Thank you! [01:32:11] Reedy: I see your creds got in your home now. Try again? [01:32:30] Yup, thanks [01:32:39] hell, in fact, we should remove pam_unix and pam_ldap from common-auth [01:32:53] AndyRussG: So, from there, explain ep_orgs; etc [01:33:24] Ryan_Lane: Indeed. [01:33:43] there's absolutely no reason for the systems to ever use password auth [01:33:55] Ryan_Lane: Hey, who should I add to review an enwiki config change in mediawiki-config/wmf-config? [01:34:02] (I'm talking about labs, and not production, of course) [01:34:13] Coren: the devs [01:34:24] unless it's an ops related setting [01:34:46] Coren: Reedy normally does them [01:34:46] Ryan_Lane: New user group and permission on enwiki [01:34:51] devs [01:35:13] is there a bug associated with this? [01:35:20] is there community consensus for it on the wiki? [01:35:26] Reedy: By an extraordinary coincidence, you've just been added to https://gerrit.wikimedia.org/r/#/c/88196 :-) [01:35:29] (03PS2) 10Reedy: Resubmit "Add support for mysqli extension" [core] - 10https://gerrit.wikimedia.org/r/90269 (owner: 10Tim Starling) [01:35:30] Ryan_Lane: yes and yes. [01:35:33] (03CR) 10Reedy: [C: 032] Resubmit "Add support for mysqli extension" [core] - 10https://gerrit.wikimedia.org/r/90269 (owner: 10Tim Starling) [01:35:39] 55432 [01:35:49] in that case you could probably do it yourself... [01:36:07] if it's not controversial in any way [01:36:32] It's not controversial, I just want to make sure I got the config 100% right before I push something on enwiki. :-P [01:36:38] * Ryan_Lane nods [01:36:43] it's just another wiki ;) [01:37:08] ... yeah. I fully intend to make the adage "You're not ops until you break the site" obsolete. :-) [01:37:34] well, I'd say the labsdb snafu has already won you that privilege [01:37:43] ;) [01:37:59] ok. heading home [01:38:19] :-P [01:39:26] For checking out the tables, this is the right place to start? [01:39:26] https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Help#Database_access [01:39:53] (03CR) 10Springle: "Can I see an example of the queries that will use the new indexes, please?" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [01:40:01] sql enwiki_p is all you need [01:40:15] (03PS2) 10Tim Starling: Fix mysqli support [core] - 10https://gerrit.wikimedia.org/r/90278 [01:42:59] Fantastic, it's all "Welcome to the MariaDB monitor..." [01:43:01] (03CR) 10Aaron Schulz: "What would the full effect of prefixing be on sorting?" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [01:44:20] Reedy: Do we have documentation on how to push a mediawiki-config change to prod? [01:45:10] sync-file wmf-config/InitialiseSettings.php summary [01:45:19] after pulling/rebasing [01:45:50] From tin? [01:46:32] Yeah in /a/common [01:46:44] The instructions for deploying actual mediawiki stuff are vaguely right [01:47:02] Just different paths (ie not under php-1.XXwmfYY, but in wmf-config) [01:48:38] (03CR) 10Springle: "Go ahead and merge if it's safe. I'll begin the schema changes now." [core] - 10https://gerrit.wikimedia.org/r/51675 (owner: 10Reedy) [01:48:58] * Coren nods. [01:49:04] Seems straightforward enough. [01:49:30] Reedy: ^ you're sure nothing in the code will care if ar_id just appears? [01:51:34] Reedy, Coren, p858snake|l, aude, thank you so much for your help. [01:51:53] sure :) [01:51:58] "copying to apaches". How scary. :-) [01:52:00] springle: It shouldn't, there is no code usage of the fields, and it's not like it's a required field that mediawiki needs to fill in a value for when inserting [01:52:19] ok then :) [01:52:40] Yeay! I didn't break the wiki! :-) [01:52:47] Might be sane to wait until after I've branched 1.22wmf22 before merging it though (12 hours or so from now) [01:53:30] (03PS24) 10Reedy: Add archive, externallinks PK [core] - 10https://gerrit.wikimedia.org/r/51675 [01:54:28] (03CR) 10Reedy: "Would be worth adding a RELEASE-NOTES entry to go with this before merging it; which I shall do after I've branched 1.22wmf22 later today" [core] - 10https://gerrit.wikimedia.org/r/51675 (owner: 10Reedy) [01:54:35] (03CR) 10Jackmcbarn: "Springle: is https://gerrit.wikimedia.org/r/#/c/85436/1/includes/logging/LogPager.php what you're asking for?" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:01:22] (03CR) 10MZMcBride: "Jackmcbarn: Can you dump a few example full SQL queries somewhere?" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:09:16] (03CR) 10Springle: "Aaron: sorting is one reason I want to see the queries. If the new indexes do get chosen for sorting, a prefix index may need a final file" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:13:00] (03CR) 10Jackmcbarn: "Here's one:" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:20:09] (03CR) 10Reedy: "mysql:wikiadmin@db1056 [enwiki]> explain SELECT /* IndexPager::buildQueryInfo (LogPager) Admin */ log_id,log_type,log_action,log_timestam" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:20:24] (03CR) 10Reedy: "*************************** 1. row ***************************" [core] - 10https://gerrit.wikimedia.org/r/85508 (owner: 10Jackmcbarn) [02:28:22] I wonder why hashar is sending emails at this time of night... [02:30:24] (03CR) 10No-L: [C: 031] "Didn't file a Bug for that, but technically it seems to be one." [extensions/LdapAuthentication] - 10https://gerrit.wikimedia.org/r/89148 (owner: 10No-L) [02:43:52] (03CR) 10Ryan Lane: [C: 032] fixes getDomain() for UseLocal = true; [extensions/LdapAuthentication] - 10https://gerrit.wikimedia.org/r/89148 (owner: 10No-L) [02:43:54] (03Merged) 10jenkins-bot: fixes getDomain() for UseLocal = true; [extensions/LdapAuthentication] - 10https://gerrit.wikimedia.org/r/89148 (owner: 10No-L) [03:10:42] (03PS1) 10Legoktm: Remove link to Special:MassMessage in log entry description [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/90282 [03:11:54] (03PS3) 10Reedy: Fix mysqli support [core] - 10https://gerrit.wikimedia.org/r/90278 (owner: 10Tim Starling) [03:43:40] (03CR) 10Legoktm: [C: 031] "So it's been about 3 weeks since the patch splitting them was deployed. Has that been enough time to move ahead with this patch now?" [extensions/Echo] - 10https://gerrit.wikimedia.org/r/84870 (owner: 10Bsitu) [03:53:52] (03CR) 10Parent5446: "I'm confused as to what the point of these assertions are. Shouldn't a bot know who they're logged in as?" [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [03:55:31] (03CR) 10Legoktm: "Ideally yes. Cookies might expire, or as [[mw:Extension:AssertEdit]] says, you accidentally run your bot on the wrong wiki. This provides " [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [03:56:34] (03CR) 10Parent5446: [C: 031] "OK, sounds good." [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [03:58:11] (03CR) 10Reedy: "Some of it probably hails from when the api (and possibly query.php before it?) were less sophisticated with the amount of information it " [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [04:29:40] (03CR) 10Santhosh: [C: 032] Add gitignore nad jshint related files [extensions/ContentTranslation] - 10https://gerrit.wikimedia.org/r/90152 (owner: 10Siebrand) [04:34:25] (03PS2) 10Santhosh: Add gitignore nad jshint related files [extensions/ContentTranslation] - 10https://gerrit.wikimedia.org/r/90152 (owner: 10Siebrand) [04:38:36] (03PS1) 10Tim Starling: Fix SQLite exception when $wgDBuser is set [core] - 10https://gerrit.wikimedia.org/r/90288 [04:39:37] (03CR) 10Tim Starling: [C: 031] "Tested, please merge." [core] - 10https://gerrit.wikimedia.org/r/90278 (owner: 10Tim Starling) [04:42:13] (03PS2) 10Tim Starling: Fix SQLite exception when $wgDBuser is set [core] - 10https://gerrit.wikimedia.org/r/90288 [05:16:21] (03CR) 10Siebrand: [C: 032 V: 032] "Per Kartik. No Jenkins yet." [extensions/ContentTranslation] - 10https://gerrit.wikimedia.org/r/90152 (owner: 10Siebrand) [05:19:04] (03CR) 10Legoktm: "(1 comment)" [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/87646 (owner: 10Legoktm) [05:22:19] (03CR) 10Siebrand: [C: 04-1] "(1 comment)" [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/90282 (owner: 10Legoktm) [05:31:31] (03CR) 10Jdlrobson: "TODOs can be done in a follow up yes. I'm still a little concerned about the inline scripts - let's talk about this tomorrow" [core] - 10https://gerrit.wikimedia.org/r/82100 (owner: 10Bartosz Dziewoński) [05:33:56] (03CR) 10Krinkle: "(2 comments)" [extensions/Oversight] - 10https://gerrit.wikimedia.org/r/18229 (owner: 10Alex Monk) [05:35:10] (03PS1) 10Legoktm: Make it possible for EditPage::getCopyrightWarning to return HTML [core] - 10https://gerrit.wikimedia.org/r/90289 [05:36:46] (03CR) 10Jdlrobson: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/82100 (owner: 10Bartosz Dziewoński) [05:37:39] (03PS2) 10Legoktm: Make it possible for EditPage::getCopyrightWarning to return HTML [core] - 10https://gerrit.wikimedia.org/r/90289 [05:43:16] (03PS2) 10Legoktm: Remove link to Special:MassMessage in log entry description [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/90282 [05:43:27] (03CR) 10Legoktm: "(1 comment)" [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/90282 (owner: 10Legoktm) [05:50:06] (03CR) 10PleaseStand: [C: 04-1] "Actually, MWExceptions still have control over how they appear to users, though even before the change, that would have been broken if an " [core] - 10https://gerrit.wikimedia.org/r/89512 (owner: 10PleaseStand) [06:10:40] Krenair: did you file those messages sent to spam on bugzilla, with full headers? [06:18:54] (03CR) 10Siebrand: [C: 04-1] "(2 comments)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [06:24:44] (03PS1) 10Pastakhov: add operator while [extensions/Foxway] (develop) - 10https://gerrit.wikimedia.org/r/90290 [06:26:32] (03CR) 10Pastakhov: [C: 032] add operator while [extensions/Foxway] (develop) - 10https://gerrit.wikimedia.org/r/90290 (owner: 10Pastakhov) [06:26:38] (03Merged) 10jenkins-bot: add operator while [extensions/Foxway] (develop) - 10https://gerrit.wikimedia.org/r/90290 (owner: 10Pastakhov) [06:43:14] (03PS13) 10Ori.livneh: Cache ResourceLoader modules in localStorage [core] - 10https://gerrit.wikimedia.org/r/86867 [06:43:31] cache them modules! [06:47:21] (03CR) 10Ori.livneh: "PS13: Debounce calls to mw.loader.store.update. This actually simplifies the implementation because it means we can stop trying to be too " [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [06:52:14] (03PS1) 10Legoktm: Merge SimpleAntiSpam extension into core [core] - 10https://gerrit.wikimedia.org/r/90291 [06:53:04] (03CR) 10Siebrand: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [06:54:51] (03CR) 10Legoktm: "Will add RELEASE-NOTES later on." [core] - 10https://gerrit.wikimedia.org/r/90291 (owner: 10Legoktm) [07:05:30] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:06:52] Krinkle: blargh, right. [07:25:00] (03CR) 10Siebrand: [C: 031] "i18n/L10n are fine. Qqq will be added automatically, as the keys already exist." [core] - 10https://gerrit.wikimedia.org/r/90291 (owner: 10Legoktm) [07:33:53] (03CR) 10Krinkle: "(4 comments)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:43:03] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:43:27] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:50:34] (03CR) 10Krinkle: "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:50:42] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90260 (owner: 10Werdna) [07:51:55] (03CR) 10Krinkle: "last thought for now: Testing (perhaps something in combination with the load.php mock we have, and something like putting stuff in storag" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [07:57:25] (03CR) 10Siebrand: [C: 04-1] "(1 comment)" [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/90174 (owner: 10Netbrain) [08:16:59] (03PS5) 10Netbrain: Tagged v0.1, did various cleanup and started on v 0.2 [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/90174 [08:17:26] (03CR) 10Ori.livneh: "(3 comments)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [08:18:10] (03CR) 10Netbrain: "(1 comment)" [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/90174 (owner: 10Netbrain) [08:21:58] (03PS1) 10Werdna: Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 [08:27:19] (03PS2) 10Werdna: WIP: Moderation confirmation dialogues [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90260 [08:30:31] (03PS1) 10Aklapper: Sync custom template/en/custom/global/user-error.html.tmpl with upstream [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/90295 [08:33:05] (03CR) 10Skizzerz: [C: 031] "I don't see any issues with this" [core] - 10https://gerrit.wikimedia.org/r/90291 (owner: 10Legoktm) [08:33:20] (03CR) 10Aklapper: [C: 031] "+1" [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/90295 (owner: 10Aklapper) [08:39:27] (03PS14) 10Ori.livneh: Cache ResourceLoader modules in localStorage [core] - 10https://gerrit.wikimedia.org/r/86867 [08:42:15] (03CR) 10Ori.livneh: "PS14: Took Krinkle's suggestions; re-introduced mw.loader.store.enabled." [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [08:46:06] yes! http://blog.wolframalpha.com/2013/10/10/gotta-compute-em-all-wolframalphas-new-data-about-pokemon/ [08:48:32] i just tried it with "what does jigglypuff evolve into?" and it works, my mind is blown [09:48:26] (03CR) 10Siebrand: [C: 031] "L10n reviewed." [extensions/SideBarMenu] - 10https://gerrit.wikimedia.org/r/90174 (owner: 10Netbrain) [09:55:00] (03PS1) 10Matthias Mullie: Click is no longer required to activate textarea [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90300 [09:55:50] (03Draft1) 10Aude: Split changes list classes into separate files [core] - 10https://gerrit.wikimedia.org/r/90299 [10:17:00] (03PS1) 10Cicalese: First commit of new extension [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90303 [10:17:55] DanielK_WMDE_, 50 Warning: assert() [function.assert]: Assertion failed in /usr/local/apache/common-local/php-1.22wmf21/extensions/WikibaseDataModel/DataModel/Claim/Claims.php on line 276 [10:18:37] that is, this assertion is being constantly broken and you guys don't know that because assertions only pester us server admins [10:21:59] (03CR) 10Cicalese: [C: 032 V: 032] First commit of new extension [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90303 (owner: 10Cicalese) [10:44:37] (03CR) 10Krinkle: [C: 04-1] "TODO: Figure out what to do with DatabaseError::getLogMessage returning false. In the old code this effectively kept it out of the logs be" [core] - 10https://gerrit.wikimedia.org/r/89621 (owner: 10Krinkle) [10:48:46] MaxSem: actually, all our test systems are (should be) set to yell about assertions. [10:48:52] aude: --^^^^ [10:49:22] then they're not doing a good service;) [10:49:28] MaxSem: actually - is this still happening? This should be fixed as of tuesday's deployment... [10:49:52] yes, 33 of latest 1000 apache errors [10:50:20] currently our second mosty spammy error [10:50:31] MaxSem: so, the apache error logs contain failed assertions, but not actual uncaught exceptions? hm... a bit silly, no? [10:50:56] exceptions are in different logs [10:50:57] i'll check in a minute, need to commit my current stuff first [10:51:20] but are they, really? i was under the impression they didn't get logged at all [10:51:30] we had problems digging up a stack trace recently [10:51:32] DanielK_WMDE_: i'm not sure how often or at all the assertions would happen on the test systems [10:51:39] uncaught exceptions _are_ logged [10:52:06] although think we do have assertions on [10:52:18] MaxSem: just completely uncaught, or also "unexpected non.edmiawiki exception" cases? [10:52:28] uncaught [10:52:38] that's no good then [10:52:55] because mediawiki catches all unknown exceptions and outputs a generic error message [10:53:02] yup [10:53:04] (and throws away anything useful for debugging) [10:53:06] and logs them [10:53:15] really? where? couldn't find the code that does this [10:53:25] solution? throw mediawiki exceptions;) [10:53:26] (well, i only spend 10 minutes looking) [10:53:33] we had a bug with a non-mediawiki exception yesterday [10:53:39] not logged, nobody could find logs [10:54:00] MaxSem: not a practical solution to use only mwexception [10:54:00] s [10:54:04] MaxSem: bad idea, actually. exceptions should be specific to the error case, not the framework. [10:54:06] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86867 (owner: 10Ori.livneh) [10:54:19] we are using quite a bit of code that should not have any dependencies on mediawiki, too [10:54:22] since we make some of our code not depend on mediawiki (e.g. can be used by bots) [10:54:34] MaxSem: i don't hink there is any good reason to not log non-mw errors [10:54:50] anyway. let me finish this so i can look for the assertion thing [10:55:08] i think we just need to see where/how to change to log all uncaught exceptions [10:55:16] indeed [10:55:24] got confused when trying to do that yesterday [10:55:33] the error reporting stuff is quite convoluted [10:55:39] * aude might poke at it in a bit but know it's complicated [10:55:46] think it is [10:55:54] might be good to write to wikitech about it [10:55:56] sure [10:56:03] i know i looked into it at one time [11:07:23] MaxSem: fixed in Idd5ffb72774a - sadly didn't make it into the branch last week. damn, i was sure it did [11:07:24] sadly, it' [11:07:32] it's not easy to backport. [11:07:47] think you can live with it for 2 more weeks? [11:07:52] week and half [11:08:32] I do wonder though why it would happen so often - should be triggered only by an edge case, namely two statements in a data item being exactly the same [11:08:56] actually.... [11:09:07] * aude wonders too [11:10:27] (03CR) 10Krinkle: "It is." [core] - 10https://gerrit.wikimedia.org/r/71184 (owner: 10Alex Monk) [11:14:37] (03CR) 10Krinkle: "Let me know whether or not you're interested in addressing this." [core] - 10https://gerrit.wikimedia.org/r/71184 (owner: 10Alex Monk) [11:23:59] (03PS1) 10Zfilipin: Updated Ruby gems [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90311 [12:03:14] !g I54c3de6d836702ffbe3044bc58a38e83e758bc33 [12:03:14] https://gerrit.wikimedia.org/r/#q,I54c3de6d836702ffbe3044bc58a38e83e758bc33,n,z [12:03:31] (03PS1) 10Krinkle: resourceloader: Use state "error" instead of "missing" in case of exceptions [core] - 10https://gerrit.wikimedia.org/r/90320 [12:05:15] (03CR) 10Gergő Tisza: [C: 031] "Looks good. We might want to filter out hidden categories later; stuff like "uploaded with FlickrLickr" is probably of little interest to " [extensions/CommonsMetadata] - 10https://gerrit.wikimedia.org/r/90213 (owner: 10MarkTraceur) [12:07:12] (03CR) 10Siebrand: "(3 comments)" [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90303 (owner: 10Cicalese) [12:07:57] (03CR) 10Siebrand: [C: 031] "L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90300 (owner: 10Matthias Mullie) [12:08:05] (03CR) 10jenkins-bot: [V: 04-1] resourceloader: Use state "error" instead of "missing" in case of exceptions [core] - 10https://gerrit.wikimedia.org/r/90320 (owner: 10Krinkle) [12:08:20] Krinkle: got an oppinion about https://gerrit.wikimedia.org/r/#/c/74170/ ? [12:08:54] this issue is causing annoyance time and time again - namely, causign test failures without any error report. or any output, really. [12:09:48] (03PS4) 10Daniel Kinzler: Detect leakage of error_reporting state in tests. [core] - 10https://gerrit.wikimedia.org/r/74170 [12:10:34] DanielK_WMDE_: Looks good to me, except I don't like the configuration toggle. I hope we can fix remaining failures now, and then just introduce this. [12:10:35] (03CR) 10jenkins-bot: [V: 04-1] Detect leakage of error_reporting state in tests. [core] - 10https://gerrit.wikimedia.org/r/74170 (owner: 10Daniel Kinzler) [12:11:12] but if it happens to be common and non-trivial to fix (e.g. there are cases where it isn't really a bug, but "intentional") then the config toggle might make sense. [12:11:20] Is that is the case? [12:12:40] (03PS1) 10Zfilipin: Updated Ruby gems [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90322 [12:13:27] Krinkle: no, this should never be intentional. The config toggle is there because it's generally preferred for deployment to have a config switch for changing from old to new behavior. [12:14:24] Krinkle: I do think it should be enabled per default, but that would mean I'd have to find and fix all such problems myself, before submitting this patch. [12:14:34] With the patch merged and the switch in place, others can help with that [12:15:21] GAH! I hate gerrit! It reverted my commit because i failed to reload the page before changing the commit message! [12:15:27] *stupid* [12:15:38] (03PS5) 10Daniel Kinzler: Detect leakage of error_reporting state in tests. [core] - 10https://gerrit.wikimedia.org/r/74170 [12:15:55] ok, now, i hope [12:16:19] Krinkle: care to +1? [12:16:29] this has been sitting there for months now :) [12:24:36] hashar: loha [12:27:09] hashar: oh, you are back! Would you have another look at https://gerrit.wikimedia.org/r/#/c/74170/ ? [12:27:15] it would be really useful to have that in. [12:34:46] (03PS3) 10Hashar: wrapper around MediaWiki core phpunit [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90146 [12:35:23] DanielK_WMDE_: I think hashar is ignoring us atm [12:36:05] (03CR) 10Hashar: [C: 032] "Good enough for now, will amend later if need be." [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90146 (owner: 10Hashar) [12:36:08] (03Merged) 10jenkins-bot: wrapper around MediaWiki core phpunit [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90146 (owner: 10Hashar) [12:38:21] (03PS3) 10Werdna: WIP: Moderation confirmation dialogues. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90260 [12:38:23] (03PS1) 10Werdna: Fix notice in WorkflowLoader that was inexplicably only showing up for moderation [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90324 [12:38:26] AzaToth: DanielK_WMDE_ sorry I am busy this afternoon [12:38:38] AzaToth: DanielK_WMDE_ got to finish a bunch of shell scripts [12:39:39] hashar: ok, but i'll bug you about it again :) [12:40:17] heh [12:40:34] I'm anti-busy all day [12:42:56] (03PS1) 10Hashar: mw setting injecter was not initially closing PHP tag [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90325 [12:43:04] (03CR) 10Hashar: [C: 032] mw setting injecter was not initially closing PHP tag [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90325 (owner: 10Hashar) [12:43:09] (03Merged) 10jenkins-bot: mw setting injecter was not initially closing PHP tag [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90325 (owner: 10Hashar) [12:45:24] 12:44:44 mkdir: cannot create directory `/srv/ssd/jenkins-slave/workspace/mediawiki-core-regression-master/log': File exists [12:45:25] bahh [12:58:42] (03PS1) 10Hashar: mw-apply-settings: copy settings in dir log, not as a file [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90327 [12:58:51] (03CR) 10Hashar: [C: 032] mw-apply-settings: copy settings in dir log, not as a file [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90327 (owner: 10Hashar) [12:58:54] (03Merged) 10jenkins-bot: mw-apply-settings: copy settings in dir log, not as a file [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90327 (owner: 10Hashar) [13:01:38] (03PS1) 10Hashar: mw-apply-settings: make sure log dir exist [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90329 [13:01:44] (03CR) 10Hashar: [C: 032] mw-apply-settings: make sure log dir exist [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90329 (owner: 10Hashar) [13:01:47] (03Merged) 10jenkins-bot: mw-apply-settings: make sure log dir exist [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90329 (owner: 10Hashar) [13:02:32] I should really be more careful and test locally :(- [13:04:12] hashar: http://www.commitstrip.com/en/2013/10/11/mieux-et-moins-cher-que-les-tests-unitaires/ [13:04:29] Nikerabbit: french includedç [13:04:48] that is me :-] [13:08:53] Nikerabbit: nice one [13:22:12] * werdna tests hashar locally [13:23:01] (03PS9) 10Matthias Mullie: Add topic participants' info, amount of comment [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 [13:24:42] (03PS10) 10Matthias Mullie: Add topic participants' info, amount of comment [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 [13:31:29] (03PS15) 10Gergő Tisza: Add "extended" file metadata to API [core] - 10https://gerrit.wikimedia.org/r/78926 (owner: 10Brian Wolff) [13:31:30] (03PS5) 10Gergő Tisza: Add extmetadata to the ForeignAPIFile iiprops [core] - 10https://gerrit.wikimedia.org/r/88889 (owner: 10MarkTraceur) [13:33:14] werdna: no need. Just deploy hashar and wait for the screams [13:33:25] (03CR) 10Gergő Tisza: "Patch set 4 was rebased." [core] - 10https://gerrit.wikimedia.org/r/88889 (owner: 10MarkTraceur) [13:33:40] Real Developers test on production, etc. [13:33:41] (03CR) 10Matthias Mullie: "The BagOStuff::get calls should have been greatly reduced now." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [13:40:21] (03CR) 10Gergő Tisza: [C: 04-1] "Descending deeper into array format hell, FormatMetadata::fetchExtendedMetadata() returns a different format for local and remote files. T" [core] - 10https://gerrit.wikimedia.org/r/78926 (owner: 10Brian Wolff) [13:40:56] (03CR) 10Matthias Mullie: [C: 032] Fix notice in WorkflowLoader that was inexplicably only showing up for moderation [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90324 (owner: 10Werdna) [13:41:01] (03Merged) 10jenkins-bot: Fix notice in WorkflowLoader that was inexplicably only showing up for moderation [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90324 (owner: 10Werdna) [13:43:34] (03PS1) 10Hashar: mw-run-phpunit*: make sure we compress mw log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90336 [13:43:44] (03CR) 10Hashar: [C: 032] mw-run-phpunit*: make sure we compress mw log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90336 (owner: 10Hashar) [13:43:47] (03Merged) 10jenkins-bot: mw-run-phpunit*: make sure we compress mw log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90336 (owner: 10Hashar) [13:45:25] (03CR) 10Matthias Mullie: [C: 032] Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:45:29] (03CR) 10jenkins-bot: [V: 04-1] Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:47:08] (03PS2) 10Matthias Mullie: Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:47:13] (03CR) 10jenkins-bot: [V: 04-1] Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:48:05] (03PS1) 10Hashar: ignore gzip errors when compressing log [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90337 [13:48:12] (03CR) 10Hashar: [C: 032] ignore gzip errors when compressing log [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90337 (owner: 10Hashar) [13:48:15] (03Merged) 10jenkins-bot: ignore gzip errors when compressing log [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90337 (owner: 10Hashar) [13:48:46] (03PS3) 10Hashar: Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:48:55] (03CR) 10Hashar: "Jenkins issue, sorry." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:50:03] (03CR) 10Matthias Mullie: [C: 032] "Thanks ;)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:50:05] !jenkins mediawiki-core-code-coverage [13:50:05] https://integration.wikimedia.org/ci/job/mediawiki-core-code-coverage [13:50:11] (03Merged) 10jenkins-bot: Fix for bug 55530. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90294 (owner: 10Werdna) [13:57:24] (03CR) 10Krinkle: "But is one of them run always? e.g. do we need to listen in both?" [extensions/AbuseFilter] - 10https://gerrit.wikimedia.org/r/71945 (owner: 10Hoo man) [14:00:29] (03PS1) 10Hashar: publisher phpunit-junit-2 for mw-apply-settings macro [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90342 [14:00:31] (03PS1) 10Hashar: mw regression jobs now use shell script [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90343 [14:00:49] (03CR) 10jenkins-bot: [V: 04-1] mw regression jobs now use shell script [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90343 (owner: 10Hashar) [14:00:51] (03CR) 10Hashar: [C: 032] macro mw-apply-settings [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90341 (owner: 10Hashar) [14:00:55] (03CR) 10Hashar: [C: 032] publisher phpunit-junit-2 for mw-apply-settings macro [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90342 (owner: 10Hashar) [14:01:00] (03CR) 10Hashar: [C: 032] mw regression jobs now use shell script [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90343 (owner: 10Hashar) [14:02:23] (03Merged) 10jenkins-bot: macro mw-apply-settings [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90341 (owner: 10Hashar) [14:02:25] (03Merged) 10jenkins-bot: publisher phpunit-junit-2 for mw-apply-settings macro [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90342 (owner: 10Hashar) [14:02:38] (03Merged) 10jenkins-bot: mw regression jobs now use shell script [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90343 (owner: 10Hashar) [14:12:52] the echo's user talk message notification email is much less informative than the old one... [14:13:17] https://bugzilla.wikimedia.org/show_bug.cgi?id=55838 https://bugzilla.wikimedia.org/show_bug.cgi?id=55827 [14:19:11] (03CR) 10Zfilipin: "Feel free to make smaller or bigger patch sets, what ever makes sense to you. I prefer smaller patch sets, but that is only personal prefe" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:21:53] AaronSchulz, is there a bug somewhere about converting use of user_text fields to UserCache calls? [14:22:46] e.g. https://gerrit.wikimedia.org/r/#/c/34032/ [14:24:59] (03CR) 10Zfilipin: "(1 comment)" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:25:44] (03CR) 10Anomie: "I see there are rows in logging on enwiki that have log_user_text empty, but it seems for all that do have a log_user_text the value match" [core] - 10https://gerrit.wikimedia.org/r/71184 (owner: 10Alex Monk) [14:26:21] (03CR) 10Chad: [C: 032] Resubmit "Add support for mysqli extension" [core] - 10https://gerrit.wikimedia.org/r/90269 (owner: 10Tim Starling) [14:26:26] (03CR) 10Zfilipin: "(1 comment)" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:26:51] (03CR) 10Chad: [C: 032] Fix mysqli support [core] - 10https://gerrit.wikimedia.org/r/90278 (owner: 10Tim Starling) [14:29:06] (03PS1) 10Ebrahim: Native digits on 'articlefeedbackv5-article-feedback-link' [extensions/ArticleFeedbackv5] - 10https://gerrit.wikimedia.org/r/90349 [14:29:45] (03Merged) 10jenkins-bot: Resubmit "Add support for mysqli extension" [core] - 10https://gerrit.wikimedia.org/r/90269 (owner: 10Tim Starling) [14:30:47] (03PS9) 10Zfilipin: Created a test for ProveIt gadget [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:31:15] AaronSchulz, nvm, got it: https://bugzilla.wikimedia.org/show_bug.cgi?id=31863 [14:31:47] !jenkins mwext-Wikibase-qunit [14:31:47] https://integration.wikimedia.org/ci/job/mwext-Wikibase-qunit [14:32:19] (03Merged) 10jenkins-bot: Fix mysqli support [core] - 10https://gerrit.wikimedia.org/r/90278 (owner: 10Tim Starling) [14:32:50] (03CR) 10Ebrahim: "As you said before, this is not a good solution specially because this message is using {{PLURAL:{{ http://translatewiki.net/w/i.php?title" [extensions/ArticleFeedbackv5] - 10https://gerrit.wikimedia.org/r/90349 (owner: 10Ebrahim) [14:35:17] (03CR) 10Alex Monk: "Actually according to https://bugzilla.wikimedia.org/show_bug.cgi?id=31863 we want to cut down on the use of user_text stuff." [core] - 10https://gerrit.wikimedia.org/r/71184 (owner: 10Alex Monk) [14:35:50] !j [14:35:50] There is no such key, you probably want to try: !2.0, !allowoverride, !centralauth, !cia, !git, !gitaccount, !help, !html9, !imabot, !importhelp, !javascript, !jobqueue, !jquery, !labs, !mediawiki, !morbo, !nosql, !padindex, !queue, !secrets, !selinux, !support, !testswarm, !tidyforidiots, !tooquiet, !tss, !undo, !unicode_test, !unreviewedcode, !vector, !webscale, !welcome, !wmf, !worstcase, !yoda, !visualeditor, !ask, !jobs, !farm, !blink, !jsduck, !job, !ajax, !repos, !jjb, !card, !jenkins, [14:36:00] !html9 [14:36:00] http://html9responsiveboilerstrapjs.com/ [14:36:15] (03CR) 10Zfilipin: "Replaced double quotes with single quotes (according to our coding convention)." [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:38:21] (03CR) 10Zfilipin: [C: 031] "The scenarios run fine on my machine and the code looks good. Chris, Matt, feel free to merge this into master, if you do not see anything" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/87436 (owner: 10Plavi) [14:39:23] (03PS1) 10Matthias Mullie: Fix moderation permissions [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90351 [14:39:28] (03PS1) 10Hashar: properly set $wgDebugLogFile path under Apache [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90352 [14:39:40] (03CR) 10Hashar: [C: 032] properly set $wgDebugLogFile path under Apache [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90352 (owner: 10Hashar) [14:39:50] (03Merged) 10jenkins-bot: properly set $wgDebugLogFile path under Apache [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90352 (owner: 10Hashar) [14:41:12] (03CR) 10Addshore: [C: 031] Split changes list classes into separate files [core] - 10https://gerrit.wikimedia.org/r/90299 (owner: 10Aude) [14:42:36] greg-g: Is someone assigned to https://bugzilla.wikimedia.org/show_bug.cgi?id=55808 ? [14:51:33] (03PS1) 10Cmcmahon: do not depend on a particular URL in case of redirect [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90354 [14:53:55] !secrets [14:53:55] 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 [14:54:01] oh [14:56:41] hashar: so... [15:06:18] (03CR) 10Zfilipin: [C: 032] do not depend on a particular URL in case of redirect [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90354 (owner: 10Cmcmahon) [15:12:23] (03PS1) 10Hashar: apache debug log permission need to be done by Job [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90355 [15:12:39] (03CR) 10Hashar: [C: 032] apache debug log permission need to be done by Job [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90355 (owner: 10Hashar) [15:12:44] (03Merged) 10jenkins-bot: apache debug log permission need to be done by Job [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90355 (owner: 10Hashar) [15:14:58] (03CR) 10Anomie: [C: 04-1] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [15:15:11] Krinkle: no, I do not, going to bring up in today's MW Core meeting... [15:18:37] (03PS2) 10Cmcmahon: Updated Ruby gems [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90322 (owner: 10Zfilipin) [15:18:49] (03CR) 10Cmcmahon: [C: 032] "maintenance" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90322 (owner: 10Zfilipin) [15:19:51] Nikerabbit, hi .. [15:19:53] (03PS2) 10Cmcmahon: Updated Ruby gems [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90311 (owner: 10Zfilipin) [15:20:04] (03CR) 10Cmcmahon: [C: 032] "maintenance" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90311 (owner: 10Zfilipin) [15:20:34] (03Merged) 10jenkins-bot: Updated Ruby gems [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90311 (owner: 10Zfilipin) [15:23:26] subbu: yes? [15:23:53] did you figure out the parser api issue you reported? [15:26:20] subbu: nope [15:26:50] subbu: I'm quite busy with all other things so any tips are welcome [15:27:07] what is the url of the api call? [15:27:23] (03PS1) 10Reedy: Update Wikidata related branches [tools/release] - 10https://gerrit.wikimedia.org/r/90356 [15:27:39] subbu: let me check [15:27:52] damn I think I closed it [15:28:03] (03CR) 10Reedy: [C: 032] Update Wikidata related branches [tools/release] - 10https://gerrit.wikimedia.org/r/90356 (owner: 10Reedy) [15:28:05] (03Merged) 10jenkins-bot: Update Wikidata related branches [tools/release] - 10https://gerrit.wikimedia.org/r/90356 (owner: 10Reedy) [15:28:23] but step 1) make a translatable page 2) action=edit&title=that_page&text=some text with tags [15:35:10] (03PS1) 10Anomie: Pass rights needed to protectedpagewarning [core] - 10https://gerrit.wikimedia.org/r/90357 [15:35:33] Nikerabbit: have you had a chance to look at https://gerrit.wikimedia.org/r/#/c/89641/ ? [15:36:28] ori-l: not the new version [15:36:42] (03CR) 10Gergő Tisza: "MarkTraceur: that was an issue with the API; the extension does not change metadata from a foreign file and should not cause problems as f" [extensions/CommonsMetadata] - 10https://gerrit.wikimedia.org/r/80403 (owner: 10Brian Wolff) [15:37:21] Nikerabbit: could you? would be nice to get it out. also, translate bug (https://bugzilla.wikimedia.org/show_bug.cgi?id=55808): who's the person who should look into that? is there someone? [15:40:52] if all else fails, I'm going to bring it up doing the MW Core meeting, but suggestions welcome (on the translate bug) [15:43:51] !jenkins mwext-Wikibase-repoapi-tests [15:43:51] https://integration.wikimedia.org/ci/job/mwext-Wikibase-repoapi-tests [15:44:21] (03CR) 10Gergő Tisza: [C: 031] Add extmetadata to the ForeignAPIFile iiprops [core] - 10https://gerrit.wikimedia.org/r/88889 (owner: 10MarkTraceur) [15:47:47] Nikerabbit: it might be worth trying to use the tag hook directly and just see if there is still a reason not to use it [15:48:25] the issues you ran into in the past might be fixed now- I vaguely remember there being fixes around headings from extensions, section editing etc [15:50:26] (03PS1) 10Hashar: mw exts now have injected settings + artifacts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90358 [15:50:30] gwicke: I guess that is the only option now [15:50:53] gwicke, ori-l, greg-g: I'll have a look at that bug next week, or do you think it is more urgent? [15:52:25] Nikerabbit: I guess not terribly urgent, people are working around it and it's been around for a bit [15:52:47] Nikerabbit: I'll still bring it up with MW Core people and see what they think, but assume you're on point for now :) [15:53:16] Parsoid might run into those errors now [15:53:25] greg-g: what's the meaning of "on point"? [15:53:42] Nikerabbit: sorry, idiom, means "person with responsibility" [15:53:45] is action=parse triggering it? [15:53:49] gwicke: yes [15:54:00] yes, then we get errors [15:54:05] gwicke: funnily enough without this patch action=parse would have given hookaborted always [15:54:29] we never tried to evaluate the tags before you registered them [15:54:54] would it be better to undo the previous edit while we work on better fix? [15:55:05] Nikerabbit, I think the problem is when .. shows up in translcusion params. [15:55:13] maybe- I don't really know how common those tags are [15:55:31] https://meta.wikimedia.org/w/api.php?action=parse&text=%3Ctranslate%3Efoo%3C/translate%3E works properly [15:55:51] (03PS2) 10Hashar: mw exts now have injected settings + artifacts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90358 [15:56:07] action=edit seems to be doing something differently, I just don't know what [15:56:18] Nikerabbit, this also bombs: https://meta.wikimedia.org/w/api.php?action=parse&text={{messagebox|%20l-width=%2040px|%20image=%20[[File:Correct.svg|40px|link=]]|%20text=%20%27%27%27%3Ctranslate%3E%3C!--T:1--%3EThis%20page%20is%20currently%20a%20draft.%3C/translate%3E%27%27%27%20%3Ctranslate%3E%3C!--T:2--%3EMore%20information%20pertaining%20to%20this%20may%20be%20available%20on%20the%20[[{{TALKPAGENAME}}|talk%20page]].%3C/translate%3E}} [15:56:20] something about the ParseBeforeStrip hook [15:56:28] subbu: interesting- do we have a page that has a lot of translate tags for testing on parsoid.wmflabs.org? [15:56:35] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [core] - 10https://gerrit.wikimedia.org/r/90357 (owner: 10Anomie) [15:56:35] and yes there seems to be multiple ways to trigger the exception [15:57:05] gwicke, using that wikitext there should be sufficient. [15:57:13] from https://meta.wikimedia.org/w/index.php?title=Template:Draft&action=edit [16:00:00] we'll still round-trip it properly, but it will obviously show the error rather than the expected output [16:00:07] so not the end of the world, but not nice either [16:00:36] this is also being cached, and we currently don't have a good way to selectively re-render just broken pages [16:00:58] Nikerabbit, gwicke yes, maybe reverting that patch for now would be good to prevent these crashers .. I imagine this can be debugged using a local mediawiki install now? [16:01:11] I am off [16:01:13] see you tonight [16:01:27] (03PS1) 10Raimond Spekking: Add messages for new group 'templateeditor' [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/90359 [16:02:11] subbu, Nikerabbit: yeah, +1 for temporarily reverting from me too [16:02:43] this is a simplified test case: https://meta.wikimedia.org/w/api.php?action=parse&text={{messagebox|%20text=%3Ctranslate%3E%3C!--T:1--%3Efoo%3C/translate%3E}} [16:02:51] subbu: yes I Can debug it locally [16:03:02] (03CR) 10Siebrand: [C: 032] Add messages for new group 'templateeditor' [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/90359 (owner: 10Raimond Spekking) [16:03:15] ori-l: whii [16:03:29] gwicke, actually even this crashes it (without the messagebox) https://meta.wikimedia.org/w/api.php?action=parse&text=%3Ctranslate%3E%3C!--T:1--%3Efoo%3C/translate%3E [16:04:15] (03Merged) 10jenkins-bot: Add messages for new group 'templateeditor' [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/90359 (owner: 10Raimond Spekking) [16:05:52] (03CR) 10Raimond Spekking: "Getting on my local wiki:" [extensions/UploadWizard] - 10https://gerrit.wikimedia.org/r/89853 (owner: 10Yuvipanda) [16:14:45] (03PS2) 10Legoktm: Merge SimpleAntiSpam extension into core [core] - 10https://gerrit.wikimedia.org/r/90291 [16:31:29] (03CR) 10Hashar: [C: 032] "deployed" [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90358 (owner: 10Hashar) [16:33:12] (03Merged) 10jenkins-bot: mw exts now have injected settings + artifacts [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90358 (owner: 10Hashar) [16:34:17] Nikerabbit: and https://gerrit.wikimedia.org/r/#/c/90090/ ! last blocker :P [16:37:17] ori-l: that's unfortunately blocked by integration tests for the font applying [16:37:38] you might want to cherry pick your changes [16:38:13] okay, will do [16:40:55] Nikerabbit: oh, you mean into one of the production branches? [16:41:22] I thought you meant cherry-pick the specific upstream change [16:42:28] Nikerabbit: if that is what you meant, then: https://gerrit.wikimedia.org/r/#/c/90365/ . if not, let me know [16:42:49] ori-l: upstream [16:43:02] 90365 then :) [16:56:44] (03CR) 10EBernhardson: [C: 032] "from 5k debug entries down to 600(many of which were there previously), good enough i would say :)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [16:56:47] (03CR) 10jenkins-bot: [V: 04-1] Add topic participants' info, amount of comment [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [16:57:01] (03CR) 10MarkTraceur: "(3 comments)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [16:58:42] (03CR) 10MarkTraceur: "Adding Fabrice to the review to respond to or address the copy-related comment" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 (owner: 10MarkTraceur) [17:00:56] (03PS11) 10EBernhardson: Add topic participants' info, amount of comment [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [17:01:54] (03CR) 10EBernhardson: [C: 032] "needed a tiny rebase" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [17:02:04] (03Merged) 10jenkins-bot: Add topic participants' info, amount of comment [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89220 (owner: 10Matthias Mullie) [17:15:19] jdlrobson: Are there pending issues with your patch for BetaFeatures that I can/need to halp with? [17:15:40] 1s marktraceur just helping adam bass with something then i'm all yours (dr0ptp4kt ) [17:15:48] Sure sure [17:15:53] Async! [17:16:19] jdlrobson thx btw [17:18:05] (03CR) 10Fabriceflorin: [C: 04-1] "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 (owner: 10MarkTraceur) [17:20:40] (03CR) 10Legoktm: "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [17:21:01] (03PS2) 10Legoktm: Merge AssertEdit extension into core [core] - 10https://gerrit.wikimedia.org/r/90263 [17:24:11] (03CR) 10Krinkle: [C: 04-1] "Code looks good (untested) but I don't think we should introduce a global config for this. Find the problems, fix them, then introduce thi" [core] - 10https://gerrit.wikimedia.org/r/74170 (owner: 10Daniel Kinzler) [17:25:44] (03PS2) 10Jforrester: Shift divs about [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/87279 (owner: 10MarkTraceur) [17:34:53] (03PS1) 10Ori.livneh: Update ULS to master [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90373 [17:36:11] Nikerabbit: if you +1 ^ i'll try to get this out today, either at the tail end of the platform window or during LD [17:36:39] Oh shit [17:36:41] Ty ubuntu [17:36:43] and php [17:36:50] ( ! ) Notice: Use of undefined constant JSON_HEX_TAG - assumed 'JSON_HEX_TAG' in /var/www/wiki/mediawiki/core/includes/json/FormatJson.php on line 149 [17:36:59] We haven't resolved this one yet, have we? [17:37:05] ( ! ) Fatal error: Call to undefined function json_encode() in /var/www/wiki/mediawiki/core/includes/json/FormatJson.php on line 150 [17:37:25] xdebug apparently started working again [17:37:28] (03CR) 10Ori.livneh: "Changes:" [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90373 (owner: 10Ori.livneh) [17:37:46] (03CR) 10Nikerabbit: [C: 031] Update ULS to master [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90373 (owner: 10Ori.livneh) [17:38:13] woot, thank you [17:39:31] I'm sure we had a bug for this!? [17:41:20] i think so, let's see [17:41:38] The solution is apt-get install php5-json [17:42:11] (03PS25) 10Reedy: Add archive, externallinks PK [core] - 10https://gerrit.wikimedia.org/r/51675 [17:42:40] Needs attaching to the 5.5 tracker bug at https://bugzilla.wikimedia.org/show_bug.cgi?id=49975 [17:42:46] https://bugzilla.wikimedia.org/show_bug.cgi?id=47431 [17:43:26] (03PS3) 10MarkTraceur: Shift divs about [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/87279 [17:43:52] (03PS26) 10Reedy: Add archive, externallinks PK [core] - 10https://gerrit.wikimedia.org/r/51675 [17:44:19] (03CR) 10Reedy: [C: 032] Add archive, externallinks PK [core] - 10https://gerrit.wikimedia.org/r/51675 (owner: 10Reedy) [17:44:46] (03PS1) 10Cmcmahon: restore working ULS tests [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90380 [17:45:11] Thanks ori-l [17:50:25] (03Merged) 10jenkins-bot: Add archive, externallinks PK [core] - 10https://gerrit.wikimedia.org/r/51675 (owner: 10Reedy) [17:57:09] (03CR) 10Reedy: [C: 032 V: 032] Update ULS to master [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90373 (owner: 10Ori.livneh) [17:57:42] <3 <3 <3 [17:59:14] (03CR) 10BryanDavis: "(2 comments)" [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90147 (owner: 10Dan-nl) [18:00:41] how should an updated upstream be handled when there are like 400 new commits? [18:02:48] (03PS1) 10Cicalese: Fixed internationalization issues. [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90383 [18:03:01] Such as? [18:04:18] was looking into buck again, and google has rebased 400 commits from facebook, removed two of its own commits and added 2-4 new ones [18:05:17] AzaToth: in your commit message, indicate the range of commits you are pulling in, e.g.: 2722493...0deb96b, and provide a link, like: https://github.com/jquery/jquery/compare/45aa20e...083edd60 [18:05:48] squashed I assume then [18:05:51] and ideally at least outline the changes ("rewrite of foo engine and speed improvements") [18:05:55] yes, not 400 gerrit changes [18:06:37] lol [18:06:57] Why not use a remote? [18:06:58] the outline of the changes will be "updates needed to actually build gerrit" [18:07:00] uh [18:07:01] submodule [18:07:06] or that, yeah [18:07:50] Reedy: because git buildpackage doesn't handle submodules [18:08:36] AzaToth: you need to give a fuller picture of what you're trying to do if you want good advice [18:09:50] ori-l: same as we do every night... [18:09:59] ori-l: as you know, to build gerrit you now need buck, which is not releasd yet, made by facebook (to be used for android), modified by google to be used for gerrit, modified by us to be installable [18:10:29] puff, puff, pass [18:10:41] ツ [18:12:47] (03PS1) 10Cicalese: Fixed internationalization issues. [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90384 [18:16:12] (03Abandoned) 10Cicalese: Fixed internationalization issues. [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90383 (owner: 10Cicalese) [18:16:47] (03CR) 10Cicalese: [C: 032 V: 032] Fixed internationalization issues. [extensions/TitleIcon] - 10https://gerrit.wikimedia.org/r/90384 (owner: 10Cicalese) [18:19:18] (03PS2) 10MarkTraceur: Add logo to feature preference, fix copy [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 [18:19:19] (03CR) 10jenkins-bot: [V: 04-1] Add logo to feature preference, fix copy [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 (owner: 10MarkTraceur) [18:21:40] (03PS3) 10MarkTraceur: Add logo to feature preference, fix copy [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 [18:22:03] (03CR) 10MarkTraceur: "Manual rebase (accursed l10n updates)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/87276 (owner: 10MarkTraceur) [18:24:14] siebrand: So stupid question: Why aren't the license messages in core? UploadWizard could use them, too... [18:25:19] Oh, actually, no [18:25:23] They're useless for this [18:25:27] And for UW [18:27:22] (03CR) 10EBernhardson: "Apologies for missing the prod migration, got lost in the shuffle when we changed ideas on how this should work. A new schema change bug " [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson) [18:27:36] (03CR) 10MarkTraceur: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [18:34:09] (03PS12) 10Reedy: Create maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [18:34:54] (03PS1) 10Aaron Schulz: Reduce excess lock contention and chance for failure on rename [core] - 10https://gerrit.wikimedia.org/r/90385 [18:35:23] (03CR) 10Aaron Schulz: "e.g:" [core] - 10https://gerrit.wikimedia.org/r/90385 (owner: 10Aaron Schulz) [18:37:27] (03CR) 10Reedy: [C: 031] "This is good to merge now, dependant commit was merged earlier" [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [18:38:05] yo marktraceur, is the timeline of deploying betafeatures to testwikis next week still doable? [18:38:31] greg-g: I don't know, jdlrobson still hasn't answered my question from earlier and I don't know what James_F and edsanders|away think either [18:38:40] well then [18:38:53] marktraceur: have them know by today, kthx :) [18:39:02] *nod* [18:39:51] marktraceur: I think we may have to push back. :-( [18:40:10] James_F: I'm not sad about that, you shouldn't be either [18:40:21] marktraceur: I'm sad to the extent that it's my fault. [18:40:24] It's realistic [18:45:30] (03PS13) 10SuchetaG: mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [18:47:22] (03PS1) 10Legoktm: Remove period from help, breaks the link and is inconsistent [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90388 [18:54:30] (03PS14) 10SuchetaG: mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [18:58:20] (03PS15) 10SuchetaG: mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [19:05:44] (03PS8) 10leucosticte: Add InterwikiMap extension, v1.0.6 [extensions/InterwikiMap] - 10https://gerrit.wikimedia.org/r/35024 [19:11:22] Sigh [19:11:32] * marktraceur bites bullet, rebases string of commits [19:11:59] wtf is "string of commits"? [19:12:35] AzaToth: I have three commits that depend on each other, and another string of three commits that consists of the same base and two different ones [19:12:55] I want to add one commit from the first chain into the middle of the second chain [19:13:00] I have no idea how badly this is going to go [19:13:05] I guess "really really badly" [19:13:37] heh [19:13:58] (03PS2) 10MarkTraceur: Add license to the header [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 [19:14:23] (03CR) 10MarkTraceur: "Note that I'm sort of in the middle of fixing things - standby for a rebase, then another fix patchset" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [19:15:20] So you want to use A-B-C and A-E-F and make A-E-A'-B'-C'-F? [19:15:47] Erm [19:15:50] So [19:15:59] dunno what you mean by "add one commit" [19:16:00] I want to have F based on E and also on B [19:16:14] C can still be based on only-B [19:16:20] But I think there's no way to do that [19:16:56] Urgh [19:17:06] I'll rebase E onto B and base C and F off of it [19:17:11] I think what you want to do it to make an interaticive rebase on A-E-F and cherry pick [19:18:14] It seems unnecessarily complicated [19:18:19] These all need to get merged anyway [19:18:20] wait, you mean you want to stand on F and _merge_ B? [19:18:33] AzaToth: I mean...OK. [19:18:42] I need both E and B for different reasons [19:18:46] In the next version of F [19:18:54] But E and B don't rely on each other [19:19:08] So it's like E and B could both be based only on A, but F needs both of them [19:19:38] well, F is based on E now, if if you merge B into F, then F is based on E and B [19:19:50] Right [19:19:58] But will Gerrit be _super_ confused? :P [19:20:10] I suggest yes. [19:20:13] heh [19:20:23] probably not [19:20:24] AzaToth: I can try though. What would the commands look like? [19:20:38] git checkout F; git merge B [19:21:04] Sounds like fun [19:21:08] * marktraceur does [19:21:21] well, I assume you are standing on F now, so the checkout isn't needed [19:21:49] so git merge SHA1-of-B should suffice [19:23:43] Hrm [19:24:03] Should I commit the merge separately, or as an amend? [19:24:15] separate [19:26:18] marktraceur: using bash? [19:26:30] or any relevant shell? [19:27:03] I'm using bash, yeah [19:28:28] marktraceur: https://gist.github.com/azatoth/7030805 [19:29:07] $(__git_ps1 " «%s»") plus the set variables will help you see if you fnuck up a merge/rebase [19:30:43] marktraceur: http://imgur.com/IHNzKVH [19:31:03] RoanKattouw, Namaste. [19:31:20] Hm [19:31:22] (03CR) 10Aaron Schulz: [C: 04-1] "Can you just add a flag to populateRevisionSha1 instead of a new file?" [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [19:31:43] "I dunno 'bout you, but namaste here any longer than I have to" (HIMYM reference of the day) [19:31:55] :) [19:31:59] sucheta: Howdy [19:32:07] marktraceur: so after a rebase, it says "azatoth@azaboxen:~/build/wmf/operations_debs_buck «upstream-google *$% u+400-4»$" you are fucked [19:32:12] marktraceur: I think that's a Barney quote? I'm trying to recall the context there [19:32:27] i.e. 400 new commits and 4 lost [19:33:10] marktraceur, I am trying really really hard to not get into HIMYM too much. [19:33:23] RoanKattouw, You have stuff to review. [19:33:36] RoanKattouw, https://gerrit.wikimedia.org/r/#/c/64582/12 [19:33:58] RoanKattouw: No, it's a Marshall quote, from when they went to Ted/Stella wedding at the yoga collective [19:34:05] Err [19:34:08] Ooooh [19:34:08] Spoiler alerts. [19:34:10] Right [19:34:21] sucheta: There's no such thing as "too much" with that show [19:34:36] I fully agree [19:34:48] * RoanKattouw is caught up to and including S09E03, FYI [19:34:57] Which might mean I'm a bit behind if new stuff came out recently [19:35:02] Probably 1 or 2 episodes [19:35:11] greg-g: I have a rather large centralnotice deply that I'd like to push out today -- may I do that after the LD? [19:35:33] huh, ok [19:35:42] mwalker: what you talkin' 'bout? [19:36:10] (does that just collapose into "talkinbout" with the consecutive 's?) [19:36:16] (03CR) 10Reedy: "I presume you mean populateRevisionLength? In the same way populateRevisionSha1 does both tables?" [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [19:36:39] prolly :p [19:36:51] mwalker: :) [19:37:05] its patches https://gerrit.wikimedia.org/r/#/c/87308/ https://gerrit.wikimedia.org/r/#/c/87366/ and https://gerrit.wikimedia.org/r/#/c/87360/ [19:37:12] (03CR) 10Aaron Schulz: "right, populateRevisionLength" [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [19:37:27] marktraceur, No it happens to me with any show I start to like. I go on watching episodes after episodes, till my eyes start to hurt. [19:37:29] so, a bug fix for something; and then a new feature which will also require a schema alter on a small table [19:37:37] ah [19:38:09] so, not like a large centralnotice notice going out that might increase load somewhere, right? [19:38:15] sucheta: That's just normal TV-watching in the 21st century! [19:38:15] (03PS5) 10Cmcmahon: Update browsertests for string change [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89389 (owner: 10Spage) [19:38:46] greg-g: not unless it does an ori and has unintended consequences with caching :p [19:38:49] (03CR) 10Cmcmahon: [C: 032] Update browsertests for string change [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89389 (owner: 10Spage) [19:38:53] but I'm not altering a RL module [19:38:57] mwalker: hah :) [19:38:58] k [19:39:03] (03CR) 10Aaron Schulz: [C: 032] If $wgShowHostnames is true, add parsed by to NewPP limit report [core] - 10https://gerrit.wikimedia.org/r/89751 (owner: 10Reedy) [19:40:34] so, I wasn't going to complain, but adam added it to today's calendar at 3pm: https://wikitech.wikimedia.org/w/index.php?title=Deployments&diff=86110&oldid=86044 [19:40:51] (I already archived this week's calendar down to the bottom, prepping for tomorrow/next week [19:40:54] +) [19:41:29] mwalker: ^ [19:41:50] heh; he said he did; but I didn't see it [19:41:55] I'll do it at three then :) [19:42:08] (03CR) 10Bartosz Dziewoński: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/90289 (owner: 10Legoktm) [19:42:09] k [19:42:26] you can be the guy who says "next time, ask first, greg'll add it to the calendar" ;) [19:42:48] *nods* [19:43:18] mwalker: just because sometimes I have super secret deploys that aren't on the calendar (security) [19:43:24] (03PS1) 10MarkTraceur: Merge repo info patch into extmeta patch [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90425 [19:43:33] sucheta: Is there a difference between .then( null, function () { .... } ); and .fail( function () { ... } ); ? I'm reading the jQuery docs but I can't find any differences offhand [19:43:53] what about the super super secret deploys! 0.o [19:43:59] we don't talk about those [19:44:07] tim just does them [19:44:25] sucheta: Ooooh I see it now, nm I'm an idiot [19:44:43] heh [19:45:07] greg-g: By day, he's a mediawiki developer...by night, he's STEALTH DEPLOY MAN [19:45:35] well, they're the same time, since he lives in Aus [19:46:11] (03Merged) 10jenkins-bot: Update browsertests for string change [extensions/Flow] - 10https://gerrit.wikimedia.org/r/89389 (owner: 10Spage) [19:46:45] :D [19:48:26] sucheta: That's a really clever use of .then() actually, I didn't know you could substitute promises like that [19:48:40] (03Merged) 10jenkins-bot: If $wgShowHostnames is true, add parsed by to NewPP limit report [core] - 10https://gerrit.wikimedia.org/r/89751 (owner: 10Reedy) [19:50:51] (03PS3) 10MarkTraceur: Add license to the header [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 [19:50:57] * marktraceur gasps for air [19:50:58] Git [19:51:00] Why [19:51:27] (03CR) 10MarkTraceur: "Rebased onto merge of two patches, links to file info page incoming." [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 (owner: 10MarkTraceur) [19:55:04] (03PS2) 10Dan-nl: corrected issue where batch upload stops after 9th image uploaded [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90147 [19:56:17] (03PS2) 10Cmcmahon: Fixed mailing list address [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90138 (owner: 10Zfilipin) [19:56:34] (03CR) 10Catrope: [C: 031] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [19:56:55] (03CR) 10Cmcmahon: [C: 032] "documentation" [qa/browsertests] - 10https://gerrit.wikimedia.org/r/90138 (owner: 10Zfilipin) [19:57:26] sucheta: Commented. The code looks awesome, you've taught me things about jQuery today :) just one minor thing about the way hasOwn is used and about the jsonp:false thing [19:57:44] hello [19:57:51] Salut, hashar [19:58:04] has anyone heard any complaint of tests falling over the last few hours ? [19:58:11] I updated most jenkins jobs earlier [19:58:25] I haven't [19:58:41] hashar: We would have seen walls of Jenkins -1 reports in here, so probably it's OK [19:58:56] yeah I bookmarked a gerrit search for verified - 1 : https://gerrit.wikimedia.org/r/#/q/status:open+label:Verified-1,n,z :] [19:59:05] marktraceur: unless he fnucked it up totally [19:59:12] so I guess I caught most issues before generalizing the deployment \O/ [19:59:53] Yay [20:00:04] hashar: I assume you went by the comic ツ [20:00:11] http://www.commitstrip.com/en/2013/10/11/mieux-et-moins-cher-que-les-tests-unitaires/ [20:00:53] RoanKattouw, *Blush* thank you. And yes, I meant to remove jsonp:false. Fixing that. [20:01:32] * RoanKattouw makes a mental note to look into refactoring the promises used in VE's initialization using .then() [20:01:45] Aww [20:02:08] Seriously it's like three nested promises that resolve each other or something [20:02:36] It seems like the kind of thing that's for [20:02:55] I think I do have $.when.apply() in there somewhere though [20:02:56] (03PS13) 10Reedy: Create maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [20:03:10] Yup: $.when.apply( $, promises ).done( this.modulesReady.resolve ).fail( this.modulesReady.reject ); [20:03:13] * RoanKattouw <3 JS [20:03:14] (03PS1) 10Hashar: gzip --force, to overwrite existing log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90428 [20:03:37] * YuviPanda makes RoanKattouw rewrite all the JS stuff in GWT [20:04:21] * RoanKattouw bangs his head against the wall for every time GWT makes him write !important [20:04:21] (03PS1) 10Hashar: gzip --force, to overwrite existing log files [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90429 [20:04:26] BANG-important [20:04:44] (03PS14) 10Reedy: Create maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [20:05:18] Hi, all...! Does anyone know what tools are used, if any, for checking the consistency of WP databases on a regular basis? [20:05:29] (03CR) 10Hashar: [C: 032] gzip --force, to overwrite existing log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90428 (owner: 10Hashar) [20:05:49] WP databases? [20:05:50] RoanKattouw: i wonder if insurance covers GWT related injuries [20:05:56] AndyRussG: None AFAIK [20:06:01] (03Merged) 10jenkins-bot: gzip --force, to overwrite existing log files [integration/jenkins] - 10https://gerrit.wikimedia.org/r/90428 (owner: 10Hashar) [20:06:18] (03PS2) 10Hashar: gzip --force, to overwrite existing log files [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90429 [20:06:20] Reedy: None? [20:06:48] Yup [20:07:21] (03CR) 10Hashar: [C: 032] gzip --force, to overwrite existing log files [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90429 (owner: 10Hashar) [20:07:26] YuviPanda: If I am physically in the office I guess CA worker's comp law might apply [20:07:35] Work-related injuries and all :) [20:08:32] RoanKattouw: hehe :P [20:09:33] Nothing regularly checks over the tables to make sure relationships have the right arity and such? [20:10:14] Nothing at all does [20:10:20] RoanKattouw: which config var specifies the location of load.php (or is used to derive the location of load.php)? [20:10:26] $wgServer, anything else? [20:10:54] ori-l: $wgLoadScript [20:11:01] * ori-l facepalms. Thanks. [20:11:11] ori-l: Are you interested in configuring it, or in computing its value for programmatic use? [20:11:52] drdee is using vagrant and getting URLs that omit the port number [20:11:56] trying to figure out why [20:13:17] (03PS1) 10Reedy: Minor updates to populateRevisionSha1.php to use newer available features [core] - 10https://gerrit.wikimedia.org/r/90431 [20:13:34] (03CR) 10jenkins-bot: [V: 04-1] Create maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [20:14:07] (03Merged) 10jenkins-bot: gzip --force, to overwrite existing log files [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/90429 (owner: 10Hashar) [20:16:09] (03PS2) 10Reedy: Minor updates to populateRevisionSha1.php to use newer available features [core] - 10https://gerrit.wikimedia.org/r/90431 [20:17:43] So basically if there are some tables that may need a regular check, I can just use whatever tool I like? [20:18:20] legoktm: do you know how long it usually takes for an Echo notification to show up? [20:18:58] YuviPanda: Are you using the job queue? [20:19:08] legoktm: on the cluster, I mean [20:19:20] it should be instantaneous [20:19:23] ah, okay [20:19:31] the email might take a while [20:19:34] but web should be right away [20:19:35] legoktm: just drawing up mental plans for close-to-realtime-notifications on the app [20:19:45] :D [20:19:56] YuviPanda: ugh, why would the mediawiki module _require_ a server_url parameter and then not do anything with it? that's a very ugly change [20:20:43] (03PS15) 10Reedy: Update populateRevisionLength maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [20:21:58] ori-l: true, I think that patch was sitting there for so long that I didn't notice it. [20:21:59] sorry [20:21:59] (03CR) 10Anomie: "That is a compelling argument, I'll have to give it some thought." [core] - 10https://gerrit.wikimedia.org/r/90263 (owner: 10Legoktm) [20:22:58] (03CR) 10EBernhardson: [C: 032] Remove period from help, breaks the link and is inconsistent [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90388 (owner: 10Legoktm) [20:23:00] (03PS16) 10Reedy: Update populateRevisionLength maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [20:23:49] (03Merged) 10jenkins-bot: Remove period from help, breaks the link and is inconsistent [extensions/Flow] - 10https://gerrit.wikimedia.org/r/90388 (owner: 10Legoktm) [20:24:52] (03CR) 10Reedy: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/90431 (owner: 10Reedy) [20:24:57] YuviPanda: sorry, my power went out and i lost scrollback >.> [20:25:10] YuviPanda: do you set $forwarded_port in labs? [20:25:36] ori-l: nope, 'coz there's no port forwarding going on [20:26:38] (03CR) 10BryanDavis: [C: 032] "(1 comment)" [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90147 (owner: 10Dan-nl) [20:26:50] YuviPanda: so that's another wart: the mediawiki role specifies $server_url = "http://127.0.0.1:${::forwarded_port}" which doesn't work since forwarded_port is undefined, but it gets passed to a module that doesn't do anything with it [20:26:51] (03Merged) 10jenkins-bot: corrected issue where batch upload stops after 9th image uploaded [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90147 (owner: 10Dan-nl) [20:27:21] ori-l: that patch was definitely undercooked, and in hindsight i should've -1'd it for a while. [20:27:27] ori-l: let me fix it up, download gonna complete in 30s [20:27:32] (Virtualbox) [20:27:49] ok, i have half of a fix written, i'll put it up for you to review/amend [20:28:34] ok [20:28:40] (03Abandoned) 10Reedy: Improve method parameter type hints [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/83969 (owner: 10Reedy) [20:29:18] ori-l: I'll head to sleep in a couple of minutes (2AM keyboard cutoff), but I'll amend/review tomorrow when I wake up [20:29:30] (03Abandoned) 10Reedy: More function level type hints [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/84482 (owner: 10Reedy) [20:30:19] YuviPanda: k, good night [20:30:26] (03PS16) 10SuchetaG: mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [20:31:12] night [20:31:17] (03PS2) 10Reedy: Fixup some bits of code, parameter type documentation etc [extensions/DPLforum] - 10https://gerrit.wikimedia.org/r/47311 [20:32:11] (03PS3) 10Reedy: Fixup some bits of code, parameter type documentation etc [extensions/DPLforum] - 10https://gerrit.wikimedia.org/r/47311 [20:32:58] (03PS2) 10Reedy: Generate "$wgServer" from SiteConfiguration using wikiid (dbname) [extensions/SecurePoll] - 10https://gerrit.wikimedia.org/r/67586 [20:33:03] (03CR) 10Reedy: [C: 032] Generate "$wgServer" from SiteConfiguration using wikiid (dbname) [extensions/SecurePoll] - 10https://gerrit.wikimedia.org/r/67586 (owner: 10Reedy) [20:33:05] (03Merged) 10jenkins-bot: Generate "$wgServer" from SiteConfiguration using wikiid (dbname) [extensions/SecurePoll] - 10https://gerrit.wikimedia.org/r/67586 (owner: 10Reedy) [20:34:13] RoanKattouw,Poke Poke Poke. [20:34:40] (03Abandoned) 10Reedy: More minor cleanup and documentation hints [extensions/OpenID] - 10https://gerrit.wikimedia.org/r/53518 (owner: 10Reedy) [20:34:45] (03CR) 10Catrope: [C: 031] mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [20:35:18] sucheta: Looks great. I'll do a bit of testing before I +2 [20:35:19] (03PS2) 10Reedy: Update mpdf to 5.7.1 [extensions/Mpdf] - 10https://gerrit.wikimedia.org/r/83829 [20:35:24] (03CR) 10Reedy: [C: 032 V: 032] Update mpdf to 5.7.1 [extensions/Mpdf] - 10https://gerrit.wikimedia.org/r/83829 (owner: 10Reedy) [20:35:32] (03PS2) 10Reedy: Parameter type hints [extensions/Mpdf] - 10https://gerrit.wikimedia.org/r/83825 [20:35:38] (03CR) 10Reedy: [C: 032 V: 032] Parameter type hints [extensions/Mpdf] - 10https://gerrit.wikimedia.org/r/83825 (owner: 10Reedy) [20:39:34] RoanKattouw, Alright. [20:39:55] RoanKattouw, When are you reaching Poona, BTW? [20:40:33] I have tickets as of like 24h ago, let me see [20:40:40] (03PS4) 10Reedy: Rename ListUsers to AllUsers [core] - 10https://gerrit.wikimedia.org/r/83584 [20:41:06] I land at BOM on Saturday Nov 16 at 9:40pm [20:41:37] So, that plus however long it takes to get to Poona from there (2 hours?), late Saturday evening I guess [20:43:52] (03CR) 10GWicke: "I didn't check in detail, but if the syntax warnings are outside the code you touched then it should be fine to ignore them." [core] - 10https://gerrit.wikimedia.org/r/89141 (owner: 10Arlolra) [20:44:09] AaronSchulz: How is mwscriptwikiset different from foreachwikiindblist ? [20:44:26] AaronSchulz: Oh, I guess it uses the script from the correct version? [20:44:45] RoanKattouw: right [20:45:41] brion, RoanKattouw, AaronSchulz, ori-l: what is your opinion on s/input/wikitext/ and s/result/html/ by default in parserTests? (see https://gerrit.wikimedia.org/r/#/c/89141/) [20:46:07] gwicke: why change them? [20:46:27] brion: so that we don't need to process 'result' to 'input' any more [20:46:33] and so that we can add 'json' output as well [20:46:49] or 'pageprops', whatever the name will be [20:47:04] i don't understand the need to change existing section names to add new ones? [20:47:20] it is just a clean-up [20:47:28] the old names will continue to work [20:47:37] RoanKattouw, Oh, so you will be spending the night at the airport? [20:48:03] brion: that diff will test your browser :) [20:48:25] brion: with the PHP parser input and result made sense, but for Parsoid we test wt2html, html2wt, wt2wt and html2html [20:48:28] sucheta: I think for now the plan is to take ground transportation to Poona after landing [20:48:43] But we may also do an airport hotel, we haven't discussed that yet [20:48:50] gwicke: as far as i know, it'll still work to input a "result" and output an "input" [20:48:52] (I'm traveling together with James_F|Away and possibly rmoen ) [20:48:59] gwicke: it sounds sensible to me if the change can be made cleanly without complications, but i've never touched that file, so consider me a gerrit +0.5 [20:49:12] it's a giant diff for no benefit that i can see [20:49:23] and may complicate peoples' maintenance of extension tests [20:49:31] anyway that' just my thought :) [20:49:38] Oh. I see. I shall be reaching on 17th afternoon. [20:49:42] i don't mind if other people want to deal with a fluctuating syntax [20:49:52] brion: we could delay switching the syntax in parserTests, and start using the new one for the tests we are adding [20:49:55] That should be roughly when I come back to life [20:50:02] It's a 15-hour flight from New York to Mumbai [20:50:04] we'd have a mix in the same file for a while [20:50:08] And I don't exactly live close to New York [20:50:16] gwicke: how does matches[1] work now? [20:50:27] (03CR) 10Brion VIBBER: "I see no benefit to doing this, but I won't object if everybody actively working on maintaining tests doesn't mind." [core] - 10https://gerrit.wikimedia.org/r/89141 (owner: 10Arlolra) [20:50:31] ah, maybe it's in the other file [20:50:32] (Another 5 hours and change from SF to New York) [20:50:52] Ouch. I know. This is usually pathetic. [20:51:14] gwicke: 'active working on maintaining tests' disqualifies me (appropriately), so I'm sitting it out [20:51:24] hehe [20:51:36] Then again that's not all that much worse than Israel was, I suppose [20:51:43] That was 4.5 hours to Toronto then 11 hours [20:52:10] it's mostly Parsoid folks editing that file these days [20:52:36] ...actually this would be quite a bit longer then. Oh well, I'm sure I'll live :) [20:55:10] (03PS1) 10Ori.livneh: Pass --server to install.php if $::forwarded_port is set [vagrant] - 10https://gerrit.wikimedia.org/r/90433 [20:55:29] (03CR) 10Subramanya Sastry: "When testing in Parsoid, the notion of input and result is not one-directional. Sometimes HTML is input and wikitext is the result. So, th" [core] - 10https://gerrit.wikimedia.org/r/89141 (owner: 10Arlolra) [20:56:00] (03CR) 10Ori.livneh: [C: 032] Pass --server to install.php if $::forwarded_port is set [vagrant] - 10https://gerrit.wikimedia.org/r/90433 (owner: 10Ori.livneh) [20:56:03] (03Merged) 10jenkins-bot: Pass --server to install.php if $::forwarded_port is set [vagrant] - 10https://gerrit.wikimedia.org/r/90433 (owner: 10Ori.livneh) [20:56:15] (03Abandoned) 10Hashar: jenkins test .. should fail. [core] - 10https://gerrit.wikimedia.org/r/90151 (owner: 10Hashar) [20:57:55] legoktm, around? [20:58:25] Eloquence: Hi [20:58:47] heya - is the version of MM with copyrightwarning deployed somewhere? I don't see it on test2, and can't easily replicate it locally [20:59:05] Eloquence: let me update the copy on my testwiki [21:00:38] Eloquence: http://legoktm.instance-proxy.wmflabs.org/wiki/Special:MassMessage login as "admin" pass: "test" [21:01:42] perfect, thanks [21:03:31] has there been any previous investigation into websocket daemons for push notification? essentially allowing client side javascript to subscribe to specific channels and pushing events (like say, 'article-edited' or whatever) to those channels? [21:03:43] (03CR) 10Catrope: [C: 032] mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [21:04:29] ebernhardson: I don't think we've done any web socket stuff in production for wikimedia yet [21:04:40] but i'm very interested in that working for a number of potential things [21:04:53] so give a shout if you do such research :D [21:05:09] IRC! [21:05:27] ebernhardson: werdna did something with xmpp a long time ago [21:05:29] for Echo [21:05:30] not sure [21:05:49] brion: it very likely will end up on the flow roadmap to at least investigate and propose, i'm just getting an idea now how much roadmap i should give it. sounds like alot :) [21:05:50] * YuviPanda|zzz puts 'websocket support in tooolabs back in his todo list... somewhere' [21:06:06] ebernhardson: haven't seen anyone doing websocket stuff with PHP :D [21:06:14] ebernhardson: you can do SSE with PHP tho, and it is halfway there [21:06:33] YuviPanda|zzz: random current idea, is node.js daemon and pushing messages to it via redis pub/sub [21:06:34] Daniel Kinzler did something with XML RC [21:06:36] YuviPanda|zzz: that is not an investigated idea, just a random one [21:06:47] the best kind of ideas [21:07:29] now to go sleep for real [21:07:29] (03Merged) 10jenkins-bot: mediawiki.api: add postWithToken and getToken [core] - 10https://gerrit.wikimedia.org/r/64582 (owner: 10Nikerabbit) [21:07:46] but we could also just write a js IRC client and let freenode suffer :P [21:12:21] (03CR) 10Dzahn: [C: 032] Sync custom template/en/custom/global/user-error.html.tmpl with upstream [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/90295 (owner: 10Aklapper) [21:12:30] (03CR) 10Dzahn: [V: 032] Sync custom template/en/custom/global/user-error.html.tmpl with upstream [wikimedia/bugzilla/modifications] - 10https://gerrit.wikimedia.org/r/90295 (owner: 10Aklapper) [21:12:58] Thank you RoanKattouw . ^_^ [21:13:04] greg-g: you know how I was wondering if I could do my deployment at 430 instead of at 3? can I go ahead and reinsert that request -- I'm seeing some very funny behaviour with the current setup that I'd like to test further before upsetting it [21:14:21] You're welcome :) [21:14:23] brion: around? bored? [21:14:39] or maybe RoanKattouw? https://gerrit.wikimedia.org/r/#/c/84315/ wants love :( [21:14:40] RoanKattouw: have you filled out the visa application yet ? [21:15:38] I have, yeah [21:15:58] You mean that wonderful on line form with maxlength set such that the address doesn't fit? [21:16:20] RoanKattouw: Yes, EXACTLY! [21:16:46] RoanKattouw: Which part of the address is the most important [21:16:47] ? [21:16:51] I don't know [21:16:54] I just did something [21:17:03] I also had to abbreviate the countries I'd visited [21:17:06] Essentially, my first ever patch on core, it was. :D [21:17:07] because there were too many [21:17:18] sucheta: Yay :) [21:17:29] sucheta: I wouldn't have guessed based on the code quality :) [21:17:38] RoanKattouw: Think it would go over well if i modified the maxlength property? [21:17:43] I don't know [21:17:47] I decided not to mess with that [21:18:01] property -> attribute [21:18:13] It didn't seem like a particularly robust or well-developed web app [21:18:29] It is terrible [21:19:06] adjust the HTML so the maxlength is just long enough and see if the server takes your intput? [21:19:06] sucheta: (by which I mean, the quality was high, you know what you're doing :) ) [21:19:14] legoktm: I was tempted to do that [21:19:27] But I have little faith in that very fragile-looking web ap [21:19:31] heheh [21:19:34] It was down for a couple days this week, too [21:19:51] And I wasn't exactly going to do cheeky things to the government that I'm hoping will give me a visa [21:20:18] mwalker: yes, good argument to do so [21:20:27] :) [21:20:48] RoanKattouw, :)) This makes me happy, thank you. [21:20:56] Ed and David were at the consulate in London today, they got crap for the invitation letter not stating the exact name of the place they were at correctly [21:21:40] Consulate General vs High Commission or whatever [21:22:50] RoanKattouw: I heard. Not fun. [21:24:06] It sounds like they were fine [21:24:26] I'm glad I won't have to spend 4 hours sitting in a coffee shop across the street from a consulate though [21:24:27] RoanKattouw: what did you put for the Phone number ? [21:24:39] AIUI we just have to deliver paperwork to that company and they handle it [21:24:41] rmoen: Phone number where? [21:24:52] RoanKattouw: on the form just under addresss [21:24:59] Address of what/whom? [21:25:12] They ask for multiple [21:25:35] RoanKattouw: I assume the Red Hat address under the fields labeled "Name of the Company in India*" [21:25:53] Oh, yeah [21:25:58] They gave us a phone number with that address too [21:26:01] It starts with +91 IIRC [21:26:04] * RoanKattouw looks it up [21:26:09] RoanKattouw: Yeah i dont have that [21:26:22] RoanKattouw: It's not on my invitation letter [21:27:11] * rmoen moves conversation to pm [21:27:11] rmoen: It should be in the WMF letter [21:30:43] why would you want to go to England? [21:32:36] AzaToth: Indian consulate in London :) so people who live in the UK getting a visa to go to India [21:33:28] ah [21:33:42] why would you want to go to India?? [21:33:52] ツ [21:33:54] But rmoen and myself live in the US so we're having a local company handle this stuff for us, apparently [21:34:16] * RoanKattouw has to actually get back to the US first, though, before giving his passport to consulate people [21:34:48] whats happening in india? [21:35:31] Some i18n thing [21:35:48] I'm not quite sure, I've managed to not be able to go every single previous time [21:35:57] hehe [21:36:20] so no l10n then I assume? [21:36:38] Probably that too [21:37:03] It's called the language summit I think [21:42:59] (03Draft2) 10Wikinaut: Bug 55763 [extensions/RSS] - 10https://gerrit.wikimedia.org/r/90441 [21:44:41] (03CR) 10Wikinaut: [C: 031 V: 031] Bug 55763 [extensions/RSS] - 10https://gerrit.wikimedia.org/r/90441 (owner: 10Wikinaut) [21:47:13] (03CR) 10Diederik: [C: 032 V: 032] "Looks great! ty for your swift response." [extensions/RSS] - 10https://gerrit.wikimedia.org/r/90441 (owner: 10Wikinaut) [21:48:05] (03PS7) 10MarkAHershberger: Include short descriptions for extensions bundled in the release [core] - 10https://gerrit.wikimedia.org/r/89340 [21:49:37] (03CR) 10Wikinaut: ":-))" [extensions/RSS] - 10https://gerrit.wikimedia.org/r/90441 (owner: 10Wikinaut) [21:50:28] (03Abandoned) 10MarkAHershberger: Fix up cite-desc message so it can be displayed without the parser. [extensions/Cite] - 10https://gerrit.wikimedia.org/r/89727 (owner: 10MarkAHershberger) [21:50:30] (03CR) 10Amire80: [C: 04-1] "The change in shared.css should be separate. It can affect a lot of other things, and needs to be tested with all the relevant browsers." [core] - 10https://gerrit.wikimedia.org/r/76714 (owner: 10Ebrahim) [21:51:10] (03Abandoned) 10MarkAHershberger: Fix up gadgets-desc message so it can be displayed without the parser. [extensions/Gadgets] - 10https://gerrit.wikimedia.org/r/89728 (owner: 10MarkAHershberger) [21:51:52] (03Abandoned) 10MarkAHershberger: Fix up interwiki-desc message so it can be displayed without the parser. [extensions/Interwiki] - 10https://gerrit.wikimedia.org/r/89729 (owner: 10MarkAHershberger) [21:52:43] (03Abandoned) 10MarkAHershberger: Fix up nuke-desc message so it can be displayed without the parser. [extensions/Nuke] - 10https://gerrit.wikimedia.org/r/89730 (owner: 10MarkAHershberger) [21:54:27] (03Abandoned) 10MarkAHershberger: Fix up renameuser-desc message so it can be displayed without the parser. [extensions/Renameuser] - 10https://gerrit.wikimedia.org/r/89731 (owner: 10MarkAHershberger) [21:57:00] (03PS4) 10MarkTraceur: Add license to the header [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90267 [21:57:51] (03CR) 10MarkAHershberger: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/89340 (owner: 10MarkAHershberger) [21:58:55] greg-g: bonjour. I'm gonna apply a small schema change in ten minutes... CentralNotice/patches/patch-campaign_throttle.sql [22:01:56] !g 89215 [22:01:57] https://gerrit.wikimedia.org/r/#q,89215,n,z [22:04:39] awight: ok [22:05:31] whoohoo! [22:05:51] (03PS2) 10Hashar: mediawiki-release [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/89215 [22:05:53] (03PS4) 10Hashar: tarball experiment [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/86160 [22:07:40] (03CR) 10Aaron Schulz: [C: 032] Add $wgEnableParserLimitReporting to control whether the NewPP limit report is included as a HTML comment. [core] - 10https://gerrit.wikimedia.org/r/89758 (owner: 10Reedy) [22:08:10] (03CR) 10jenkins-bot: [V: 04-1] Add $wgEnableParserLimitReporting to control whether the NewPP limit report is included as a HTML comment. [core] - 10https://gerrit.wikimedia.org/r/89758 (owner: 10Reedy) [22:08:41] Grrr [22:09:29] (03CR) 10MarkAHershberger: "Chad, The issue of serialized files should be addressed elsewhere separate since we're going to ship 1.22 with serialized files -- hopeful" [core] - 10https://gerrit.wikimedia.org/r/76663 (owner: 10MarkAHershberger) [22:13:22] (03PS4) 10EBernhardson: New hook accomidates non-revision rc querys [core] - 10https://gerrit.wikimedia.org/r/90051 [22:13:43] (03CR) 10EBernhardson: "Added documentation to patch and on mediawiki.org" [core] - 10https://gerrit.wikimedia.org/r/90051 (owner: 10EBernhardson) [22:14:03] (03PS2) 10Reedy: Add $wgEnableParserLimitReporting to control whether the NewPP limit report is included as a HTML comment. [core] - 10https://gerrit.wikimedia.org/r/89758 [22:14:04] (03CR) 10jenkins-bot: [V: 04-1] New hook accomidates non-revision rc querys [core] - 10https://gerrit.wikimedia.org/r/90051 (owner: 10EBernhardson) [22:14:51] (03CR) 10Bartosz Dziewoński: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/82100 (owner: 10Bartosz Dziewoński) [22:17:37] (03PS5) 10EBernhardson: New hook accomidates non-revision rc querys [core] - 10https://gerrit.wikimedia.org/r/90051 [22:19:32] (03CR) 10Aaron Schulz: [C: 04-1] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/54111 (owner: 10Reedy) [22:22:01] (03CR) 10Hashar: "I tend to dislike that hack. Cant we just skip loading the plural system whenever the XML file is not available?" [core] - 10https://gerrit.wikimedia.org/r/76663 (owner: 10MarkAHershberger) [22:22:11] (03CR) 10Hashar: [C: 04-1] Install without XML support [core] - 10https://gerrit.wikimedia.org/r/76663 (owner: 10MarkAHershberger) [22:23:16] (03PS17) 10Reedy: Update populateRevisionLength maintenance script to populate archive.ar_len [core] - 10https://gerrit.wikimedia.org/r/54111 [22:31:38] (03PS1) 10Reedy: Update Translate to master [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90446 [22:32:05] (03PS1) 10Reedy: Update Translate to master [core] (wmf/1.22wmf21) - 10https://gerrit.wikimedia.org/r/90447 [22:32:24] (03CR) 10Reedy: [C: 032 V: 032] Update Translate to master [core] (wmf/1.22wmf22) - 10https://gerrit.wikimedia.org/r/90446 (owner: 10Reedy) [22:32:36] (03CR) 10Reedy: [C: 032 V: 032] Update Translate to master [core] (wmf/1.22wmf21) - 10https://gerrit.wikimedia.org/r/90447 (owner: 10Reedy) [22:35:45] (03CR) 10Aaron Schulz: [C: 032] Add $wgEnableParserLimitReporting to control whether the NewPP limit report is included as a HTML comment. [core] - 10https://gerrit.wikimedia.org/r/89758 (owner: 10Reedy) [22:36:48] (03PS3) 10Aaron Schulz: Fix SQLite exception when $wgDBuser is set [core] - 10https://gerrit.wikimedia.org/r/90288 (owner: 10Tim Starling) [22:38:48] (03CR) 10BryanDavis: [C: 04-1] "(1 comment)" [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90149 (owner: 10Dan-nl) [22:39:07] (03PS6) 10EBernhardson: New hook accomidates non-revision rc querys [core] - 10https://gerrit.wikimedia.org/r/90051 [22:40:10] (03CR) 10Bsitu: "(1 comment)" [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/88223 (owner: 10Legoktm) [22:42:04] (03CR) 10Aaron Schulz: [C: 032] Fix SQLite exception when $wgDBuser is set [core] - 10https://gerrit.wikimedia.org/r/90288 (owner: 10Tim Starling) [22:42:12] AaronSchulz: [22:42:15] are too late [22:42:45] AaronSchulz: I was wondering whether we could write a test when using a sqlite backend that would set the $wgDBUser and assert the exception [22:43:06] (03Merged) 10jenkins-bot: Add $wgEnableParserLimitReporting to control whether the NewPP limit report is included as a HTML comment. [core] - 10https://gerrit.wikimedia.org/r/89758 (owner: 10Reedy) [22:48:14] (03Merged) 10jenkins-bot: Fix SQLite exception when $wgDBuser is set [core] - 10https://gerrit.wikimedia.org/r/90288 (owner: 10Tim Starling) [22:52:18] AaronSchulz: http://trotify.com/ [22:52:37] AzaToth: LEEEERROOOOYYY JENKIIIINNNNSSS [22:52:53] 'a wooden device that sits on your front wheel and makes your bike clop like a horse' [22:53:07] AzaToth: for the Jenkins debian jobs, I finally found a way to fetch the proper source artifacts :-]]]]]]]]] [22:54:40] AzaToth: would hack that tomorrow :-] good night! [22:54:45] * hashar is off [23:03:47] gwicke: haha [23:08:01] (03PS6) 10MaxSem: Make TOC hideable [core] - 10https://gerrit.wikimedia.org/r/80578 [23:12:19] (03CR) 10jenkins-bot: [V: 04-1] Make TOC hideable [core] - 10https://gerrit.wikimedia.org/r/80578 (owner: 10MaxSem) [23:30:55] (03PS1) 10EBernhardson: Remove 'Debug' hook from wfDebug and wfDebugLog [core] - 10https://gerrit.wikimedia.org/r/90453 [23:31:52] (03CR) 10Aaron Schulz: "I'm not terribly familiar with this code. Is there some way to have a flag flip the input/output as tests are run? Why do we need new type" [core] - 10https://gerrit.wikimedia.org/r/89141 (owner: 10Arlolra) [23:33:15] (03PS2) 10Dan-nl: corrected issue where batch upload breaks off early [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90149 [23:39:00] (03PS1) 10MarkTraceur: Only add handlers on image thumbnails [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/90455 [23:39:42] TimStarling: does https://gerrit.wikimedia.org/r/#/c/89097/1 seem sane? [23:39:59] (03CR) 10Dan-nl: "rebased the code and added further explanation to the commit message." [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/90149 (owner: 10Dan-nl) [23:43:03] yes, it seems fine to me [23:43:11] (03PS2) 10Tim Starling: Try to make sure jobs will not get pushed twice on failures [core] - 10https://gerrit.wikimedia.org/r/89097 (owner: 10Aaron Schulz) [23:43:17] (03CR) 10Tim Starling: [C: 032] Try to make sure jobs will not get pushed twice on failures [core] - 10https://gerrit.wikimedia.org/r/89097 (owner: 10Aaron Schulz) [23:45:52] TimStarling: that just leaves https://gerrit.wikimedia.org/r/#/c/69448/ blocking 49788 [23:47:30] (03PS1) 10Werdna: Add Wiki ID variable [extensions/AbuseFilter] - 10https://gerrit.wikimedia.org/r/90457 [23:48:24] (03Merged) 10jenkins-bot: Try to make sure jobs will not get pushed twice on failures [core] - 10https://gerrit.wikimedia.org/r/89097 (owner: 10Aaron Schulz) [23:49:12] TimStarling: anyway, there doesn't seem to be a way to configure custom PRAGMAs for sqlite [23:49:56] * AaronSchulz is curious how fast PRAGMA synchronous=NORMAL; PRAGMA journal=WAL would work [23:58:24] (03PS7) 10MarkTraceur: Put the username into the info list [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/86172 [23:58:48] (03PS2) 10MarkTraceur: Fetch extended metadata from the API [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/89103