[00:02:46] (03PS1) 10Andrew Bogott: Hosts created by addHost() are private. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/86208 [00:03:08] (03CR) 10Andrew Bogott: [C: 032] Hosts created by addHost() are private. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/86208 (owner: 10Andrew Bogott) [00:03:19] (03Merged) 10jenkins-bot: Hosts created by addHost() are private. [extensions/OpenStackManager] - 10https://gerrit.wikimedia.org/r/86208 (owner: 10Andrew Bogott) [00:05:49] huh. [00:06:08] either something's being cached, or i'm losing my mind [00:06:27] the little arrow seems to be correctly flipped on IE6 [00:06:31] (03PS6) 10Bsitu: Add primary/secondary link, email, bundling to post reply [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86064 [00:06:40] actually, that would kinda make sense [00:06:50] the fallback non-embedded url is janus-flipped [00:06:59] the primary embedded one isn't [00:07:16] so, that patch works better on IE6 that other browsers. hooray [00:08:01] (03CR) 10Bartosz Dziewoński: "The fallback appears to still work on IE6 – and funnily, it works even better than on new browsers, as the arrow is correctly flipped in R" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [00:09:44] (03CR) 10Jdlrobson: "OK fixing up now then merge?! :D" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [00:11:52] (03CR) 10Bartosz Dziewoński: "Ori filed bug 54673 about letting us just use the old @embed annotations in LESS. If it turns out to be easy to fix, I'd wait for it to av" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [00:14:50] hmmmmmm ori-l [00:15:00] couldn't we just run CSSJanus before LESS compiler? [00:15:20] all it does is apply some regex replacements, they should still work with LESS syntax [00:15:27] (hopefully, probably) [00:15:34] oh my god, of course [00:15:45] excuse while i go drown myself [00:16:04] that's just an idea, not sure it'll work [00:16:09] why wouldn't it? [00:16:10] not sure it'll work well enough for production* :D [00:16:15] because less does stuff? [00:16:18] i dunno [00:16:48] well, [00:16:50] you @import [00:17:01] so you'd need to make sure each file is transformed upon import [00:17:08] rather than only transform the root node [00:17:12] which wouldn't work, obviously [00:17:13] it doesn't have the crazy "font: {family:arial; size:15px}" syntax Sass has, it's pretty close to CSS [00:17:28] so it could work, maybe [00:17:38] variables would mess stuff up, though [00:17:54] (and mixins/functions too) [00:18:14] it would sorta kinda work as long as we don't try using either of these [00:18:24] yeah, it might not be dramatically simpler than what i suggested in the bug you cited above [00:18:43] the author of lessphp basically told us to ping him if we need anything merged upstream [00:18:52] and the re-ordering / deduping of comments is arguably a bug [00:19:05] so maybe that's still the better approach [00:19:28] ori-l: ping - issue with less embed mixin - if you run embed('image/foo-ltr.png') it doesn't get flipped in rtl mode - any ideas? [00:19:45] yeah, i think that fixing that in the compiler is a better idea after all [00:19:50] forget what i just said. :P [00:19:52] jdlrobson: That's what we've been discussing for the past hour :) [00:20:30] jdlrobson: you'll have to sit on your hands a few more days, sorry :/ [00:20:42] arhhfhfhhaz [00:21:06] want to summarize the current stance? [00:21:55] jdlrobson: unless someone comes up with something better, we need to fix lessc not to reorder our annotations. [00:22:11] (aka, https://bugzilla.wikimedia.org/show_bug.cgi?id=54673) [00:22:14] * jdlrobson starts to think [00:22:37] sure. LESS opens up some interesting possibilities with respect to embedding and flipping content, but those will take some time to scope out and develop, and we don't want to add technical debt by introducing additional syntax that we'd later have to deprecate [00:22:54] Ori suggested writing a less function for url(), which actually works and doesn't cause an infinite loop even though it outputs url(), but that still have the foo-ltr issue [00:22:58] so the short-term solution is to have flip/embed annotations in LESS handled correctly [00:23:27] RoanKattouw: it also needs to produce two rules [00:23:39] yeah, integrating our cssjanus with less sounds like a good idea too, but that'd probably take some time to get right [00:23:59] (that ugly mess of regexes really needs to go) [00:24:08] Yeah [00:24:12] less is an actual parser [00:24:16] (and having a real parser would go long way towards this) [00:24:22] Writing CSSJanus on top of an actual parser would be infinitely better [00:24:28] RoanKattouw: :S [00:24:30] :D [00:24:33] could you not have a global called @dir ? [00:24:34] jinx [00:24:36] yeah, I agree [00:24:40] and then do image-@dir.png [00:24:43] jdlrobson: You don't always want that [00:24:50] You'd need @antidir too, I guess [00:24:54] We do stuff like: [00:25:14] .leftarrow { background-image: url(arrow-ltr.png); } .rightarrow { background-image: url(arrow-rtl.png); } [00:25:34] Where by "leftarrow" and "rightarrow" really what I mean is "prevarrow" and "nextarrow" [00:26:05] hrm, @dir / @counterdir is viable [00:26:37] but i'd rather have a url() function (or equivalent) handle it [00:26:43] CSSJanus also offers flipping of foo-left to foo-right and v.v., but for some reason we never turned that on, only -ltr and -rtl [00:27:50] you'd probably want to do something like [00:28:21] .flip-image(@url) when (@rtl) { ... stuff .. "@url-rtl.png" } [00:28:39] or .flip-image(@url) when (direction() == 'rtl') { ... stuff .. "@url-rtl.png" } [00:28:46] ori-l: Does that mean you'd have to define everything twice? [00:28:54] Because that's the problem CSSJanus is designed to avoid [00:28:57] no, these would be mixins [00:29:02] Oh, I see [00:29:09] when() is like pattern matching in Haskell, kind of [00:29:16] almost exactly like, yes [00:29:20] they're even called 'guards', like in haskell [00:29:34] Except that multiple variants can execute whereas in Haskell it's the first match [00:29:42] yep [00:29:52] because that's like CSS selectors [00:29:59] Of course :) [00:30:00] I think the fact that you could approach this in multiple ways is a good reason to not try and solve it all before we have more LESS experience [00:30:11] and thus to support the existing annotations [00:30:25] Yes [00:31:06] sadly yes … :) [00:31:25] how best to decide this - should we get together and thrash this out sometime next week? [00:31:36] (or even better tomorrow :)) [00:31:49] bug 54673 is probably the way forward [00:32:26] ori-l: would this need an upstream change to less php? [00:32:34] or something we have complete control over? [00:32:46] yes, but we've already made several, and the author told us to ping him if we need anything merged [00:32:51] he's pretty excited about the mediawiki integration [00:35:52] sweet [00:36:05] ori-l: so are you currently owning getting that bug fixed or is someone else looking into it? [00:37:00] well, because i lack moral fiber, i'm probably going to not do anything for the next 24 hours, on the off-chance that brion's curiosity is tickled enough for him to fix it [00:37:09] if it isn't, i'll do it this weekend [00:37:40] :) [00:37:49] i'm kidding, happy to do it [00:37:58] Gotta get ori-l some Moralmucil [00:41:01] marktraceur: a while ago I was yelling at openoffice about a bug; and you told me to file a report against it -- so I did. they just got back to me and said that the version of LO I'm running has been EOL'd and so a fix will not be backported to it... however, the version of ubuntu I'm on has not been EOL'd; but they certainly are not going to update their version [00:41:09] whoooo [00:41:13] opensource! [00:41:42] mwalker: But they fixed it. [00:42:21] mwalker: Ubuntu 12.04 has mediawiki 1.15, if someone reported a bug in 1.15, we wouldn't even have the fucking conversation, that bug is set in stone now [00:42:32] sure [00:42:45] I think it's more a bitch about ubuntu's packaging than anything else [00:42:45] You upgrade to 1.19 like a sane person, or you don't get the bugfixes. [00:42:49] Indeed. [00:43:28] lol, I like the "Like a sane person" part of that [00:46:08] bawolff: Relatively sane. Compared to running 1.15. [00:47:00] Could be worse. They could be running 1.5 (I have seen people come in here, saying they are currently running 1.5 and asking if they should update) [00:47:21] Or that really old debian package where they moved all the css files to a non-web accessible location [00:47:35] Heh [00:51:44] (03CR) 10JGonera: [C: 04-1] "I think we should try to avoid adding too many custom modifications to lessphp. We could potentially think about adding some of those func" [core] - 10https://gerrit.wikimedia.org/r/85939 (owner: 10Ori.livneh) [00:52:39] (03PS1) 10Mgrover: Added Dependencies and Hygiene [tools/release] - 10https://gerrit.wikimedia.org/r/86210 [00:54:29] (03Abandoned) 10Mgrover: added additional logic to grep the output file for Story and Bug [tools/release] - 10https://gerrit.wikimedia.org/r/75280 (owner: 10Mgrover) [01:01:05] (03PS7) 10Kaldari: Migrating Mobile thanks into Thanks extension [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 [01:08:10] (03CR) 10Anomie: [C: 04-2] "If we were to do this, it would be better to just revert I0fbc798c and actually file (and WONTFIX) the bug that that fixed." [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/86207 (owner: 10CSteipp) [01:12:32] (03CR) 10Jdlrobson: [C: 031] "Feel free to +2 along with the other change." [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 (owner: 10Kaldari) [01:31:02] (03PS8) 10Kaldari: Migrating Mobile thanks into Thanks extension [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 [01:38:40] (03PS13) 10Mwalker: Further cleanup to Special:Version [core] - 10https://gerrit.wikimedia.org/r/65868 [01:42:20] marktraceur: you want to see something a bit scary? [01:42:22] (03PS9) 10Kaldari: Migrating Mobile thanks into Thanks extension [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 [01:42:35] Maaaaybe? [01:43:04] compare patchsets 10 and 11 in https://gerrit.wikimedia.org/r/#/c/65868 [01:43:19] pay special attention to the addition of includes/AutoLoader.php [01:43:34] it's worth saying that 10 to 11 was a rebase [01:44:03] OH MAN [01:44:04] but... apparently it rebased to include the child patch https://gerrit.wikimedia.org/r/#/c/66054/ [01:44:35] This is insane! [01:44:40] Oh [01:44:43] I know what happened [01:44:47] You fucked it up. [01:44:48] (03CR) 10Kaldari: [C: 032] "merging per jon" [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 (owner: 10Kaldari) [01:44:51] (03Merged) 10jenkins-bot: Migrating Mobile thanks into Thanks extension [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/86059 (owner: 10Kaldari) [01:44:52] * marktraceur helps [01:45:42] heh; clearly I screwed up; but how? [01:46:39] oh; 10 --> 11 was a manual rebase on my end; and I must have comitted the wrong thing [01:46:40] damn [01:47:17] Probably [01:47:28] mwalker: It seems like you could reasonably have rebased and gotten two copies of the commit [01:48:18] yarp [01:48:23] thoughts on how to undo it? [01:48:40] mwalker: git rebase -i HEAD~5, delete duplicates and/or reorder everything [01:49:35] (03PS1) 10Ori.livneh: Use interval Vagrant hook to determine VirtualBox version [vagrant] - 10https://gerrit.wikimedia.org/r/86212 [01:50:14] (03CR) 10Ori.livneh: "I need to determine what versions of Vagrant this approach is compatible with." [vagrant] - 10https://gerrit.wikimedia.org/r/86212 (owner: 10Ori.livneh) [01:51:16] (03PS14) 10Mwalker: Further cleanup to Special:Version [core] - 10https://gerrit.wikimedia.org/r/65868 [01:51:41] there we go [02:22:01] (03CR) 10Mwalker: "(13 comments)" [core] - 10https://gerrit.wikimedia.org/r/66054 (owner: 10Mwalker) [02:22:48] (03PS4) 10Mwalker: Allow MW and extensions to declare included components [core] - 10https://gerrit.wikimedia.org/r/66054 [02:23:20] (03CR) 10jenkins-bot: [V: 04-1] Allow MW and extensions to declare included components [core] - 10https://gerrit.wikimedia.org/r/66054 (owner: 10Mwalker) [02:30:55] (03Abandoned) 10CSteipp: Remove returnto for basic login test [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/86207 (owner: 10CSteipp) [02:39:43] (03PS3) 10EBernhardson (WMF): Better error handling for memc and some comments [extensions/Flow] - 10https://gerrit.wikimedia.org/r/83030 [02:40:36] (03PS4) 10EBernhardson (WMF): Better error handling for memc and some comments [extensions/Flow] - 10https://gerrit.wikimedia.org/r/83030 [02:45:24] (03PS5) 10EBernhardson (WMF): Better error handling for memc and some comments [extensions/Flow] - 10https://gerrit.wikimedia.org/r/83030 [02:57:13] (03PS8) 10EBernhardson (WMF): See that a post has been edited [extensions/Flow] - 10https://gerrit.wikimedia.org/r/84893 [04:26:28] ori-l: where does mediawiki/LocalSettings.php come from? [04:26:30] * YuviPanda looks around [04:26:41] ori-l: it sets $wgServer to 127.0.0.1:8080 explicitly [04:29:10] er [04:29:25] nevermind, found it [04:29:30] it's one of the last relics of the Old Way of Doing Things [04:29:58] ori-l: exec? [04:32:18] no, I think it's just checked into git with the values hard-coded [04:32:36] ori-l: nope, don't think so [04:32:49] ori-l: it uses install.php [04:33:03] well, I have to say [04:33:11] I'm rather pleased to not remember how it works :) [04:33:22] hehe :P [04:33:29] i've made a fix, testing it now [04:33:43] destory and up is still slow, but i suppose that's more because of my ageing Air [04:37:44] bd808: gah, not setting server url in install doesn't work :( [04:37:51] bd808: works for redirect, but not for RL [04:37:59] boo [04:38:37] what if we unset($wgServer) in the vagrant LocalConfig? [04:38:43] (03PS1) 10Yuvipanda: Do not explicitly set the server url [vagrant] - 10https://gerrit.wikimedia.org/r/86214 [04:38:45] bd808: ^ [04:38:48] doesn't work [04:38:50] as in [04:38:51] it works [04:38:54] but no styles or js [04:40:25] hmmm… but http://multimedia-bd808.instance-proxy.wmflabs.org/wiki/Main_Page works? [04:40:45] I just commented the $wgServer line out [04:40:55] bd808: yeah. [04:40:56] bd808: weird? [04:42:23] (03PS3) 10Yuvipanda: Report calls to deprecated methods as errors in PHP log [vagrant] - 10https://gerrit.wikimedia.org/r/75544 (owner: 10Ori.livneh) [04:42:28] (03PS1) 10Ori.livneh: Rename labsvagrant.rb -> labs-vagrant.rb [vagrant] - 10https://gerrit.wikimedia.org/r/86215 [04:45:15] (03PS2) 10Ori.livneh: Rename labsvagrant.rb -> labs-vagrant.rb; tidy code. [vagrant] - 10https://gerrit.wikimedia.org/r/86215 [04:45:38] (03CR) 10Nikerabbit: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/86172 (owner: 10MarkTraceur) [04:45:52] (03CR) 10Ori.livneh: [C: 032] Rename labsvagrant.rb -> labs-vagrant.rb; tidy code. [vagrant] - 10https://gerrit.wikimedia.org/r/86215 (owner: 10Ori.livneh) [04:45:55] (03Merged) 10jenkins-bot: Rename labsvagrant.rb -> labs-vagrant.rb; tidy code. [vagrant] - 10https://gerrit.wikimedia.org/r/86215 (owner: 10Ori.livneh) [04:46:10] YuviPanda: can you give labs-vagrant.rb a nice comment header? [04:46:20] will do, ori-l [04:46:24] also, documentation?! geez! :P [04:47:24] ori-l: doing the associated ops/puppet change now [04:47:37] thanks [04:49:08] bd808: I wonder if it is because HTTP_HOST isn't being set, for some weird reason? [04:49:12] and hence it just picks up localhost? [04:49:28] there are two LocalSettings.php files [04:49:43] but you know this already, since you're reading carefully the lines you are changing [04:50:08] indeed, one that is autogenerated by install.php, which has the wgServer line [04:50:12] and one in vagrant/ [04:50:21] which is hardcoded [04:51:46] I've gotta run. I'll get the patch for the apache enmod dependency in first thing tomorrow. [04:51:58] thanks for all the lessons tonight YuviPanda [04:52:08] bd808: night! [04:52:23] i'm filling my 'afternoons' when nobody else is awake with coursera [04:52:27] so it's all good [04:54:14] hi YuviPanda [04:54:18] hi Nikerabbit [04:54:38] (03CR) 10Tim Starling: [C: 04-1] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/71966 (owner: 10Aaron Schulz) [05:10:08] (03CR) 10Yuvipanda: [C: 04-1] "Doesn't seem to work yet, for unknown reasons." [vagrant] - 10https://gerrit.wikimedia.org/r/86214 (owner: 10Yuvipanda) [05:15:11] (03CR) 10Nikerabbit: [C: 04-1] "Is that line number correct? Currently this patch leads to double escaping in some of the code paths." [extensions/CleanChanges] - 10https://gerrit.wikimedia.org/r/86080 (owner: 10Siebrand) [05:16:56] (03PS2) 10Amire80: Document the block duration tooltip [core] - 10https://gerrit.wikimedia.org/r/74950 [05:20:49] YuviPanda|course: studying? [05:20:58] Nikerabbit: coursera [05:21:04] Nikerabbit: functional programming in scala :D [05:21:12] oh okay, nothing special :) [05:21:31] nope :P [05:21:34] no software engineering yet [05:33:24] (03CR) 10Aude: [C: 04-1] "(4 comments)" [core] - 10https://gerrit.wikimedia.org/r/66054 (owner: 10Mwalker) [05:37:41] (03PS4) 10Tim Starling: Support rate limiting backlink purges, including null edits [core] - 10https://gerrit.wikimedia.org/r/85358 (owner: 10Aaron Schulz) [05:37:49] (03CR) 10Tim Starling: [C: 032] Support rate limiting backlink purges, including null edits [core] - 10https://gerrit.wikimedia.org/r/85358 (owner: 10Aaron Schulz) [05:42:26] (03Merged) 10jenkins-bot: Support rate limiting backlink purges, including null edits [core] - 10https://gerrit.wikimedia.org/r/85358 (owner: 10Aaron Schulz) [06:03:48] ori-l, around? [06:13:06] yurik: what's up? [06:13:44] ori-l, i had some issues with fresh win7 vagrant installation (was quiting right after starting the machine), but after deleting and restarting with log=info mode, it seems to be working [06:14:08] will keep you updated of the progress with the latest fresh win7+vritualbox+vagrant :) [06:14:33] cool [06:15:15] yurik: could you try out a patch for me? [06:15:27] https://gerrit.wikimedia.org/r/#/c/86212/ [06:15:32] ori-l, sure, but its currently running an install [06:15:44] okay [06:16:14] one of the recent changes i made was move the apt package cache to the host's shared folder [06:16:23] so the .deb files persist after vagrant destroy [06:16:50] this makes re-creating a machine from scratch a lot faster -- typically less than 5 mins for a full run [06:16:54] ori-l, would it be possible to set up two vagrant machines to share it? [06:17:20] because this is exactly what's needed when working with two git MW repos [06:17:28] i mean - one MW git repo - two machines [06:17:51] not yet, sadly. you've asked for something similar for mediawiki a long time ago and i haven't forgotten about it, but it's a limitation of some early design choices that are hard to undo [06:18:10] yep [06:18:27] well, not hard, but annoying and unpleasant :P [06:18:28] it would be nice to use both the package cache and MW git repo [06:18:31] hehe [06:42:43] (03CR) 10Tim Starling: "If this is better than Services_JSON, why not upload it to PEAR as a new version of Services_JSON?" [core] - 10https://gerrit.wikimedia.org/r/78941 (owner: 10PleaseStand) [07:17:05] (03PS1) 10Aaron Schulz: Clear out UID files for unit tests [core] - 10https://gerrit.wikimedia.org/r/86218 [07:53:35] (03CR) 10Aaron Schulz: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/71966 (owner: 10Aaron Schulz) [07:54:32] (03PS5) 10Aaron Schulz: jobqueue: improved performance of JobQueueGroup::getQueuesWithJobs() [core] - 10https://gerrit.wikimedia.org/r/71966 [08:37:01] (03CR) 10Hashar: "(3 comments)" [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [08:42:25] (03CR) 10Krinkle: "The ot" [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [08:42:52] Krinkle|detached: hm? [08:45:17] (03CR) 10Ori.livneh: "her possibility is.." [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [08:53:01] (03CR) 10Krinkle: "Another reason I think this may be too early is that right now mediawiki.ui is not finished for general adoption. I don't know what the cu" [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [08:57:07] (03PS1) 10Hashar: return uniq list of extensions [tools/release] - 10https://gerrit.wikimedia.org/r/86220 [08:57:42] (03CR) 10Hashar: "(1 comment)" [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [08:59:44] (03CR) 10Ori.livneh: "Matt: could we abandon this patch and have you resubmit it with yourself as the author? This is too remote from what I'm currently working" [core] - 10https://gerrit.wikimedia.org/r/70740 (owner: 10Ori.livneh) [09:02:16] (03PS1) 10Hashar: standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 [09:02:20] (03CR) 10jenkins-bot: [V: 04-1] standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 (owner: 10Hashar) [09:02:28] (03CR) 10Hashar: [C: 032] standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 (owner: 10Hashar) [09:02:34] (03CR) 10jenkins-bot: [V: 04-1] standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 (owner: 10Hashar) [09:03:51] (03PS2) 10Hashar: standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 [09:04:06] (03CR) 10Hashar: [C: 032] standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 (owner: 10Hashar) [09:04:10] (03Merged) 10jenkins-bot: standardize return code and set exit code [tools/release] - 10https://gerrit.wikimedia.org/r/86221 (owner: 10Hashar) [09:05:03] (03PS2) 10Hashar: return uniq list of extensions [tools/release] - 10https://gerrit.wikimedia.org/r/86220 [09:06:38] (03CR) 10Hashar: [C: 032] return uniq list of extensions [tools/release] - 10https://gerrit.wikimedia.org/r/86220 (owner: 10Hashar) [09:06:42] (03Merged) 10jenkins-bot: return uniq list of extensions [tools/release] - 10https://gerrit.wikimedia.org/r/86220 (owner: 10Hashar) [09:08:23] (03PS10) 10Hashar: script cleanup and additional options [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:08:24] (03CR) 10jenkins-bot: [V: 04-1] script cleanup and additional options [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:12:25] (03PS11) 10Hashar: script cleanup and additional options [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:22:52] Does anyone know what "remote rejected ... change 77087 missing revisions" means? [09:23:04] I've seen "remote rejected", but I think the "missing revisions" part is new for me. [09:24:53] (03CR) 10Hashar: "rebased" [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:25:02] superm401: no idea :( [09:26:24] (03PS12) 10Hashar: script cleanup and additional options [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:26:48] (03CR) 10Hashar: [C: 032] "made options.gitroot all lower case :-D" [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:26:52] (03Merged) 10jenkins-bot: script cleanup and additional options [tools/release] - 10https://gerrit.wikimedia.org/r/81619 (owner: 10MarkAHershberger) [09:30:56] (03CR) 10Krinkle: "git-review doesn't coop well with a detached head (topic: "(detached")." [core] - 10https://gerrit.wikimedia.org/r/86154 (owner: 10Brion VIBBER) [09:35:17] (03PS4) 10Krinkle: Clean-up of skins/common/IEFixes.js [core] - 10https://gerrit.wikimedia.org/r/85544 (owner: 10Ori.livneh) [09:35:39] (03CR) 10Krinkle: [C: 032] "* Removed unused local variable 'hookit'." [core] - 10https://gerrit.wikimedia.org/r/85544 (owner: 10Ori.livneh) [09:36:33] (03PS5) 10Krinkle: Remove legacy scripts from .jshintignore [core] - 10https://gerrit.wikimedia.org/r/85543 (owner: 10Ori.livneh) [09:36:54] (03CR) 10Krinkle: [C: 032] Remove legacy scripts from .jshintignore [core] - 10https://gerrit.wikimedia.org/r/85543 (owner: 10Ori.livneh) [09:38:29] (03PS3) 10Krinkle: In OutPutPage->getJSVars, maintain symmetry between JS & PHP var names [core] - 10https://gerrit.wikimedia.org/r/86070 (owner: 10Ori.livneh) [09:39:56] (03Merged) 10jenkins-bot: Clean-up of skins/common/IEFixes.js [core] - 10https://gerrit.wikimedia.org/r/85544 (owner: 10Ori.livneh) [09:42:32] (03PS4) 10Krinkle: OutputPage->getJSVars: Maintain symmetry between JS & PHP vars [core] - 10https://gerrit.wikimedia.org/r/86070 (owner: 10Ori.livneh) [09:43:10] (03CR) 10Krinkle: [C: 032] "Fixed typo in subject (OutPutPage > OutputPage), and shortened it a bit to <70." [core] - 10https://gerrit.wikimedia.org/r/86070 (owner: 10Ori.livneh) [09:43:42] (03Merged) 10jenkins-bot: Remove legacy scripts from .jshintignore [core] - 10https://gerrit.wikimedia.org/r/85543 (owner: 10Ori.livneh) [09:45:56] (03CR) 10Krinkle: "@Brion: BrowserStack, SauceLabs, CrossBrowserTesting etc. We've got access to all of those, they boot faster than my VM, are easier to set" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [09:47:14] (03CR) 10Krinkle: [C: 04-1] "I was about to merge this last week (got caught up in something else), but now it mixes different things. It was already making a few chan" [core] - 10https://gerrit.wikimedia.org/r/83591 (owner: 10Jdlrobson) [09:49:50] (03Merged) 10jenkins-bot: OutputPage->getJSVars: Maintain symmetry between JS & PHP vars [core] - 10https://gerrit.wikimedia.org/r/86070 (owner: 10Ori.livneh) [09:55:24] (03CR) 10Siebrand: [C: 032] Namespace names for Macedonian (mk) [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/86226 (owner: 10L10n-bot) [09:56:31] (03Merged) 10jenkins-bot: Namespace names for Macedonian (mk) [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/86226 (owner: 10L10n-bot) [09:57:58] (03CR) 10Siebrand: [C: 032] Namespace names for Macedonian (mk) [extensions/EducationProgram] - 10https://gerrit.wikimedia.org/r/86228 (owner: 10L10n-bot) [09:58:06] (03Merged) 10jenkins-bot: Namespace names for Macedonian (mk) [extensions/EducationProgram] - 10https://gerrit.wikimedia.org/r/86228 (owner: 10L10n-bot) [09:59:57] (03PS1) 10Hashar: wrap functions in a class [tools/release] - 10https://gerrit.wikimedia.org/r/86229 [10:01:05] (03PS1) 10Hashar: typo gitRoot -> gitroot [tools/release] - 10https://gerrit.wikimedia.org/r/86230 [10:01:20] (03CR) 10Hashar: [C: 032] typo gitRoot -> gitroot [tools/release] - 10https://gerrit.wikimedia.org/r/86230 (owner: 10Hashar) [10:01:24] (03Merged) 10jenkins-bot: typo gitRoot -> gitroot [tools/release] - 10https://gerrit.wikimedia.org/r/86230 (owner: 10Hashar) [10:04:50] (03PS2) 10Hashar: wrap functions in a class [tools/release] - 10https://gerrit.wikimedia.org/r/86229 [10:49:07] superm401: "Missing revisions" means that your local git and the remote git server get confused about which of the two has which objects. [10:49:24] superm401: Did you try "git fetch" [10:50:07] qchris, it's https://gerrit.wikimedia.org/r/#/c/77087/ [10:50:15] I have the parent, 5433b7ec299e1beb7ef7672040709ff4bf9e8285, locally. [10:50:33] And I didn't do a rebase, so the amended version has the same parent. [10:51:04] I'll try a git fsck, then git fetch. [10:51:22] Ok. [10:53:00] qchris, git fsck only finds dangling commits, which I think are okay. [10:53:49] Fetching the change you gave gives me "fatal: Couldn't find remote ref refs/changes/87/77087/9" [10:53:50] qchris, git fetch does nothing. [10:54:11] qchris, so Gerrit's corrupt? [10:54:15] So probably something to broken for that change on gerrit's side. [10:54:26] Alright, I'll file it. [10:54:33] Ok. Thanks. [11:04:33] qchris, filed: https://bugzilla.wikimedia.org/show_bug.cgi?id=54688 [11:04:42] But while retesting the fetch started working... [11:04:56] (03PS10) 10Mattflaschen: Towards OB6: show a popup CTA to test onboarders [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/77087 (owner: 10Spage) [11:04:57] (03CR) 10jenkins-bot: [V: 04-1] Towards OB6: show a popup CTA to test onboarders [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/77087 (owner: 10Spage) [11:05:05] And now git review. [11:05:23] superm401: So it is working now? [11:05:33] Yeah, don't you hate when that happens? [11:05:37] superm401: (I did start to try to fix it in the meantime)So it is working now? [11:05:45] qchris, oh, good. :) [11:05:47] Yes [11:05:51] superm401: Cool. [11:05:59] That's the commit above I was trying to push the whole time. [11:06:31] Awesome. [11:26:12] (03CR) 10Siebrand: "(1 comment)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86064 (owner: 10Bsitu) [11:27:18] (03CR) 10Siebrand: [C: 04-1] "(2 comments)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86064 (owner: 10Bsitu) [11:48:44] (03PS1) 10Krinkle: index.php: Remove copyright line and use HTTPS url for mediawiki.org [core] - 10https://gerrit.wikimedia.org/r/86242 [11:56:04] (03PS1) 10J: Fix thumbnails for low framerate Ogg Theora Videos [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/86243 [11:56:49] ori-l: that comment was cut off because the textarea unfocused for some reason while I was typing the "h" in "The other" , and then "h", "e" and "r" triggered various hotkeys in Gerrit and submitted the comment.. [11:57:48] It does that too when adding reviewers, the dropdown menu for completion sometimes steals focus and then the next keystrokes are interpreted as hotkeys in the document itself [12:05:27] Krinkle: what bothers me most is the autocompletion in search; after every tab the submit button is selected, so I must write everything myself and tab-complete only the last part (e.g. the extension name but not mediawiki/extensions/) [12:07:00] Nemo_bis: In Gerrit search, "Enter" is what you think "Tab" does. [12:07:13] hitting Enter doesn't submit, it inserts the selected item from the dropdown autocomplete [12:07:23] like you think Tab would [12:07:47] and then, to make it "easier", if you hit Enter while not in a dropdown autocomplete (e.g. right after the previous scenario), then it does act as trigger for submit [12:08:07] so you can go project:med -> project:mediawiki/core , results :) [12:08:36] thanks, I didn't know about enter, but it doesn't make much sense to use it because it has unpredictable results [12:08:47] It's a bit counter-intuitive, but I can get used to it. I think they should change it, too. But atleast now you know :) [12:08:55] plus it still adds a space so I have to backspace and continue the selection [12:09:26] hence writing it myself is still faster in most cases... but maybe I'll find a use for that [12:09:29] the space adding doesn't affect the results, though, does it? Assuming you complete to the item you want, the next thing you type would have to be separated by a space anyway, so that kinda makes sense. [12:09:49] not if I autocomplete mediawiki and want to add /extensions [12:10:15] or some other subproject like that [12:17:23] (03PS1) 10Hashar: . is not for string concatenation [tools/release] - 10https://gerrit.wikimedia.org/r/86245 [12:17:50] (03CR) 10Hashar: [C: 032] . is not for string concatenation [tools/release] - 10https://gerrit.wikimedia.org/r/86245 (owner: 10Hashar) [12:17:54] (03Merged) 10jenkins-bot: . is not for string concatenation [tools/release] - 10https://gerrit.wikimedia.org/r/86245 (owner: 10Hashar) [12:23:21] (03PS1) 10Krinkle: less: Update to upstream 6e8e724fc7 [core] - 10https://gerrit.wikimedia.org/r/86246 [12:28:54] (03PS1) 10Hashar: remove unused hashfile() [tools/release] - 10https://gerrit.wikimedia.org/r/86247 [12:29:55] (03PS1) 10Krinkle: less: Add '@since 1.22' comments [core] - 10https://gerrit.wikimedia.org/r/86248 [12:29:56] (03PS1) 10Krinkle: less: Use new addParsedFile method, replacing embeddedFiles hack [core] - 10https://gerrit.wikimedia.org/r/86249 [12:40:33] hashar: Is there some documentation of what needs to be done to have jenkins-bot verify and comment back commits for java projects? [12:42:24] qchris: I got some doc on the wiki [12:42:32] qchris: let me find a page worht a read [12:42:40] I could not find it on wikitech .-/ [12:43:21] ahh no it is still on mw.org [12:43:22] https://www.mediawiki.org/wiki/Continuous_integration/Jenkins_job_builder [12:43:47] Ok. Thanks. I'll go through that then. [12:43:47] most jenkins jobs are build by a python script named Jenkins Job Builder (abbreviated JJB) [12:44:06] you would define the jobs using a YAML definition in integration/jenkins-jobs-builder-config.git [12:44:14] then run the script to refresh the jobs [12:44:19] Cool. [12:44:24] another way is to create the job using the GUI [12:44:33] and migrate it to a YAML definition later on when you are happy with it [12:44:55] that saves you from having to learn YAML / jjb [12:44:56] The GUI allows to setup commenting back to gerrit? That's even better :-) [12:45:03] not really :-] [12:45:14] jenkins is basically a repository of jobs [12:45:19] most of them are not running on their own [12:45:22] they need to be triggered [12:45:34] which is done by Zuul. It is a daemon listening for Gerrit events (stream-events) [12:45:57] that has a workflow defined in (integration/zuul-config) to determine jobs that would be triggered for a given event [12:46:09] I know how to setup a Java project inJenkins. I guess I'll do just that and ping you next week on how to get the "commenting back to gerrit" to work. [12:46:17] The Zuul part is barely documented though :( [12:46:42] I think I got a maven job defined in JJB let me find it [12:47:00] I've done maven with Jenkis before. [12:47:03] question. when i go to https://gerrit.wikimedia.org/r/#/c/84901/ which changes just files in templates/apache and then i want to compare the diff between patch set 2 and patch set 3, so i go and click "Reference version: patch set 2" and ..all of a sudden all those unrelated files show up as being changed. ? [12:47:26] hashar: Thanks :-) [12:48:09] !gitweb integration/jenkins-job-builder-config [12:48:09] Don't use gitweb anymore, use gitblit: https://git.wikimedia.org/ [12:48:13] !git integration/jenkins-job-builder-config.git [12:48:13] MediaWiki development is using git, a distributed source control manager, starting on March 21st, 2012; details: https://www.mediawiki.org/wiki/Git_happens Instructions for using it: https://www.mediawiki.org/wiki/Git/Workflow To get an account: https://www.mediawiki.org/wiki/Project:Labsconsole_accounts [12:48:17] !repo integration/jenkins-job-builder-config.git [12:48:17] There is no such key, you probably want to try: !cia, !details, !download, !dpl, !errors, !extensiondistributor, !faq, !foreignfilerepos, !gitrepo, !newgit, !passthru, !reportbug, !reports, !svnprops, !newbug, [12:48:20] pff [12:48:23] !gitblit integration/jenkins-job-builder-config.git [12:48:24] https://gerrit.wikimedia.org/r/plugins/gitblit/ [12:48:30] rage [12:48:38] :-)) [12:48:39] !gitrepo integration/jenkins-job-builder-config [12:48:39] To request a new Git repository, use the "New repositories" page on mediawiki.org [12:48:43] !gitrepo integration/jenkins-job-builder-config.git [12:48:43] To request a new Git repository, use the "New repositories" page on mediawiki.org [12:48:46] :p [12:48:58] @del gitweb [12:50:00] !key del gitweb [12:50:00] Unable to find the specified key in db [12:50:14] !gitblit del [12:50:15] Successfully removed gitblit [12:50:21] !gitblit is https://git.wikimedia.org/summary/?r=$1 [12:50:21] Key was added [12:50:26] (03PS5) 10saper: Fix exception on collection save [extensions/Collection] - 10https://gerrit.wikimedia.org/r/84435 [12:50:30] !gitblit integration/jenkins-job-builder-config.git [12:50:30] https://git.wikimedia.org/summary/?r=integration/jenkins-job-builder-config.git [12:50:39] !gitweb del [12:50:39] Successfully removed gitweb [12:50:43] (03CR) 10saper: "Fixed white space issues, thanks." [extensions/Collection] - 10https://gerrit.wikimedia.org/r/84435 (owner: 10saper) [12:50:44] !gitweb alias gitblit [12:50:44] Created new alias for this key [12:50:46] !gitweb integration/jenkins-job-builder-config.git [12:50:47] https://git.wikimedia.org/summary/?r=integration/jenkins-job-builder-config.git [12:50:50] sorry for d4 spam [12:51:19] hashar: I liked the "spam". It showed me howe to use mw-bot. Thanks :-) [12:51:28] qchris: an example of a maven job defined in JJB https://git.wikimedia.org/blob/integration%2Fjenkins-job-builder-config.git/master/mobile.yaml [12:51:31] * qchris learns something new every day. [12:51:53] qchris: there is some doc at https://meta.wikimedia.org/wiki/Wm-bot#Infobot [12:52:01] qchris: I usually search for 'wm-bot' in google, that is the first link [12:52:02] hashar: Thanks. [12:52:11] hashar: And thanks again :-D [12:52:27] qchris: there is a few oddities in how we configure Jenkins jobs. You will need a bunch of parameters that are passed by Zuul [12:53:14] qchris: the git conf is a bit weird as well. You want to fetch from git://integration.wikimedia.org/ [12:53:25] Ok. [12:53:28] qchris: and set a refrence repository: /srv/ssd/zuul/git/ [12:54:00] If you feel brave, you can try using JJB, else do the job in jenkins GUI and we can have a look at how we can migrate it to JJB next week :) [12:54:14] it is not rocket science, but probably needs a couple hours to get started [12:54:15] hashar: /me is not too brave. [12:54:33] hashar: I'll just annoy you with my failures next week then. [12:54:52] sounds good :) [12:55:36] (03CR) 10Hashar: [C: 04-2] "some methods should be global functions since they do not need a context." [tools/release] - 10https://gerrit.wikimedia.org/r/86229 (owner: 10Hashar) [13:13:17] (03CR) 10Hashar: [C: 032] remove unused hashfile() [tools/release] - 10https://gerrit.wikimedia.org/r/86247 (owner: 10Hashar) [13:13:21] (03Merged) 10jenkins-bot: remove unused hashfile() [tools/release] - 10https://gerrit.wikimedia.org/r/86247 (owner: 10Hashar) [13:13:33] easy approval from the CLI : gerrit approve --code-review +2 f4b48b906 [13:13:34] :) [14:13:47] (03CR) 10Bartosz Dziewoński: [C: 04-1] "(1 comment)" [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/82092 (owner: 10Helder.wiki) [14:18:38] (03CR) 10Siebrand: [C: 032] Add Portuguese translations (pt and pt-BR) [extensions/EducationProgram] - 10https://gerrit.wikimedia.org/r/82087 (owner: 10Helder.wiki) [14:18:46] (03Merged) 10jenkins-bot: Add Portuguese translations (pt and pt-BR) [extensions/EducationProgram] - 10https://gerrit.wikimedia.org/r/82087 (owner: 10Helder.wiki) [14:19:43] (03CR) 10Siebrand: [C: 031] "L10n/i18n reviewed." [extensions/Collection] - 10https://gerrit.wikimedia.org/r/84435 (owner: 10saper) [14:20:38] (03CR) 10Siebrand: [C: 032] index.php: Remove copyright line and use HTTPS url for mediawiki.org [core] - 10https://gerrit.wikimedia.org/r/86242 (owner: 10Krinkle) [14:20:49] hey jarry1250 [14:21:04] YuviPanda: If you have big tools, catscan, or the wlmapi, or geohack, I can see it makes sense [14:21:44] YuviPanda: But if you have smaller tools, I guess there's now an incentive to bundle them arbitarily [14:21:58] into, say, "Jarry's Microtools" [14:22:02] jarry1250: I could see how, yeah. [14:22:28] jarry1250: I think personally I'm biased because I never really had tools on toolserver, and this approach of libraries / dependencies matches what I've done elsewhere doing things in 'production' [14:22:49] jarry1250: my tools are mostly python, so it is trivial to just put common libraries into pip and source them into the virtualenv [14:23:26] (03Merged) 10jenkins-bot: index.php: Remove copyright line and use HTTPS url for mediawiki.org [core] - 10https://gerrit.wikimedia.org/r/86242 (owner: 10Krinkle) [14:24:26] (03CR) 10Matthias Mullie: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson (WMF)) [14:24:38] YuviPanda: Yeah, for scripts I can see that -- but then, I hardly imagining a botop who runs 50 tasks from the TS converting them into 50 projects [14:24:56] jarry1250: indeed, and that's what Cyber has done, IIRC. [14:25:04] jarry1250: just macro groupings of related bots [14:25:22] jarry1250: I'm unsure what legoktm does. [14:25:23] Yes. There's a decision to be made. [14:25:34] jarry1250: I think toollabs is flexible enough in that regard. [14:26:01] YuviPanda: Oh, it's always been flexible, that's never been the issue for TS users re: the migration [14:26:03] jarry1250: however, that kinda 'macro' tools are bad, since it makes making things truly 'multi-tenant' hard [14:26:33] Well, you use the granularity of permissions, certainly [14:26:57] you can add people only to tools, not much lesser [14:27:15] Sorry, *lose [14:27:21] ah [14:27:22] :D [14:27:23] ok [14:28:10] (03CR) 10Siebrand: [C: 031] "i18n/L10n LGTM." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/84893 (owner: 10EBernhardson (WMF)) [14:28:27] jarry1250: actually, I just remembered somethin [14:28:34] But the more I think about, the more that isn't really a cost, because I either trust someone to mess around with my code or I don't. Maybe another cost to do with the tools directory lists as well? [14:28:48] jarry1250: a month or so ago, we added the feature for tool accoutns to be added to other groups' tool accounts [14:29:09] jarry1250: so you can do something like create a 'jarry-common' tool, and grant specific access to that account for other accoutns [14:29:18] it's not particularly well documented yet, but it exists [14:29:32] YuviPanda: Oooh, interesting. That's not bug #51990 is it? [14:29:37] link? [14:29:40] !b 51990 [14:29:40] https://bugzilla.wikimedia.org/51990 [14:29:55] (I'm sorry, these things are not my speciality) [14:29:59] jarry1250: aha! it is [14:30:04] (03PS24) 10Physikerwelt: Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 [14:30:09] jarry1250: and it was merged a while back [14:30:21] (03CR) 10jenkins-bot: [V: 04-1] Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 (owner: 10Physikerwelt) [14:30:28] Status: REOPENED? [14:30:36] YuviPanda: But all in all, I like that paradigm [14:30:51] jarry1250: because Ryan Lane misunderstood it and closed it, then we explained to him and then he reopened it [14:31:39] hexmode: around ? :) [14:31:41] jarry1250: indeed! needs documentation [14:31:57] hashar: for you? yes! [14:32:02] ;) [14:32:08] hexmode: I am heavily rewriting make-release.p [14:32:09] y [14:32:22] hashar: saw that, thank you [14:32:25] hexmode: I moved the function to a class and refactoring a bit [14:32:32] hexmode: I might end up breaking some parameters [14:32:54] hexmode: I also want to make it fast by avoiding to clone the repo several time and use a local repo to get files from [14:33:00] (03CR) 10Matthias Mullie: "(1 comment)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/83056 (owner: 10Matthias Mullie) [14:33:03] YuviPanda: I'll have a go at that later. On the one hand, I know precious little about service groups, etc, but on the other I'm hardly alone in that [14:33:09] hashar: makes sense [14:33:16] ...when it comes to TS developer. [14:33:20] YuviPanda: Thanks! [14:33:25] hashar: I was doing that by using gitrepo on cmdline [14:33:31] hexmode: basically, the script could be passed a local repo (equivalent of --git-root) and then use git-archive to quickly fetch all the files of a given commit / tag [14:33:40] I need that for jenkins [14:33:42] :D [14:33:47] sure [14:33:49] jarry1250: heh, yeah. just literally translate 'service groups' to 'multimaintainer tool accounts' and it should mostly work, ,I think [14:34:01] hexmode: also I have created the Jenkins part that would let us fetch a tag whenever a new tag is made [14:34:04] jarry1250: although I don't fully understand toolserver mechanisms, so I could be wrong [14:34:23] \o/ [14:34:31] hexmode: would reply to your email later tonight or tomorrow. Sorry :( [14:34:41] next [14:34:46] np... these details are good [14:34:48] rewrite wmf-make-branch to python *evil grin* [14:34:52] + tests [14:34:53] ! [14:34:53] ok, you just typed an exclamation mark with no meaning in the channel, good job. If you want to see a list of all keys, check !botbrain [14:35:01] ! del [14:35:02] !! del [14:35:02] Successfully removed [14:35:03] Unable to find the specified key in db [14:35:06] heh :D [14:35:12] ! [14:35:12] There is no such key, you probably want to try: !::1, !;, !?>, !@, !@param, !^demon, !1.17, !1.17_tracking, !1.17wmf1, !1.18, !1.18wmf1, !1.19, !1.19wmf1, !2.0, !3x, !42, !a, !abusefilter, !access, !ack, !actionpaths, !actions, !adminsettings, !ajax, !allmessages, !allowoverride, !amend, !ampersand, !analytics, !ancient, !anglebrackets, !announce, !annoyinglittlebug, !anonnotice, !answer, !api, !around, !audit, !auth, !authentication, !autoblock, !autoconfirmed, !autoload, !autothumb, !backup, !bellthecat, !berlin, !bestcase, !blacklist, !blame, !blank, !bleedingedge, !blocking, !boilerplate, !bom, !bot, !botbrain2, !botnoob, !branch, !branchpoints, !browser, !bugsearch, !bugzilla, !bunching, !bureaucracy, !bzapi, !bzsource, !cache, !captcha, !cat, !categories, !centralauth, !chansearch, !chanspam, !charinsert, !cheatsheet, !checkuser, !ci, !cia, !cite, !class, !cloak, !cms, !code, !codereview, !collapse, !colon, !commit, !commons, !compare, !compression, !config, !console, !cookies, !cooluris, !copyhelp, !copying, !copyinghelp, !copytemplates, !core, !corehack, !corn, !createandpromote, !createpage, !credit, !cron, !css, !csshideheader, !csv2wp, !damneditcount, !db, !dbtable, !debbug, !debianubuntu, !debug, !defaultsettings, !defenestrate, !deleteuser, !deploymentqueue, !deployments, !details, !developer, !developers, !devsecurity, !diff3, !dismissablesitenotice, !displaytitle, !documentation, !doesntwork, !domdocument, !downgrade, !download, !doxygen, !dpl, !draft, !dump, !dumphtml, !easy, !echo, !editbuttons, !editing, !edittoken, !email, !emailconfirmed, !enableapi, !enotif, !enwp, !errors, !es, !etherpad, !events, !evilglobals, !exclamation, !export, !exporttemplate, !extensiondistributor, !extensionmatrix, !extensions, !externalauth, !externaleditor, !externalimages, !extranamespace, !f5, !fail, !false, !faq, !feature, !feeds, !file, !filecache, !filepermissions, !filesize, !filetype, !fileurl, !firebug, !flaggedrevs, !flash, !footer, !foreignfilerepos, !fu, !gadget, !gadgets, !gallery, !gerritapi, !gerritdocs, !gerriterror, !gerrithelp, !gerritkeys, !gerritsearch, !gethelp, !gfi, !git, !git-access, !gitaccount, !gitauthor, !gitcheat, !gitemail, !gitrepo, !git-review, !giturl, !gnu, !google, !googleanalytics, !googleit, !grep, !gsocapp, !gsod, !hacker, !HAL, !hdss, !hello, !help, !helpfor, !hetdeploy, !hiddencategories, !hidetabs, !hidetoc, !highlight, !hooks, !hosting, !howto, !html, !html9, !humor, !i18n, !iamabot, !icon, !ie6, !IE9, !iframe, !iis, !iis404, !imabot, !imageauth, !imagelink, !imagemagic, !imagemap, !images, !img_auth, !import, !important, !importhelp, !importimages, !importsources, !info, !infobox, !inputbox, !install, !installlinux, !installmacosx, !instantcommons, !interface, !internalapi, !interwiki, !invalidupload, !is, !itsalwaysbugzilla, !javascript, !jobqueue, !jquery, !labs, !langlinks, !lazy, !ldap, !ldapdebug, !ldapsupport, !line, !links, !list, !listinfo, !lists, !lmgtfy, !localhost, !lockdown, !logo, !lowercase, !lsb, !lst, !lua, !lucene, !magicwords, !mailarchive, !mainpage, !mainpagetitle, !man, !manapi, !maps, !math, !maxlag, !mediawiki, !meetings, !memcached, !memory, !meta, !migrate, !mime, !moderation, !moin, !monday, !moo, !morbo, !move, !mssql, !multiplecommits, !multiplewiki, !mwbot, !mw-config, !mwdumper, !mwhelp, !mysql, !namespaces, !networkauth, !newgit, !newusergroup, !newwindow, !next, !nightly, !nofollow, !noob, !noshell, !nosql, !not, !notes, !nowiki, !nuke, !nullpath, !numberedparams, !ogg, !omgevilbug, !openid, !opw, !oracle, !oversight, !packagefix, !padindex, !pagecss, !paranoid, !parser, !parserconf, !parserfunctions, !passthru, !passwordsql, !patch, !pdhelp, !performance, !petan, !phase3, !php, !php4, !php531, !phpinfo, !phpmyadmin, !phpshell, !phpsucks, !ping, !pipetrick, !plainlinks, !player, !poem, !ppp, !prefsrant, !preventaccess, !profiling, !protect, !protocols, !punehackathon, !purge, !putty, !pywikipediabot, !query, !queue, !rangeblocking, !rawr, !rcbot, !readonly, !rebase, !recaptcha, !redirect, !redlinks, !refresh, !releasedate, !renameuser, !repeat, !reportbug, !reports, !requestcontext, !research, !resetpass, !resetpassword, !resourceloader, !rest, !revdelete, !revert, !reviewqueue, !rfc, !rights, !rightssql, !robots, !rollback, !rt, !rtam, !rtfm, !ruby, !safemode, !sal, !sandbox, !sarcasm, !sb, !scap, !scarytranscluding, !schema, !search, !secrets, !security, !selfmerge, !selinux, !semantichelp, !semanticmediawiki, !sequel, !serversuperglobal, !session, !sharedlogins, !shell, !shellrequest, !shorturls, !sitemap, !sitenotice, !siwae, !sm, !smtp, !smw, !sofixit, !songs, !sop, !sortkey, !sorttable, !sourceforge, !spam, !specialpages, !sqlite, !sqllog, !sqlsearch, !srv193, !start, !stfu, !stringfunctions, !stroopwafels, !subpages, !subst, !suggest, !sul, !sumanah, !support, !supported, !svg, !svnauthor, !svnprops, !svn-rev, !svnsearch, !syntax, !syntaxhighlight, !sysadmin, !t_paamayim_nekudotayim, !tabcompletion, !table, !tables.sql, !tabs, !tag, !tautology, !telepathy, !temp, !templates, !test, !test2, !testing, !tests, !testswarm, !testwiki, !thankyou, !the_mask, !thisbotisnotstupid, !thisisalongkey, !threads, !thumb, !tias, !tidy, !tidyforidiots, !timezone, !toc, !toolserver, !tooquiet, !transclusion, !translatewiki, !true, !ts, !tso, !tss, !tunemysql, !twn, !ui, !umask, !undo, !unicode_test, !uninstall, !unreviewedcode, !update.php, !upgrade, !upload, !uploadbyurl, !useful, !useragent, !userapproval, !variables, !vector, !version, !versionlifecycle, !viewvc, !votes, !vulnerability, !wallofshame, !webinspector, !webrequest, !webscale, !welcome, !wf, !wg, !wiki, !wiki2pdf, !wiki2xml, !wikia, !wikiacode, !wikibugs, !wikidata, !wikieditor, !wikimania, !wikimedia, !wikipmediawiki, !wip, !wmf, !wmf-deployment, !wmo, !worstcase, !wptemplates, !wtf, !wysiwyg, !xampp, !xkcd, !xml, !yoda, !sanitizer, !index.php, !wikipedia, !pastelocalsettings, !flow, !opsschool, !vagrant, !packages, !subversion, !1.21, !1.20, !1.22, !patrol, !wm-bot, !wp, !memoserv, !submodules, !cs, !5.4, !$wgServer, !visualeditor, !master, !unicorn, !rewriteproblems, !parsoid, !pong, !adsense, !github, !sourcecode, !technews, !reboot, !hook, !seen, !tryflow, !sidebar, !htmlform, !bzqs, !roadmap, !apiinternal, !apiclient, !pastebin, !dirtydiffs, !parsoid-web, !parsoid-test, !link, !smart, !autopromote, !vebug, !VE-table, !veq, !maintenance, !bugsong, !ext, !tea, !ask, !jobs, !farm, !newbug, !templateproblems, !mw, !blink, !htmllogs, !logs, !hss, !jsduck, !packagessimple, !skins, !skinning, !job, !codesearch, !ie, !bribe, !gitblit, [14:35:24] OH MY GOD [14:35:55] !1.19wmf1 [14:35:55] Branched on 10 February 2012 at r111189 from REL1_19 [14:35:57] !1.19wmf1 del [14:35:58] Successfully removed 1.19wmf1 [14:36:01] !1.19 [14:36:02] 1.19.0 released 02 May 2012 http://http://lists.wikimedia.org/pipermail/mediawiki-announce/2012-May/000115.html [14:36:04] !1.19 del [14:36:05] Successfully removed 1.19 [14:36:10] !::1 [14:36:11] localhost [14:36:14] !::1 del [14:36:14] Successfully removed ::1 [14:36:18] !; [14:36:19] when ur downloading stuff from terminal, don't forget to use quotes. Bash doesn't like URL's with special symbols [14:36:22] !; del [14:36:23] Successfully removed ; [14:36:30] okay, i stop flooding now [14:39:05] ! [14:39:05] There is no such key, you probably want to try: !?>, !@, !@param, !^demon, !1.17, !1.17_tracking, !1.17wmf1, !1.18, !1.18wmf1, !2.0, !3x, !42, !a, !abusefilter, !access, !ack, !actionpaths, !actions, !adminsettings, !ajax, !allmessages, !allowoverride, !amend, !ampersand, !analytics, !ancient, !anglebrackets, !announce, !annoyinglittlebug, !anonnotice, !answer, !api, !around, !audit, !auth, !authentication, !autoblock, !autoconfirmed, !autoload, !autothumb, !backup, !bellthecat, !berlin, !bestcase, !blacklist, !blame, !blank, !bleedingedge, !blocking, !boilerplate, !bom, !bot, !botbrain2, !botnoob, !branch, !branchpoints, !browser, !bugsearch, !bugzilla, !bunching, !bureaucracy, !bzapi, !bzsource, !cache, !captcha, !cat, !categories, !centralauth, !chansearch, !chanspam, !charinsert, !cheatsheet, !checkuser, !ci, !cia, !cite, !class, !cloak, !cms, !code, !codereview, !collapse, !colon, !commit, !commons, !compare, !compression, !config, !console, !cookies, !cooluris, !copyhelp, !copying, !copyinghelp, !copytemplates, !core, !corehack, !corn, !createandpromote, !createpage, !credit, !cron, !css, !csshideheader, !csv2wp, !damneditcount, !db, !dbtable, !debbug, !debianubuntu, !debug, !defaultsettings, !defenestrate, !deleteuser, !deploymentqueue, !deployments, !details, !developer, !developers, !devsecurity, !diff3, !dismissablesitenotice, !displaytitle, !documentation, !doesntwork, !domdocument, !downgrade, !download, !doxygen, !dpl, !draft, !dump, !dumphtml, !easy, !echo, !editbuttons, !editing, !edittoken, !email, !emailconfirmed, !enableapi, !enotif, !enwp, !errors, !es, !etherpad, !events, !evilglobals, !exclamation, !export, !exporttemplate, !extensiondistributor, !extensionmatrix, !extensions, !externalauth, !externaleditor, !externalimages, !extranamespace, !f5, !fail, !false, !faq, !feature, !feeds, !file, !filecache, !filepermissions, !filesize, !filetype, !fileurl, !firebug, !flaggedrevs, !flash, !footer, !foreignfilerepos, !fu, !gadget, !gadgets, !gallery, !gerritapi, !gerritdocs, !gerriterror, !gerrithelp, !gerritkeys, !gerritsearch, !gethelp, !gfi, !git, !git-access, !gitaccount, !gitauthor, !gitcheat, !gitemail, !gitrepo, !git-review, !giturl, !gnu, !google, !googleanalytics, !googleit, !grep, !gsocapp, !gsod, !hacker, !HAL, !hdss, !hello, !help, !helpfor, !hetdeploy, !hiddencategories, !hidetabs, !hidetoc, !highlight, !hooks, !hosting, !howto, !html, !html9, !humor, !i18n, !iamabot, !icon, !ie6, !IE9, !iframe, !iis, !iis404, !imabot, !imageauth, !imagelink, !imagemagic, !imagemap, !images, !img_auth, !import, !important, !importhelp, !importimages, !importsources, !info, !infobox, !inputbox, !install, !installlinux, !installmacosx, !instantcommons, !interface, !internalapi, !interwiki, !invalidupload, !is, !itsalwaysbugzilla, !javascript, !jobqueue, !jquery, !labs, !langlinks, !lazy, !ldap, !ldapdebug, !ldapsupport, !line, !links, !list, !listinfo, !lists, !lmgtfy, !localhost, !lockdown, !logo, !lowercase, !lsb, !lst, !lua, !lucene, !magicwords, !mailarchive, !mainpage, !mainpagetitle, !man, !manapi, !maps, !math, !maxlag, !mediawiki, !meetings, !memcached, !memory, !meta, !migrate, !mime, !moderation, !moin, !monday, !moo, !morbo, !move, !mssql, !multiplecommits, !multiplewiki, !mwbot, !mw-config, !mwdumper, !mwhelp, !mysql, !namespaces, !networkauth, !newgit, !newusergroup, !newwindow, !next, !nightly, !nofollow, !noob, !noshell, !nosql, !not, !notes, !nowiki, !nuke, !nullpath, !numberedparams, !ogg, !omgevilbug, !openid, !opw, !oracle, !oversight, !packagefix, !padindex, !pagecss, !paranoid, !parser, !parserconf, !parserfunctions, !passthru, !passwordsql, !patch, !pdhelp, !performance, !petan, !phase3, !php, !php4, !php531, !phpinfo, !phpmyadmin, !phpshell, !phpsucks, !ping, !pipetrick, !plainlinks, !player, !poem, !ppp, !prefsrant, !preventaccess, !profiling, !protect, !protocols, !punehackathon, !purge, !putty, !pywikipediabot, !query, !queue, !rangeblocking, !rawr, !rcbot, !readonly, !rebase, !recaptcha, !redirect, !redlinks, !refresh, !releasedate, !renameuser, !repeat, !reportbug, !reports, !requestcontext, !research, !resetpass, !resetpassword, !resourceloader, !rest, !revdelete, !revert, !reviewqueue, !rfc, !rights, !rightssql, !robots, !rollback, !rt, !rtam, !rtfm, !ruby, !safemode, !sal, !sandbox, !sarcasm, !sb, !scap, !scarytranscluding, !schema, !search, !secrets, !security, !selfmerge, !selinux, !semantichelp, !semanticmediawiki, !sequel, !serversuperglobal, !session, !sharedlogins, !shell, !shellrequest, !shorturls, !sitemap, !sitenotice, !siwae, !sm, !smtp, !smw, !sofixit, !songs, !sop, !sortkey, !sorttable, !sourceforge, !spam, !specialpages, !sqlite, !sqllog, !sqlsearch, !srv193, !start, !stfu, !stringfunctions, !stroopwafels, !subpages, !subst, !suggest, !sul, !sumanah, !support, !supported, !svg, !svnauthor, !svnprops, !svn-rev, !svnsearch, !syntax, !syntaxhighlight, !sysadmin, !t_paamayim_nekudotayim, !tabcompletion, !table, !tables.sql, !tabs, !tag, !tautology, !telepathy, !temp, !templates, !test, !test2, !testing, !tests, !testswarm, !testwiki, !thankyou, !the_mask, !thisbotisnotstupid, !thisisalongkey, !threads, !thumb, !tias, !tidy, !tidyforidiots, !timezone, !toc, !toolserver, !tooquiet, !transclusion, !translatewiki, !true, !ts, !tso, !tss, !tunemysql, !twn, !ui, !umask, !undo, !unicode_test, !uninstall, !unreviewedcode, !update.php, !upgrade, !upload, !uploadbyurl, !useful, !useragent, !userapproval, !variables, !vector, !version, !versionlifecycle, !viewvc, !votes, !vulnerability, !wallofshame, !webinspector, !webrequest, !webscale, !welcome, !wf, !wg, !wiki, !wiki2pdf, !wiki2xml, !wikia, !wikiacode, !wikibugs, !wikidata, !wikieditor, !wikimania, !wikimedia, !wikipmediawiki, !wip, !wmf, !wmf-deployment, !wmo, !worstcase, !wptemplates, !wtf, !wysiwyg, !xampp, !xkcd, !xml, !yoda, !sanitizer, !index.php, !wikipedia, !pastelocalsettings, !flow, !opsschool, !vagrant, !packages, !subversion, !1.21, !1.20, !1.22, !patrol, !wm-bot, !wp, !memoserv, !submodules, !cs, !5.4, !$wgServer, !visualeditor, !master, !unicorn, !rewriteproblems, !parsoid, !pong, !adsense, !github, !sourcecode, !technews, !reboot, !hook, !seen, !tryflow, !sidebar, !htmlform, !bzqs, !roadmap, !apiinternal, !apiclient, !pastebin, !dirtydiffs, !parsoid-web, !parsoid-test, !link, !smart, !autopromote, !vebug, !VE-table, !veq, !maintenance, !bugsong, !ext, !tea, !ask, !jobs, !farm, !newbug, !templateproblems, !mw, !blink, !htmllogs, !logs, !hss, !jsduck, !packagessimple, !skins, !skinning, !job, !codesearch, !ie, !bribe, !gitblit, [14:39:10] 1.17_tracking? wat [14:39:11] !1.17_tracking [14:39:12] The tracking bug for 1.17 tarball is bug 26676 [14:39:17] lol [14:39:21] !1.17_tracking del [14:39:21] Successfully removed 1.17_tracking [14:39:35] !1.18wmf1 del [14:39:35] Successfully removed 1.18wmf1 [14:39:43] !1.17wmf1 [14:39:43] Branched on 3 February 2011 at r81457 from REL1_17 http://www.mediawiki.org/wiki/Special:Code/MediaWiki/81457. See also !1.17 [14:39:56] !1.17wmf1 del [14:39:56] Successfully removed 1.17wmf1 [14:40:01] these don't look useful. [14:40:26] !ajax [14:40:26] http://www.mediawiki.org/wiki/Manual:Ajax [14:41:55] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/82619 (owner: 10Dan-nl) [14:42:49] (03PS25) 10Physikerwelt: Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 [14:43:02] (03CR) 10jenkins-bot: [V: 04-1] Backup [extensions/Math] - 10https://gerrit.wikimedia.org/r/85801 (owner: 10Physikerwelt) [14:43:38] (03PS3) 10Hashar: wrap most functions in a class [tools/release] - 10https://gerrit.wikimedia.org/r/86229 [14:43:40] (03PS1) 10Hashar: extract parse_args out of the class [tools/release] - 10https://gerrit.wikimedia.org/r/86255 [14:43:40] (03PS1) 10Hashar: ask() no more need a skip parameter, use context instead [tools/release] - 10https://gerrit.wikimedia.org/r/86256 [14:55:14] does anyone here know what lua sandbox scribunto uses? [14:56:10] what do you mean "what lua sandbox"? [14:56:47] cscott: [14:56:59] which revision out of the git repo? [14:58:17] (03PS1) 10Hashar: switch methods to be context (options) based [tools/release] - 10https://gerrit.wikimedia.org/r/86259 [15:03:32] (03CR) 10Aude: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson (WMF)) [15:07:03] apergos: IIRC scribunto uses lua instead of v8 because lua was easier to sandbox. but i just did some googling, and it doesn't seem like lua is actually very easy to sandbox at all. so i'm wondering how that works. [15:07:19] apergos: i can probably Read The Source Luke, but i thought i'd venture a question here first [15:09:31] (03CR) 10Rillke: [C: 031] "Tested locally (with only one file, however) and it worked like a charm. Thank you. This will be certainly useful for Wikimedia Commons ad" [core] - 10https://gerrit.wikimedia.org/r/85446 (owner: 10Brian Wolff) [15:11:20] legoktm: https://pagekite.net/ is localtunnel replacement [15:12:48] I thought it was done in-house, maybe that's wrong [15:14:28] cscott: one of the things I remember reading about was the fact that you can control execution times and more importantly memory allocations trivially with Lua - lua supports custom memory allocators, while v8 did not [15:14:31] cscott: http://stackoverflow.com/questions/9671793/limiting-a-lua-scripts-memory-usage [15:14:52] (latest python 3.x just added custom allocators, for example) [15:17:28] as I look at the sandbox code (I have it checked out cause i've had to build it for local installs) it seems like it is indeed homegrown [15:17:35] ie from the git log [15:18:51] YuviPanda, apergos: yeah, ok that seems right. [15:18:56] from what I remember, even homegrowing was not even possible on v8 [15:19:11] homegrown sandboxes are way sketchy, of course. [15:19:25] i think v8 can use custom allocators, but it's not well documented. [15:19:35] plus the v8 team maintains that the only use case for v8 is Chrome/*, and breaks interfaces without too much concern [15:20:00] I got that from skimming nodejs devs complaining about changes in the C++ api, so might not be fully accurate :P [15:20:05] php-spidermonkey might be a better idea [15:20:25] upstream changing is always an issue if you've got an active upstream. no way around that, really. [15:20:39] using an inactive upstream has its own set of issues. ;) [15:21:19] (03CR) 10BryanDavis: [C: 031] "Tests still passing for me on Zend PHP via mediawiki-vagrant (PHP 5.3.10-1ubuntu3.7 with Suhosin-Patch)." [core] - 10https://gerrit.wikimedia.org/r/86151 (owner: 10Chad) [15:21:32] lua is primarily meant to be an embeddable language, so the folks doing it consider 'easy to embed' a feature [15:23:23] YuviPanda: yeah, i keep hearing that. but i think it's totally wrong. [15:23:29] why so, cscott? [15:23:34] lua is 'easy to embed', but v8 is just as easy to embed, as is spidermonkey, etc. [15:23:43] what people usually mean is 'lua is easy to sandbox' ... but it's not, really. [15:23:48] sure, but lua is designed to be embedded, from what I understand [15:23:56] there are lots of unsafe OS calls calls exposed which you have to manually protect [15:24:16] where v8 / spidermonkey (for example) embed by default with *nothing* bound, so in theory they are actually *easier* to sandbox. [15:24:19] this part of my brain that keeps telling me 'dude you are arguing without full knowledge of the facts' is now quite active :D [15:24:46] i don't really want to get into a whole thing [15:25:16] since i have embedded spidermonkey before, hacked a bit on v8, but never tried to embed lua, so i don't have all the facts either [15:25:39] but i was rather surprised to see how little sandboxing info was available in the lua docs. [15:26:00] (and there seems to be a similar upstream issue, in that lua upstream can add new unsafe library stuff to the standard library which will then break your sandbox if you're not careful) [15:26:15] !ajax is https://www.mediawiki.org/wiki/Manual:Ajax [15:26:15] This key already exist - remove it, if you want to change it [15:26:21] so i totally agree that lua is designed to be embedded, but i disagree that this is unusual or actually what is needed. [15:26:25] !ajax del [15:26:25] Successfully removed ajax [15:26:29] !ajax is https://www.mediawiki.org/wiki/Manual:Ajax [15:26:29] Key was added [15:26:31] * cscott stops ranting [15:26:51] cscott: :D [15:27:15] cscott: My extent of knowledge in this case amounts to following the mailing list, so I'm far less qualified [15:27:48] maybe in my Copious Free Time (tm) i'll write php-spidermonkey and bind it to scribunto. [15:28:04] yeah, should be able to fit that into the 3 weeks of spare time you have :P [15:28:43] wait where did you get three weeks of spare time? I'm going there to get some if they have any left [15:29:18] step 1. acquire RSI [15:29:21] there's no step 2 [15:29:28] (03PS1) 10J: Pull upstream fix for File_Ogg [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/86262 [15:30:41] hmm I did that some years back but they didn't give out any [15:30:52] luckily I was able to work around the issue [15:31:01] I've the forebooding feeling that I'm going to get that sometime soon [15:31:17] (did a bunch more lugging of gear and a lot less typing for awhile at the very first signs) [15:31:33] i can't do that :( [15:31:46] my gripping strength is almost non-existant now [15:32:03] uh [15:32:15] you better start doing other stuff *now* [15:32:58] I don't know what else to do that doesn't involve a keyboard :| [15:33:32] draw on a whiteboard, write notes by hand [15:33:36] anything to break up typing [15:40:05] apergos: I'll try [15:44:12] so perhaps i'm just dense ... what does the CAP theorum have to do with edit conflicts? I would like to think i'm familiar with CAP, but i see edit conflicts as being about the machine not being able to guess how to merge two different changes [15:45:03] ebernhardson: I don't know anything about CAP either, but is this from someone named Wikid77? [15:45:09] YuviPanda: guy macon [15:45:12] ah [15:45:18] first time i ignored it, but he keeps bringing it up [15:45:24] so thought i should at least find out what hes talking about :) [15:45:49] heh. Wikid77 is the one who keeps bringing up edit conflicts into every discussion ever, and is usually ignoreable. Not sure about Guy Macon tho [15:46:34] ebernhardson: link? [15:47:16] sec gotta find the right dif [15:47:36] YuviPanda: https://en.wikipedia.org/w/index.php?title=Wikipedia_talk%3AFlow&diff=574703880&oldid=574677627 [15:48:24] and for reference, in regards to no edit conflicts the team was refering to revisions in flow being very small, and the original plan to not let anyone except the creating user edit the post. So edit conflicts would be not necessarily gone forever, but incredibly rare because you would have to conflict with yourself [15:49:33] right. [15:50:41] (03PS1) 10BryanDavis: Make apache::site for main vhost depend on a2enmod. [vagrant] - 10https://gerrit.wikimedia.org/r/86263 [15:50:50] YuviPanda: ^ [15:51:19] ebernhardson: perhaps he considers the client and the server to be parts of a 'distributed' system, and how they can not be in sync at the same time if there exists 3 parties (1 server, 2 clients) with differing info about the same comment? [15:51:21] and ebernhardson i think you're right in that flow was probably saying "edit conflicts are rare" and the commenter was trying to say "no edit conflicts are impossible, by this theorem, thus the flow team must be deliberately lying to us" [15:51:26] a bit of a stretch. [15:52:04] (03PS1) 10Hashar: abstract out version numbers [tools/release] - 10https://gerrit.wikimedia.org/r/86264 [15:52:08] bd808: testing [15:52:16] ebernhardson i think you're right in that flow was probably saying "edit conflicts are rare" and the commenter was trying to say "no edit conflicts are impossible, by this theorem, thus the flow team must be deliberately lying to us" [15:52:31] * cscott apologizes if he's repeating himself, his network connection is flakey [15:52:42] the reference to the CAP theorem is pretty bogus [15:53:03] cscott: thanks for the insight, and yea i repeat myself alot when i'm on mifi it drops all the time [15:53:41] YuviPanda: i suppose it seems possible, but yea its a stretch. [15:53:53] the CAP theorem also doesn't say edit conflicts are impossible [15:54:05] it says, "you can't provide all three of these things..." [15:54:45] so, for example, if you allowed different viewers to perhaps have different views of the order comments were made (violating 'consistency') then you could safely have both of the other two properties. hence, no edit conflicts. [15:54:45] cscott: yea thats my understanding as well, and why i first asked. I can sorta see how he might stretch the definition to fit, although i agree it doesn't [15:54:59] (03CR) 10Yuvipanda: [C: 032] "WFM!" [vagrant] - 10https://gerrit.wikimedia.org/r/86263 (owner: 10BryanDavis) [15:55:01] (03Merged) 10jenkins-bot: Make apache::site for main vhost depend on a2enmod. [vagrant] - 10https://gerrit.wikimedia.org/r/86263 (owner: 10BryanDavis) [15:55:18] bd808: merged [15:55:23] that's not entirely theoretic, that's actually how togetherjs (and most other realtime interaction systems) handle things [15:55:25] bd808: think you can take a stab at wgServer? [15:55:29] YuviPanda: thx [15:55:34] cscott: really? interesting. [15:55:44] YuviPanda: Yeah I'm going to look at that next [15:55:45] you sometimes have to reorder operations, because the server sees A and then B, where you saw B and then A. [15:55:59] YuviPanda: spinning up new instance now... [15:56:02] so you try to structure your operations so that reordering is safe [15:56:16] bd808: sweet. wgServer also affects local things, so you'd want to test that there too [15:56:42] YuviPanda: ok. Will do. [15:56:49] ty bd808 [15:56:58] eg, if both A and B added a new comment *at the exact same time* then somebody (usually the server) has to make an arbitrary decision about who was "first". One or the other of them has to reorder their local history to make that work. [15:57:47] so, sure the CAP theorem is totally correct: both A and B can't *both* be first. but it's also totally irrelevant in practice. [15:57:55] cscott: that makes alot of sense actually [15:58:37] (03PS1) 10Anomie: For caching, avoid passing returnto to Special:CentralAutoLogin [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/86265 [15:58:42] (03PS2) 10Hashar: abstract out version numbers [tools/release] - 10https://gerrit.wikimedia.org/r/86264 [15:58:48] (03PS4) 10Tpt: (bug 45470) Adds support of JSON encoding and decoding [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/85426 [16:00:08] bd808: ori-l also promised to email the lists about a different way for managing extensions. The current one has problems - you can not enable UW and Scribunto roles together anymore, since they both require CodeEditor [16:00:12] puppet will complain and bail out [16:00:39] Yuck. I was worried about that. [16:01:03] he's going to have to make his pretty and simple system much more complex [16:01:06] bd808: there's also the MLEB role patch in gerrit that runs into a similar problem with CLDR [16:01:26] bd808: heh, indeed. we're essentially using puppet for something it wasn't well suited to (perhaps?) - deploying code [16:01:56] It's good at it, but it would really like us to make everything native packages [16:01:56] i dont suppose, anyone use xhprof with mediawiki? I tried to use xdebug with the profiling trigger but kcachegrind mis-interprets the results (e.g. when recursively calling the same function, it gets marked in kcachegrind as 400% of runtime of whatever) [16:02:24] also for some reason kcachegrind is not associating the callers and the calle's, i get two entrys for each function so the graphing all fails :S [16:03:22] (03CR) 10Hashar: "I have extracted some methods as function in following patchsets." [tools/release] - 10https://gerrit.wikimedia.org/r/86229 (owner: 10Hashar) [16:03:53] ebernhardson: I haven't used either xhprof or xdebug to profile MW yet. I've been getting by with StartProfile.php traces so far. Sorry. [16:05:15] bd808: hmm, i suppose i should try that at least, havn't yet :) [16:05:26] but then i have to add profilng calls everywhere :P [16:09:04] bd808: do you plan to look into dataloss when moving at https://bugzilla.wikimedia.org/show_bug.cgi?id=54674 ? [16:09:19] (03PS3) 10Hashar: abstract out version numbers [tools/release] - 10https://gerrit.wikimedia.org/r/86264 [16:09:20] (03PS2) 10Hashar: switch methods to be context (options) based [tools/release] - 10https://gerrit.wikimedia.org/r/86259 [16:10:02] andre__: I could try. Someone else (AaronSchulz?) might be able to figure it out faster. [16:10:14] (03CR) 10Hashar: "Forgot to move git build root out of makeRelease. Refer to it directly in export() and patchExport()" [tools/release] - 10https://gerrit.wikimedia.org/r/86264 (owner: 10Hashar) [16:10:27] andre__: I'm learning a lot but not at full speed yet [16:10:34] yeah, I understand :) [16:11:25] I think the core problem is not that swift failed but that the "transaction" didn't roll back when it did [16:12:25] eek, looks like a pandora box [16:16:06] andre__: FWIW I don't think this is "new". I'm pretty sure that AaronSchulz and robla know other bugs caused by same root. [16:17:21] hmm, I see [16:18:52] AaronSchulz: what do you think about https://bugzilla.wikimedia.org/show_bug.cgi?id=54674 ? Known problem or new? [16:21:13] (03PS1) 10Hashar: let us point extension root to some other place [tools/release] - 10https://gerrit.wikimedia.org/r/86269 [16:30:34] (03PS1) 10Hashar: let us override the tar command [tools/release] - 10https://gerrit.wikimedia.org/r/86270 [16:30:42] (03CR) 10MarkTraceur: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/86172 (owner: 10MarkTraceur) [16:32:02] andre__: I think a LESS component (nested under MediaWiki) in BZ might be warranted [16:32:56] (03CR) 10EBernhardson (WMF): "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85787 (owner: 10EBernhardson (WMF)) [16:35:19] qgil: sorry, I'm someplace the net sucks [16:35:27] qgil: can't repeat the comms handover - sorry [16:36:03] sumanah, of course, I read the log [16:36:50] sumanah, I feel stoopid but well [16:36:53] :) [16:37:53] (03PS1) 10Hashar: skip tarignore if it is not existing [tools/release] - 10https://gerrit.wikimedia.org/r/86271 [16:44:16] (03CR) 10Rillke: [C: 04-1] "(10 comments)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/78341 (owner: 10MarkTraceur) [16:48:06] Thanks so much rillke! [16:50:25] rillke: Does IE8 not support SVG? Bloody hell. [16:50:31] I can probably write a fallback [16:51:27] Krinkle|detached or RoanKattouw_away, do you have advice about supporting IE8 for icons? (I'm using SVGs now but could make them into PNGs and write a fallback) [16:52:16] (03PS4) 10Hashar: abstract out version numbers [tools/release] - 10https://gerrit.wikimedia.org/r/86264 [16:52:17] (03PS2) 10Hashar: let us override the tar command [tools/release] - 10https://gerrit.wikimedia.org/r/86270 [16:52:18] (03PS2) 10Hashar: skip tarignore if it is not existing [tools/release] - 10https://gerrit.wikimedia.org/r/86271 [16:52:19] (03PS2) 10Hashar: let us point extension root to some other place [tools/release] - 10https://gerrit.wikimedia.org/r/86269 [16:52:20] (03PS1) 10Hashar: option to skip gpg signature (--dont-sign) [tools/release] - 10https://gerrit.wikimedia.org/r/86275 [16:52:31] and I call it an end [16:52:43] hexmode: I will review my patches later on [16:52:48] hexmode: feel free to give it a try :-D [16:53:23] hexmode: I am pinging qa-l [16:54:18] (03CR) 10Brion VIBBER: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/78341 (owner: 10MarkTraceur) [16:56:53] and I call it an end ! [16:56:55] *wave* [17:00:59] (03CR) 10MarkTraceur: "(9 comments)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/78341 (owner: 10MarkTraceur) [17:02:40] * brion is excited to see new media viewer happening at last :D [17:03:51] is this part of what mdale was working on all these years? or did his stuff ever get folded in? [17:03:53] I kinda lost track [17:05:56] apergos: this is for images too, I think [17:06:22] apergos: and yes, I think my right hand is near a critical point. Typing one handed now [17:11:36] brion: if you have a sec, could you test https://gerrit.wikimedia.org/r/#/c/86212/ ? [17:12:22] Yuvi, when (today) did you last take a ten minute break? [17:12:27] er YuviPanda [17:12:52] apergos: about 10 mins ago. lay down for about 25mins, had set an alarm for standup [17:12:58] ok [17:12:59] ori-l: ah good lemme test that [17:13:09] i saw it last night but was on my phone, too lazy to come downstairs and test ;) [17:13:19] you're supposed to do this every hour: 10 minutes of not typing [17:14:00] I might already be doing it [17:14:02] * ori-l wisely declines to make an inappropriate joke. [17:14:29] I switched to a newer keyboard that has a very different feel about a week ago. perhaps that [17:14:31] * YuviPanda switches back [17:14:44] * apergos refrains from handing ori-l a pamphlet :-P [17:15:19] maybe the angle or something is wrong [17:19:28] <^d> Is git-review's config really in ~/.config/git-review/git-review.conf seriously? [17:19:34] <^d> That seems...braindead. [17:20:41] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/82040 (owner: 10Krinkle) [17:20:49] ^d: why? [17:21:00] anyone happen to know which class to use or how to convert an mwstore path to a file system path? [17:21:13] <^d> ori-l: Because reading ~/.gitconfig would've made too much sense? [17:21:24] (03PS9) 10Krinkle: Title: Add byte class to unicode class conversion for js [core] - 10https://gerrit.wikimedia.org/r/82040 [17:22:49] (03CR) 10Krinkle: [C: 031] "Status? This is blocking mw.Title and subsequently things in other extensions that are all unable to deal with various titles (could be wo" [core] - 10https://gerrit.wikimedia.org/r/82040 (owner: 10Krinkle) [17:23:16] (03PS12) 10Krinkle: mw.Title: Rewrite from scratch (porting logic from Title.php) [core] - 10https://gerrit.wikimedia.org/r/83047 [17:23:24] YuviPanda: http://www.howtostretch.com/p/wrist-stretches for better hand health [17:23:49] also get a brace to wear when you sleep if your having active pain [17:24:17] dan-nl: I think there's a function for that in FileRepoGroup [17:24:19] Sec [17:24:21] ok vagrant is up, let's try provisioning... [17:24:54] $ cat /etc/virtualbox-version [17:24:55] 4.2.18 [17:24:59] Or I'm drunk and there's no class called FileRepoGroup [17:24:59] ori-l: it woooooorks :D [17:25:03] (03CR) 10Nikerabbit: "(1 comment)" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/86172 (owner: 10MarkTraceur) [17:25:12] <^d> marktraceur: You don't look drunk [17:25:28] ^d: not a bad point :P [17:25:31] RepoGroup [17:25:33] ori-l, nope, didn't work :( [17:25:48] yurik: what didn't? are you testing the same patch brion is? [17:25:55] no - vagrant :) [17:25:57] marktraceur: thanks [17:25:59] (03CR) 10Brion VIBBER: [C: 031 V: 032] "Confirmed this works on my Mac (10.8.5) with VirtualBox 4.2.18" [vagrant] - 10https://gerrit.wikimedia.org/r/86212 (owner: 10Ori.livneh) [17:26:00] * ^d wants his patch merged :) [17:26:10] ori-l, i was testing master [17:26:22] dan-nl: I suspect https://doc.wikimedia.org/mediawiki-core/master/php/html/classRepoGroup.html#a99acff1d5db4d878602c21df2b49d29a *might* do something useful [17:26:31] But at the very least it's the right class to look at [17:26:33] ori-l, should i try the patch? [17:26:53] I guess you could manually run through each repo, I'm pretty *bloody* sure FileRepo objects will have a way to do this [17:26:56] <^d> ori-l: On a totally different note...I got bored last night and started trying to get the dba_* functions implemented for hhvm :) [17:26:59] <^d> https://github.com/demon/hiphop-php/commit/81be449bf8d2978f06cda7d5fab73e50e3337aa7 [17:27:10] * ebernhardson cheers [17:27:29] * brion hmms [17:27:32] yurik: so the log you sent me is up-to-date? [17:27:36] why do we not have a vagrant role for hhvm? [17:27:40] that'd be bitchin' [17:27:41] ori-l, that's what my email was about - vagrant master doesn't work [17:27:46] OK [17:27:49] brion: do i hear a volunteer? :) [17:27:50] Maybe I'm massively wrong [17:27:57] ori-l, it doesn't start apach2 service [17:28:01] Where the hell did I see the mwstore thing [17:28:05] ebernhardson: ain't nobody got time for that :D [17:28:05] brion: I keep poking ^d about hhvm for vagrant [17:28:23] <^d> Hey, and I said it was a good idea! [17:28:28] :) [17:28:32] brion: there's a bug for it. there's a bash script for fetching and compiling the dependencies for precise and i think the ability to use the script from upstream verbatim trumps puppetization [17:28:33] who would've thought ;) [17:28:35] marktraceur: ja, i've been trying to get the fs path … been able to use the key returned from the hash to get the mwstore path, but not the fs path … will look into the class you mentioned [17:28:39] <^d> Then I got sucked down a rabbit hole of trying to get vagrant + parallels working :) [17:28:40] i'll throw something in BZ so ^d can get occasional annoying pings about it [17:28:47] oh it's already there yay :D [17:28:48] brion: but still, the script could be put in place by puppet [17:28:56] <^d> I've also got to get one setup for elastic. [17:29:04] ^d: YES PLZ [17:29:09] brion: as an addition, if you add a vagrant role now, it automagically can be used in labs too! [17:29:34] ^d: I sent manybubbles a few notes on that but he's been busy squishing bugs [17:29:49] <^d> Yeah he's busy. Vagrant is low priority for us atm. [17:29:57] <^d> But maybe I can do this today for hhvm + es. [17:29:58] ori-l: i don't see the bug in mediawiki-vagrant component ... [17:30:08] marktraceur: https://commons.wikimedia.org/wiki/MediaWiki:BindPolyfill.js [17:30:09] that'd be cool [17:30:10] ori-l: they have it in ops-puppet would love to reuse [17:30:12] brion: hrm, maybe i lied [17:30:16] ori-l: btw, we can no longer enable both UW and scribunto component, because of conflicting wikieditor [17:30:16] heh [17:30:38] i really need help maintaining mw-v [17:30:47] rillke: The answer is just to use an anonymous function, really [17:30:52] as in, maintaining it, not adding new features [17:30:53] Easy enough [17:30:56] hah :P [17:30:57] YuviPanda: extract wikieditor to own role and include from both? [17:31:24] bd808: there's an open patchset where mleb did that for CLDR, and Ori mentioned a 'better' way [17:31:30] <^d> Ok so I installed virtualbox and vagrant and cloned mw-v. [17:31:53] "better" [17:31:53] marktraceur: there is another issue. should I comment there or here? [17:32:08] rillke: There is probably better 'cause I won't forget [17:32:17] Or I could put it up on the roadmap if it's a big issue [17:32:25] (03CR) 10Ori.livneh: [C: 032] "Should work with 1.1+. Thanks for testing." [vagrant] - 10https://gerrit.wikimedia.org/r/86212 (owner: 10Ori.livneh) [17:32:27] (03Merged) 10jenkins-bot: Use interval Vagrant hook to determine VirtualBox version [vagrant] - 10https://gerrit.wikimedia.org/r/86212 (owner: 10Ori.livneh) [17:32:32] rillke: Either way works though :( [17:32:34] :)* [17:32:39] Frickin' keyboard [17:32:48] ori-l: ok i filed https://bugzilla.wikimedia.org/show_bug.cgi?id=54715 [17:32:58] i saw, thank you [17:32:59] ! [17:33:00] There is no such key, you probably want to try: !?>, !@, !@param, !^demon, !1.17, !1.18, !2.0, !3x, !42, !a, !abusefilter, !access, !ack, !actionpaths, !actions, !adminsettings, !allmessages, !allowoverride, !amend, !ampersand, !analytics, !ancient, !anglebrackets, !announce, !annoyinglittlebug, !anonnotice, !answer, !api, !around, !audit, !auth, !authentication, !autoblock, !autoconfirmed, !autoload, !autothumb, !backup, !bellthecat, !berlin, !bestcase, !blacklist, !blame, !blank, !bleedingedge, !blocking, !boilerplate, !bom, !bot, !botbrain2, !botnoob, !branch, !branchpoints, !browser, !bugsearch, !bugzilla, !bunching, !bureaucracy, !bzapi, !bzsource, !cache, !captcha, !cat, !categories, !centralauth, !chansearch, !chanspam, !charinsert, !cheatsheet, !checkuser, !ci, !cia, !cite, !class, !cloak, !cms, !code, !codereview, !collapse, !colon, !commit, !commons, !compare, !compression, !config, !console, !cookies, !cooluris, !copyhelp, !copying, !copyinghelp, !copytemplates, !core, !corehack, !corn, !createandpromote, !createpage, !credit, !cron, !css, !csshideheader, !csv2wp, !damneditcount, !db, !dbtable, !debbug, !debianubuntu, !debug, !defaultsettings, !defenestrate, !deleteuser, !deploymentqueue, !deployments, !details, !developer, !developers, !devsecurity, !diff3, !dismissablesitenotice, !displaytitle, !documentation, !doesntwork, !domdocument, !downgrade, !download, !doxygen, !dpl, !draft, !dump, !dumphtml, !easy, !echo, !editbuttons, !editing, !edittoken, !email, !emailconfirmed, !enableapi, !enotif, !enwp, !errors, !es, !etherpad, !events, !evilglobals, !exclamation, !export, !exporttemplate, !extensiondistributor, !extensionmatrix, !extensions, !externalauth, !externaleditor, !externalimages, !extranamespace, !f5, !fail, !false, !faq, !feature, !feeds, !file, !filecache, !filepermissions, !filesize, !filetype, !fileurl, !firebug, !flaggedrevs, !flash, !footer, !foreignfilerepos, !fu, !gadget, !gadgets, !gallery, !gerritapi, !gerritdocs, !gerriterror, !gerrithelp, !gerritkeys, !gerritsearch, !gethelp, !gfi, !git, !git-access, !gitaccount, !gitauthor, !gitcheat, !gitemail, !gitrepo, !git-review, !giturl, !gnu, !google, !googleanalytics, !googleit, !grep, !gsocapp, !gsod, !hacker, !HAL, !hdss, !hello, !help, !helpfor, !hetdeploy, !hiddencategories, !hidetabs, !hidetoc, !highlight, !hooks, !hosting, !howto, !html, !html9, !humor, !i18n, !iamabot, !icon, !ie6, !IE9, !iframe, !iis, !iis404, !imabot, !imageauth, !imagelink, !imagemagic, !imagemap, !images, !img_auth, !import, !important, !importhelp, !importimages, !importsources, !info, !infobox, !inputbox, !install, !installlinux, !installmacosx, !instantcommons, !interface, !internalapi, !interwiki, !invalidupload, !is, !itsalwaysbugzilla, !javascript, !jobqueue, !jquery, !labs, !langlinks, !lazy, !ldap, !ldapdebug, !ldapsupport, !line, !links, !list, !listinfo, !lists, !lmgtfy, !localhost, !lockdown, !logo, !lowercase, !lsb, !lst, !lua, !lucene, !magicwords, !mailarchive, !mainpage, !mainpagetitle, !man, !manapi, !maps, !math, !maxlag, !mediawiki, !meetings, !memcached, !memory, !meta, !migrate, !mime, !moderation, !moin, !monday, !moo, !morbo, !move, !mssql, !multiplecommits, !multiplewiki, !mwbot, !mw-config, !mwdumper, !mwhelp, !mysql, !namespaces, !networkauth, !newgit, !newusergroup, !newwindow, !next, !nightly, !nofollow, !noob, !noshell, !nosql, !not, !notes, !nowiki, !nuke, !nullpath, !numberedparams, !ogg, !omgevilbug, !openid, !opw, !oracle, !oversight, !packagefix, !padindex, !pagecss, !paranoid, !parser, !parserconf, !parserfunctions, !passthru, !passwordsql, !patch, !pdhelp, !performance, !petan, !phase3, !php, !php4, !php531, !phpinfo, !phpmyadmin, !phpshell, !phpsucks, !ping, !pipetrick, !plainlinks, !player, !poem, !ppp, !prefsrant, !preventaccess, !profiling, !protect, !protocols, !punehackathon, !purge, !putty, !pywikipediabot, !query, !queue, !rangeblocking, !rawr, !rcbot, !readonly, !rebase, !recaptcha, !redirect, !redlinks, !refresh, !releasedate, !renameuser, !repeat, !reportbug, !reports, !requestcontext, !research, !resetpass, !resetpassword, !resourceloader, !rest, !revdelete, !revert, !reviewqueue, !rfc, !rights, !rightssql, !robots, !rollback, !rt, !rtam, !rtfm, !ruby, !safemode, !sal, !sandbox, !sarcasm, !sb, !scap, !scarytranscluding, !schema, !search, !secrets, !security, !selfmerge, !selinux, !semantichelp, !semanticmediawiki, !sequel, !serversuperglobal, !session, !sharedlogins, !shell, !shellrequest, !shorturls, !sitemap, !sitenotice, !siwae, !sm, !smtp, !smw, !sofixit, !songs, !sop, !sortkey, !sorttable, !sourceforge, !spam, !specialpages, !sqlite, !sqllog, !sqlsearch, !srv193, !start, !stfu, !stringfunctions, !stroopwafels, !subpages, !subst, !suggest, !sul, !sumanah, !support, !supported, !svg, !svnauthor, !svnprops, !svn-rev, !svnsearch, !syntax, !syntaxhighlight, !sysadmin, !t_paamayim_nekudotayim, !tabcompletion, !table, !tables.sql, !tabs, !tag, !tautology, !telepathy, !temp, !templates, !test, !test2, !testing, !tests, !testswarm, !testwiki, !thankyou, !the_mask, !thisbotisnotstupid, !thisisalongkey, !threads, !thumb, !tias, !tidy, !tidyforidiots, !timezone, !toc, !toolserver, !tooquiet, !transclusion, !translatewiki, !true, !ts, !tso, !tss, !tunemysql, !twn, !ui, !umask, !undo, !unicode_test, !uninstall, !unreviewedcode, !update.php, !upgrade, !upload, !uploadbyurl, !useful, !useragent, !userapproval, !variables, !vector, !version, !versionlifecycle, !viewvc, !votes, !vulnerability, !wallofshame, !webinspector, !webrequest, !webscale, !welcome, !wf, !wg, !wiki, !wiki2pdf, !wiki2xml, !wikia, !wikiacode, !wikibugs, !wikidata, !wikieditor, !wikimania, !wikimedia, !wikipmediawiki, !wip, !wmf, !wmf-deployment, !wmo, !worstcase, !wptemplates, !wtf, !wysiwyg, !xampp, !xkcd, !xml, !yoda, !sanitizer, !index.php, !wikipedia, !pastelocalsettings, !flow, !opsschool, !vagrant, !packages, !subversion, !1.21, !1.20, !1.22, !patrol, !wm-bot, !wp, !memoserv, !submodules, !cs, !5.4, !$wgServer, !visualeditor, !master, !unicorn, !rewriteproblems, !parsoid, !pong, !adsense, !github, !sourcecode, !technews, !reboot, !hook, !seen, !tryflow, !sidebar, !htmlform, !bzqs, !roadmap, !apiinternal, !apiclient, !pastebin, !dirtydiffs, !parsoid-web, !parsoid-test, !link, !smart, !autopromote, !vebug, !VE-table, !veq, !maintenance, !bugsong, !ext, !tea, !ask, !jobs, !farm, !newbug, !templateproblems, !mw, !blink, !htmllogs, !logs, !hss, !jsduck, !packagessimple, !skins, !skinning, !job, !codesearch, !ie, !bribe, !gitblit, !ajax, [17:33:13] <^d> The heck? [17:33:14] !berlin [17:33:15] Registration is now open for the Berlin hackathon, 1-3 June 2012. See https://www.mediawiki.org/wiki/Berlin_Hackathon_2012 [17:33:15] I'm slowly killing many of these [17:33:17] rillke: do you have a few moments to discuss Module:Languages? [17:33:17] !! [17:33:17] There is no such key, you probably want to try: !?>, !anonnotice, !bom, !cms, !console, !cooluris, !credit, !css, !exclamation, !export, !fail, !firebug, !footer, !fu, !hello, !help, !hidetabs, !iamabot, !iframe, !images, !import, !important, !install, !lockdown, !lst, !lucene, !mainpagetitle, !maps, !math, !maxlag, !moin, !mwdumper, !namespaces, !next, !nosql, !omgevilbug, !oracle, !parserfunctions, !pipetrick, !ppp, !queue, !readonly, !releasedate, !resetpass, !secrets, !shorturls, !sitenotice, !sm, !tabcompletion, !templates, !tooquiet, !ts, !tss, !vector, !welcome, !wikibugs, !wikimedia, !pastelocalsettings, !unicorn, !rewriteproblems, !parsoid, !jobs, !hss, !skins, !skinning, !bribe, [17:33:19] !berlin del [17:33:19] Successfully removed berlin [17:33:23] <^d> !^demon [17:33:23] !nosql [17:33:23] Demon is a champ, everybody knows that. [17:33:24] we should migrate to nosql then we wouldn't have all these tables/fields problems and could have more agile development! and everything would just work [17:33:27] !! is an exclamation point [17:33:27] Key was added [17:33:34] ! [17:33:34] There is no such key, you probably want to try: !?>, !@, !@param, !^demon, !1.17, !1.18, !2.0, !3x, !42, !a, !abusefilter, !access, !ack, !actionpaths, !actions, !adminsettings, !allmessages, !allowoverride, !amend, !ampersand, !analytics, !ancient, !anglebrackets, !announce, !annoyinglittlebug, !anonnotice, !answer, !api, !around, !audit, !auth, !authentication, !autoblock, !autoconfirmed, !autoload, !autothumb, !backup, !bellthecat, !bestcase, !blacklist, !blame, !blank, !bleedingedge, !blocking, !boilerplate, !bom, !bot, !botbrain2, !botnoob, !branch, !branchpoints, !browser, !bugsearch, !bugzilla, !bunching, !bureaucracy, !bzapi, !bzsource, !cache, !captcha, !cat, !categories, !centralauth, !chansearch, !chanspam, !charinsert, !cheatsheet, !checkuser, !ci, !cia, !cite, !class, !cloak, !cms, !code, !codereview, !collapse, !colon, !commit, !commons, !compare, !compression, !config, !console, !cookies, !cooluris, !copyhelp, !copying, !copyinghelp, !copytemplates, !core, !corehack, !corn, !createandpromote, !createpage, !credit, !cron, !css, !csshideheader, !csv2wp, !damneditcount, !db, !dbtable, !debbug, !debianubuntu, !debug, !defaultsettings, !defenestrate, !deleteuser, !deploymentqueue, !deployments, !details, !developer, !developers, !devsecurity, !diff3, !dismissablesitenotice, !displaytitle, !documentation, !doesntwork, !domdocument, !downgrade, !download, !doxygen, !dpl, !draft, !dump, !dumphtml, !easy, !echo, !editbuttons, !editing, !edittoken, !email, !emailconfirmed, !enableapi, !enotif, !enwp, !errors, !es, !etherpad, !events, !evilglobals, !exclamation, !export, !exporttemplate, !extensiondistributor, !extensionmatrix, !extensions, !externalauth, !externaleditor, !externalimages, !extranamespace, !f5, !fail, !false, !faq, !feature, !feeds, !file, !filecache, !filepermissions, !filesize, !filetype, !fileurl, !firebug, !flaggedrevs, !flash, !footer, !foreignfilerepos, !fu, !gadget, !gadgets, !gallery, !gerritapi, !gerritdocs, !gerriterror, !gerrithelp, !gerritkeys, !gerritsearch, !gethelp, !gfi, !git, !git-access, !gitaccount, !gitauthor, !gitcheat, !gitemail, !gitrepo, !git-review, !giturl, !gnu, !google, !googleanalytics, !googleit, !grep, !gsocapp, !gsod, !hacker, !HAL, !hdss, !hello, !help, !helpfor, !hetdeploy, !hiddencategories, !hidetabs, !hidetoc, !highlight, !hooks, !hosting, !howto, !html, !html9, !humor, !i18n, !iamabot, !icon, !ie6, !IE9, !iframe, !iis, !iis404, !imabot, !imageauth, !imagelink, !imagemagic, !imagemap, !images, !img_auth, !import, !important, !importhelp, !importimages, !importsources, !info, !infobox, !inputbox, !install, !installlinux, !installmacosx, !instantcommons, !interface, !internalapi, !interwiki, !invalidupload, !is, !itsalwaysbugzilla, !javascript, !jobqueue, !jquery, !labs, !langlinks, !lazy, !ldap, !ldapdebug, !ldapsupport, !line, !links, !list, !listinfo, !lists, !lmgtfy, !localhost, !lockdown, !logo, !lowercase, !lsb, !lst, !lua, !lucene, !magicwords, !mailarchive, !mainpage, !mainpagetitle, !man, !manapi, !maps, !math, !maxlag, !mediawiki, !meetings, !memcached, !memory, !meta, !migrate, !mime, !moderation, !moin, !monday, !moo, !morbo, !move, !mssql, !multiplecommits, !multiplewiki, !mwbot, !mw-config, !mwdumper, !mwhelp, !mysql, !namespaces, !networkauth, !newgit, !newusergroup, !newwindow, !next, !nightly, !nofollow, !noob, !noshell, !nosql, !not, !notes, !nowiki, !nuke, !nullpath, !numberedparams, !ogg, !omgevilbug, !openid, !opw, !oracle, !oversight, !packagefix, !padindex, !pagecss, !paranoid, !parser, !parserconf, !parserfunctions, !passthru, !passwordsql, !patch, !pdhelp, !performance, !petan, !phase3, !php, !php4, !php531, !phpinfo, !phpmyadmin, !phpshell, !phpsucks, !ping, !pipetrick, !plainlinks, !player, !poem, !ppp, !prefsrant, !preventaccess, !profiling, !protect, !protocols, !punehackathon, !purge, !putty, !pywikipediabot, !query, !queue, !rangeblocking, !rawr, !rcbot, !readonly, !rebase, !recaptcha, !redirect, !redlinks, !refresh, !releasedate, !renameuser, !repeat, !reportbug, !reports, !requestcontext, !research, !resetpass, !resetpassword, !resourceloader, !rest, !revdelete, !revert, !reviewqueue, !rfc, !rights, !rightssql, !robots, !rollback, !rt, !rtam, !rtfm, !ruby, !safemode, !sal, !sandbox, !sarcasm, !sb, !scap, !scarytranscluding, !schema, !search, !secrets, !security, !selfmerge, !selinux, !semantichelp, !semanticmediawiki, !sequel, !serversuperglobal, !session, !sharedlogins, !shell, !shellrequest, !shorturls, !sitemap, !sitenotice, !siwae, !sm, !smtp, !smw, !sofixit, !songs, !sop, !sortkey, !sorttable, !sourceforge, !spam, !specialpages, !sqlite, !sqllog, !sqlsearch, !srv193, !start, !stfu, !stringfunctions, !stroopwafels, !subpages, !subst, !suggest, !sul, !sumanah, !support, !supported, !svg, !svnauthor, !svnprops, !svn-rev, !svnsearch, !syntax, !syntaxhighlight, !sysadmin, !t_paamayim_nekudotayim, !tabcompletion, !table, !tables.sql, !tabs, !tag, !tautology, !telepathy, !temp, !templates, !test, !test2, !testing, !tests, !testswarm, !testwiki, !thankyou, !the_mask, !thisbotisnotstupid, !thisisalongkey, !threads, !thumb, !tias, !tidy, !tidyforidiots, !timezone, !toc, !toolserver, !tooquiet, !transclusion, !translatewiki, !true, !ts, !tso, !tss, !tunemysql, !twn, !ui, !umask, !undo, !unicode_test, !uninstall, !unreviewedcode, !update.php, !upgrade, !upload, !uploadbyurl, !useful, !useragent, !userapproval, !variables, !vector, !version, !versionlifecycle, !viewvc, !votes, !vulnerability, !wallofshame, !webinspector, !webrequest, !webscale, !welcome, !wf, !wg, !wiki, !wiki2pdf, !wiki2xml, !wikia, !wikiacode, !wikibugs, !wikidata, !wikieditor, !wikimania, !wikimedia, !wikipmediawiki, !wip, !wmf, !wmf-deployment, !wmo, !worstcase, !wptemplates, !wtf, !wysiwyg, !xampp, !xkcd, !xml, !yoda, !sanitizer, !index.php, !wikipedia, !pastelocalsettings, !flow, !opsschool, !vagrant, !packages, !subversion, !1.21, !1.20, !1.22, !patrol, !wm-bot, !wp, !memoserv, !submodules, !cs, !5.4, !$wgServer, !visualeditor, !master, !unicorn, !rewriteproblems, !parsoid, !pong, !adsense, !github, !sourcecode, !technews, !reboot, !hook, !seen, !tryflow, !sidebar, !htmlform, !bzqs, !roadmap, !apiinternal, !apiclient, !pastebin, !dirtydiffs, !parsoid-web, !parsoid-test, !link, !smart, !autopromote, !vebug, !VE-table, !veq, !maintenance, !bugsong, !ext, !tea, !ask, !jobs, !farm, !newbug, !templateproblems, !mw, !blink, !htmllogs, !logs, !hss, !jsduck, !packagessimple, !skins, !skinning, !job, !codesearch, !ie, !bribe, !gitblit, !ajax, !!, [17:33:42] ! is an exclamation point [17:33:42] Key was added [17:33:43] marktraceur: Using a bind polyfill can save you a lot of work. [17:33:43] wheeee [17:33:48] maybe you have to say !! [17:33:48] <^d> !^demon del [17:33:49] Successfully removed ^demon [17:33:57] * brion hides from the bot [17:34:01] rillke: Not really, it's two lines [17:34:02] !lst [17:34:02] `!e Labeled_Section_Transclusion`` [17:34:09] !lst [17:34:09] `!e Labeled_Section_Transclusion`` [17:34:11] oh dear [17:34:14] !lst del [17:34:14] Successfully removed lst [17:34:35] marktraceur: You don't intend updating this library at some time? [17:34:55] Yeah, I'm going to [17:35:00] <^d> brion: speaking of hhvm, I wrote a patch to fix ~400 phpunit tests. bd808 +1'd already. [17:35:02] <^d> https://gerrit.wikimedia.org/r/#/c/86151/ [17:35:13] whee [17:35:14] * rillke meant updating from source [17:35:22] "I have no clue how this ever actually worked under Zend" hah [17:35:51] ori-l: could you file a ticket to request a LESS component in Bugzilla? See https://www.mediawiki.org/wiki/Bug_management/Project_Maintainers#To_add_a_project_or_component - TIA. [17:36:15] andre__: will do [17:36:16] * rillke sees that Mark is also the author there [17:36:50] rillke: Yeah [17:36:56] Hence, not a lot of problem. :) [17:36:59] (03CR) 10Anomie: "I'm still not a fan of the implementation of friprop. Now you explicitly ask for none and you get all of them, including ones like "local"" [core] - 10https://gerrit.wikimedia.org/r/85344 (owner: 10MarkTraceur) [17:37:09] :) [17:37:20] anomie: 'local' isn't included anymore, I think [17:37:36] And why would you ask for none? That makes no sense. [17:37:47] Checking for existence? [17:38:19] marktraceur: When I just tried it on my local test wiki, it gave me 'local' in the results. As for asking for none, sure it makes no sense, but returning everything doesn't make sense either. [17:38:31] Urgh. [17:38:31] (03CR) 10Rillke: "There is an issue with the alignment of the lightbox: It is always aligned to the top of the document. When having scrolled-down and click" [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/78341 (owner: 10MarkTraceur) [17:38:32] 'kay [17:38:46] i'm sure theres a function but i'm not seeing it, what can i pass a string plus a length to and it returns the string, possibly shortened with a "..." at the end? [17:38:49] And I would be OK with returning the local repo, but it seems questionably useful [17:39:39] True, but on the other hand I can't come up with a reason not to include it. [17:40:01] Just In Case someone has a questionable use for it ;) [17:41:46] (03CR) 10Bartosz Dziewoński: [C: 032] "Okay, let's do this. The code is straightforward, tests apparently pass, and I see no reason why this function would be unwanted." [core] - 10https://gerrit.wikimedia.org/r/45677 (owner: 10Parent5446) [17:42:08] (03PS13) 10Catrope: mw.Title: Rewrite from scratch (porting logic from Title.php) [core] - 10https://gerrit.wikimedia.org/r/83047 (owner: 10Krinkle) [17:42:26] (03PS14) 10Krinkle: mw.Title: Rewrite from scratch (porting logic from Title.php) [core] - 10https://gerrit.wikimedia.org/r/83047 [17:42:42] marktraceur: à propos Function.prototype.bind, could you review https://gerrit.wikimedia.org/r/#/c/85324/ ? [17:43:30] rillke: I'm not sure I can today, but maybe this weekend sometime [17:43:37] If I can bring myself to [17:43:43] (03PS2) 10Nikerabbit: Tests for project card actions [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86194 [17:44:13] (03PS15) 10Krinkle: mw.Title: Rewrite from scratch (porting logic from Title.php) [core] - 10https://gerrit.wikimedia.org/r/83047 [17:45:13] (03CR) 10Krinkle: "* Clean up some doc comments." [core] - 10https://gerrit.wikimedia.org/r/83047 (owner: 10Krinkle) [17:45:28] (03CR) 10Krinkle: "* Re-applied Roan's commit-message change." [core] - 10https://gerrit.wikimedia.org/r/83047 (owner: 10Krinkle) [17:45:29] (03Merged) 10jenkins-bot: Added new MWTimestamp::getRelativeTimestamp for pure relative. [core] - 10https://gerrit.wikimedia.org/r/45677 (owner: 10Parent5446) [17:45:58] ori-l, , should i test your patch? [17:46:13] (03CR) 10Krinkle: [C: 031] "Changes by others look good." [core] - 10https://gerrit.wikimedia.org/r/83047 (owner: 10Krinkle) [17:46:17] yurik: unrelated to your problem; i'm lookg at your logs tho, just a sec [17:46:33] thx :)( [17:47:36] (03PS4) 10Krinkle: When a module is blocking (position=top), never defer its execution [core] - 10https://gerrit.wikimedia.org/r/75270 (owner: 10Catrope) [17:48:56] (03CR) 10Anomie: [C: 032] "Other code paths look good too." [core] - 10https://gerrit.wikimedia.org/r/86021 (owner: 10Brion VIBBER) [17:49:22] (03PS3) 10Krinkle: bug 54378: remove duplicate file extensions from output messages [core] - 10https://gerrit.wikimedia.org/r/86021 (owner: 10Brion VIBBER) [17:49:33] whee [17:49:44] (03CR) 10Krinkle: [C: 032] "Oh, clash. re-+2" [core] - 10https://gerrit.wikimedia.org/r/86021 (owner: 10Brion VIBBER) [17:50:07] yurik: what happens if you just run 'vagrant provision'? [17:54:19] (03CR) 10Catrope: [C: 032] Title: Add byte class to unicode class conversion for js [core] - 10https://gerrit.wikimedia.org/r/82040 (owner: 10Krinkle) [17:55:46] (03PS1) 10Ori.livneh: Apache service: hasstatus => true [vagrant] - 10https://gerrit.wikimedia.org/r/86278 [17:56:03] yurik: ping [17:56:14] Ffs [17:56:26] (03Merged) 10jenkins-bot: Remove duplicate file extensions from output messages [core] - 10https://gerrit.wikimedia.org/r/86021 (owner: 10Brion VIBBER) [17:56:34] (03PS5) 10Krinkle: resourceloader: Never defer execution for top loading modules [core] - 10https://gerrit.wikimedia.org/r/75270 (owner: 10Catrope) [17:56:54] RoanKattouw: Isn't there a bug report about ^ ? [17:56:59] (03CR) 10Krinkle: [C: 032] resourceloader: Never defer execution for top loading modules [core] - 10https://gerrit.wikimedia.org/r/75270 (owner: 10Catrope) [17:57:06] Not that I'm aware of [17:57:13] ori-l, now - nothing because i recreated the machine with debug enabled [17:57:16] I noticed it because it broke VE's plugin infrastructure [17:57:25] ori-l, i could re-run it of course [17:57:30] and most top-loading modules don't have CSS [17:57:43] yurik: rerun what? [17:58:03] ori-l, but the point is that when i start it apache doesnt start (except for the first time). Rerun - provision [17:58:24] yeah, please do rerun it [17:58:46] (03Merged) 10jenkins-bot: Title: Add byte class to unicode class conversion for js [core] - 10https://gerrit.wikimedia.org/r/82040 (owner: 10Krinkle) [17:59:02] Anyone know if we've got a nice way to just get the simplified backtrace (like we print to fatal.log etc) on demand? debug_print_backtrace() etc are far too verbose [17:59:56] Reedy: wfGetAllCallers() or whatever it's called [18:00:02] Reedy: is the issue in https://bugzilla.wikimedia.org/54178 still occurring on wmf18? [18:00:10] Aha [18:01:16] brion: meeting on hangout! [18:01:25] Why doesn't my irc client think aude is in here... [18:01:32] heh [18:01:47] Now it does [18:01:58] * aude always here [18:02:30] It would look like the logs we have for "today" (ie all after moving to wmf18) are empty of those errors.. [18:02:37] ok, good [18:03:02] the problem was with deleted items but appears to me it is handled correctly in wmf18 [18:03:16] (03PS3) 10Amire80: Tests for project card actions [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86194 (owner: 10Nikerabbit) [18:03:50] (03Merged) 10jenkins-bot: resourceloader: Never defer execution for top loading modules [core] - 10https://gerrit.wikimedia.org/r/75270 (owner: 10Catrope) [18:04:16] (03CR) 10Amire80: [C: 032] Tests for project card actions [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86194 (owner: 10Nikerabbit) [18:04:34] 2013-09-27 18:04:05 mw1110 mediawikiwiki: require/MediaWiki::run/MediaWiki::main/MediaWiki::performRequest/MediaWiki::performAction/EditAction::show/EditPage::edit/EditPage::getEditPermissionErrors/wfArrayDiff2/array_udiff/wfArrayDiff2_cmp [18:04:44] ori-l, sent [18:04:46] RoanKattouw: Not brilliant but certainly more readable than 1000s of lines :) [18:04:53] (03Merged) 10jenkins-bot: Tests for project card actions [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86194 (owner: 10Nikerabbit) [18:05:56] ori-l, were you able to debug with vagrant's debug role? [18:06:00] and phpstorm? [18:07:45] I wish more people would hit rebase before merging stuff [18:07:58] https://git.wikimedia.org/summary/mediawiki%2Fcore.git [18:08:04] Half of the short log are are just merges [18:11:12] Perhaps gerrit should do that automatically (or gerrit-bot should as part of the post +2 steps) [18:11:26] Seems like an automated sort of task that should be handled by a machine [18:13:10] yurik: no, I don't have phpstorm [18:13:14] I asked you to test it, as I recall [18:13:46] ori-l, i did - result - it doesn't work :) [18:13:56] sorry couldn't get to it earlier :( [18:14:15] can you report it? [18:14:21] are you using any other debugger? [18:14:27] I could try to figure out the patch [18:14:35] because it doesn't seem to be right [18:18:12] did you open the port in Vagrantfile? [18:20:17] ori-l, yes, thats' why it didn't work - from what i understand, xdebug makes an outbound connection to the debugger [18:20:27] and the debugger listens on 9000 [18:20:34] which means noone else should grab that port [18:20:40] so fix it :P [18:21:01] ori-l, i will , was just curious where you got those settings from [18:21:08] thought they were working for you [18:22:00] (03PS5) 10Mwalker: Allow MW and extensions to declare included components [core] - 10https://gerrit.wikimedia.org/r/66054 [18:23:31] (03CR) 10Mwalker: [C: 04-1] "I need to address Aude's comments from the last patch" [core] - 10https://gerrit.wikimedia.org/r/66054 (owner: 10Mwalker) [18:25:10] bawolff: gerrit can do that [18:25:20] but for some reason it's not turned on for core [18:25:26] it is enabled for some operations repos or something [18:25:30] ^d would know [18:25:37] Reedy: ^ [18:25:51] Reedy: git log --no-merges [18:27:58] MatmaRex: https://git.wikimedia.org/summary/mediawiki%2Fcore.git [18:28:04] ori-l, the point is that "vagrant up" still doesn't start apache service for some reason, even though vagrant provision did start it [18:28:26] (03PS4) 10Amire80: Test for project selector [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86199 (owner: 10Nikerabbit) [18:29:14] (03CR) 10Amire80: [C: 032] Test for project selector [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86199 (owner: 10Nikerabbit) [18:29:19] (03Merged) 10jenkins-bot: Test for project selector [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86199 (owner: 10Nikerabbit) [18:29:34] Reedy: last time i checked gitblit was open-source. :P [18:29:35] (03PS3) 10Amire80: Implement test for search [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86203 (owner: 10Nikerabbit) [18:33:15] yurik: that's vagrant's design. the provisioner (i.e., puppet) runs in two cases: [18:33:25] 1) when explicitly invoked via 'vagrant provision' [18:33:38] (03CR) 10Amire80: [C: 032] Implement test for search [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86203 (owner: 10Nikerabbit) [18:33:44] (03Merged) 10jenkins-bot: Implement test for search [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86203 (owner: 10Nikerabbit) [18:33:56] 2) on 'vagrant up', when the machine does not exist or is halted (i.e., fully shut down) [18:34:06] it does not run if the machine is merely suspended. [18:34:16] i think the behavior is confusing, but it's upstream's fault [18:35:05] (03PS1) 10Anomie: Adjust styling on jQuery authorize dialog [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/86285 [18:35:54] (03CR) 10CSteipp: [C: 032] For caching, avoid passing returnto to Special:CentralAutoLogin [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/86265 (owner: 10Anomie) [18:36:00] ori-l, the email i sent you today was from a cold start - the machine was halted and rebooted [18:36:00] (03Merged) 10jenkins-bot: For caching, avoid passing returnto to Special:CentralAutoLogin [extensions/CentralAuth] - 10https://gerrit.wikimedia.org/r/86265 (owner: 10Anomie) [18:36:26] there was no suspension [18:36:52] (03CR) 10Anomie: "(1 comment)" [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/86285 (owner: 10Anomie) [18:37:00] yurik: OK, i'll investigate it [18:37:17] can you try vagrant halt / vagrant up again? [18:37:27] i know it works now, just wondering if it still requires explicit provisioning [18:38:15] hey, that's cool, you can delete patchsets on draft changes in gerrit [18:38:20] ori-l, i tried it before - yes, it does. I didn't before. Will do it again though [18:38:24] (03CR) 10Ori.livneh: [C: 032] Apache service: hasstatus => true [vagrant] - 10https://gerrit.wikimedia.org/r/86278 (owner: 10Ori.livneh) [18:38:26] (03Merged) 10jenkins-bot: Apache service: hasstatus => true [vagrant] - 10https://gerrit.wikimedia.org/r/86278 (owner: 10Ori.livneh) [18:38:36] (03PS1) 10Aaron Schulz: Hide filebackend listing exceptions for thumbnail purges [core] - 10https://gerrit.wikimedia.org/r/86286 [18:38:39] (03PS1) 10Ori.livneh: update-guest-additions: allow version to be specified as command-line argument [vagrant] - 10https://gerrit.wikimedia.org/r/86287 [18:38:48] ^ brion, fyi (doesn't need review) [18:38:58] whee [18:39:27] (03CR) 10Ori.livneh: [C: 032] update-guest-additions: allow version to be specified as command-line argument [vagrant] - 10https://gerrit.wikimedia.org/r/86287 (owner: 10Ori.livneh) [18:39:33] (03Merged) 10jenkins-bot: update-guest-additions: allow version to be specified as command-line argument [vagrant] - 10https://gerrit.wikimedia.org/r/86287 (owner: 10Ori.livneh) [18:39:34] * Reedy finds something to throw at Nemo_bis [18:40:10] (03CR) 10Ori.livneh: [C: 032] Use $.each instead of [].forEach to avoid distracting reviewers [extensions/NavigationTiming] - 10https://gerrit.wikimedia.org/r/86189 (owner: 10Ori.livneh) [18:40:29] (03Merged) 10jenkins-bot: Use $.each instead of [].forEach to avoid distracting reviewers [extensions/NavigationTiming] - 10https://gerrit.wikimedia.org/r/86189 (owner: 10Ori.livneh) [18:46:29] Reedy: there should be plenty of choice, both for objects and reasons to throw them; what was your choice here? :) [18:46:42] * Nemo_bis now hides below the desk [18:46:46] Cars are always a fun choice [18:46:53] expensive [18:46:56] bd808|LUNCH: http://ganglia.wikimedia.org/latest/graph.php?r=day&z=xlarge&c=Text+caches+eqiad&h=cp1038.eqiad.wmnet&jr=&js=&v=402856450&m=vhtcpd_inpkts_recvd&vl=pkts&ti=Multicast+packets [18:47:01] Nemo_bis: You've caused a lot of noise in the logs [18:47:09] I have some boxes full of small toy cars though [18:47:11] Mostly when they're blocked from editing by globalblocking [18:47:21] yay, blocking works :P [18:47:37] Though, one round of errors have been fixed [18:47:37] your fixes have to be backported I'm afraid? [18:47:43] only one? :O [18:47:48] 1 round [18:47:50] I thought you merged the last call too [18:47:52] which was 3 lines of code [18:48:21] (03PS6) 10MarkTraceur: Add meta=filerepoinfo for getting file repo info [core] - 10https://gerrit.wikimedia.org/r/85344 [18:48:24] anomie: Let's try that again. :) [18:49:01] We have too many repos [18:49:20] Reedy: Yo dawg I herd you like repos. [18:49:29] Imma rm -rf them [18:49:50] <^d> MatmaRex: Huh? [18:51:20] ^d: Reedy was complaining that gerrit doesn't require fast-forward merges. [18:51:32] also [18:51:34] @param $context RequestContext available since 1.18, will become compulsory in 1.19. [18:51:40] guess if that parameter is compulsory. [18:51:46] s/1.19/1.23/ [18:51:47] Of course not [18:51:58] <^d> I turned that on for operations/puppet awhile ago but it sucked. [18:51:58] the comment actually says "available since 1.18, will become compulsory in 1.18" [18:52:06] <^d> We had to turn it back off. [18:52:09] <^d> Way too much rebasing. [18:52:14] because of some overeager search&replace in 2011 [18:52:22] BY REEDY. [18:52:25] gj [18:52:32] {{sofixit}} [18:52:35] And he would've gotten away with it too! [18:52:54] * Reedy blames git blame [18:52:55] i'll make it say 1.23, then submit a commit that makes it to be merged in november. [18:53:02] <^d> If it weren't for that meddling MatmaRex! [18:53:11] heh [18:53:21] also [18:53:33] i think i managed to rip out 50 lines of hairy code from Preferences [18:53:35] (03PD4) 10Bartosz Dziewoński: [WIP] Rip out crappy user preferences code [core] - 10https://gerrit.wikimedia.org/r/86282 [18:53:41] i'm the best ^ [18:53:47] <^d> PD4? [18:53:49] * marktraceur looks at it skeptically [18:53:52] <^d> Shouldn't that be PS? [18:54:00] ^d: i published a draft change, so dunno. [18:54:14] <^d> Oh well. [18:54:22] MatmaRex: O NO U DIDINT [18:54:29] <^d> Wonder why I don't see PS1-3. [18:54:44] I think AzaToth added that code [18:55:08] ^d: because i deleted them. [18:55:12] i just learned that you can [18:55:25] DAMN IT [18:55:28] so you won't see jenkins failures for when i made a stupid typo and didn't run my code ~ [18:55:30] Nemo_bis: I just found another one [18:55:31] *glare* [18:55:40] * bawolff mutters stupid feature... :P [18:55:55] anyway guys, review it if you're looking at that already [18:56:00] people shouldn't be able to rewrite history [18:56:14] wfRunHooks( 'TorBlockBlockedMsg', &$blockedMsg ); [18:56:14] $hookError = array( 'permissionserrors', $blockedMsg, array( $ip ) ); [18:56:18] it will work assuming we don't have invalid data in user_options table [18:56:18] we mostly threw that out when we started reccomending --amend [18:56:21] we don't, right? [18:56:25] s/reccomending/requiring/ [18:56:27] Ignore that [18:56:42] MatmaRex: :P [18:56:45] /ignore that [18:56:46] :D [18:57:05] <^d> You're only rewriting local history with --amend [18:57:14] yeah, it's not like you amend and then push... [18:57:16] oh wait [18:57:18] (03PS1) 10Reedy: Fix error in call to wfRunHooks() [extensions/TorBlock] - 10https://gerrit.wikimedia.org/r/86291 [18:57:27] <^d> YuviPanda: You amend, push, gerrit treats it as a new PS. [18:57:27] * Nemo_bis whistles [18:57:31] <^d> And only one goes in to master. [18:57:37] <^d> It's not like your amending and pushing to master. [18:57:39] <^d> Which *is* evil. [18:57:43] indeed [18:57:53] well, what gerrit does is evil too, IMO [18:58:03] but there's no point in ranting about that :P [18:58:14] i'll just go to sleep instead. [18:58:34] hmmm [18:58:53] i wonder if i should use wfWarn() or wfLogWarning() for that [18:59:09] i guess we do want those warning in production, so let's go with the second one [18:59:43] ori-l, is there an easy way to disable debug once its enabled? [18:59:57] i suspect that turning role off wouldn't change config [19:00:17] night [19:04:07] marktraceur: I didn't know you could sanely do functions-inside-of-functions in PHP. [19:04:57] anomie: I didn't know you could sanely do anything in PHP [19:05:09] But I tried this and it doesn't appear to crash. :P [19:05:24] you can have functions inside functions? [19:05:40] doesn't that just define a function in the outer scope? (or are you talking about closures?) [19:05:43] (03CR) 10CSteipp: [C: 032] "(1 comment)" [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/86285 (owner: 10Anomie) [19:05:49] (03Merged) 10jenkins-bot: Adjust styling on jQuery authorize dialog [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/86285 (owner: 10Anomie) [19:06:20] marktraceur: I had thought they'd be global, but I tested and it seems to work [19:06:47] Will someone please merge https://gerrit.wikimedia.org/r/86291 as it's a very simple commit [19:06:54] MatmaRex: I do use closures in this patch, but no, we're talking about a function within a function [19:07:01] anomie: Meaning they aren't global? [19:07:27] marktraceur: At least, they don't turn up in get_defined_functions() [19:07:32] Excellent [19:08:00] (03CR) 10Chad: [C: 032 V: 032] Fix error in call to wfRunHooks() [extensions/TorBlock] - 10https://gerrit.wikimedia.org/r/86291 (owner: 10Reedy) [19:08:11] <^d> Reedy: ^ [19:08:58] (03PS1) 10Reedy: Fix up noise in wfArrayDiff2_cmp due to messages being wrapped in an array [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/86296 [19:09:00] (03CR) 10jenkins-bot: [V: 04-1] Fix up noise in wfArrayDiff2_cmp due to messages being wrapped in an array [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/86296 (owner: 10Reedy) [19:09:23] Thanks [19:09:51] (03PS1) 10Reedy: Update TorBlock to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86297 [19:11:43] (03PS1) 10Reedy: In wfArrayDiff2_cmp only compare $a and $b as string if they both are strings [core] - 10https://gerrit.wikimedia.org/r/86298 [19:12:13] (03CR) 10Reedy: [C: 032 V: 032] Update TorBlock to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86297 (owner: 10Reedy) [19:15:31] (03PS2) 10Reedy: In wfArrayDiff2_cmp only compare $a and $b as string if they both are strings [core] - 10https://gerrit.wikimedia.org/r/86298 [19:15:46] marktraceur: $props = array_flip( $params['prop'] ); and then array_intersect_keys( $repo->getInfo(), $props ); ;) [19:15:53] Reedy: are you going to be around in about 45 mins? and if so, can you look over a change to set ukwikimedia to read only except for user talk pages? I can commit the change earlier and you can tell me where it's wrong/overkill/etc... if you have time. [19:16:04] Grrr [19:16:12] anomie: I was trying to figure out how to do something like that [19:16:19] <^d> apergos: Sup with ukwikimedia? [19:16:25] they're moving it [19:16:30] rt ticket's been pending for awhile [19:16:33] <^d> Oh? [19:16:37] uh huh, off site [19:16:47] <^d> Can I close the "search sucks for ukwikimedia" as invalid then? [19:16:51] yes! [19:17:00] <^d> (I really hope they're not moving because of that!) [19:17:03] haha [19:17:10] supposed to leave user talk open so people can request their accounts be recreated at the new site [19:17:22] I don't actually know the back story [19:17:32] Do we get to treat them like the child that ran away? ;) [19:17:46] Kelson is in wikimedia-tech and could tell you [19:17:46] <^d> Whoops. [19:17:50] <^d> uk != ua [19:17:55] uh no it isn't [19:18:02] (03PS1) 10Nikerabbit: Remove sortable as it makes the table not functional [extensions/InviteSignup] - 10https://gerrit.wikimedia.org/r/86300 [19:18:05] <^d> ua.wikimedia.org is the one I have a bug for. [19:18:09] ah too bad [19:18:31] so what I did was um [19:18:53] anomie: Are you sure array_intersect_keys is what you mean? [19:19:03] (03CR) 10Anomie: [C: 04-1] "(2 comments)" [core] - 10https://gerrit.wikimedia.org/r/85344 (owner: 10MarkTraceur) [19:19:45] Now that you ask, no... [19:19:47] in InitialiseSettings I have a stanza that turns off most rights (move, edit, etc) for *, user, sysop, medlem, (dunno what that is), and leaves em in place for crat, also leaves edit in place for user [19:20:16] then I also have editallpages = true for crats only, and finally [19:20:27] marktraceur: Ah, it's array_intersect_key, no 's' [19:20:39] No, I don't think that does what we want [19:20:40] (03CR) 10Amire80: [C: 032] Remove sortable as it makes the table not functional [extensions/InviteSignup] - 10https://gerrit.wikimedia.org/r/86300 (owner: 10Nikerabbit) [19:20:55] Oh, damn it [19:20:56] It does [19:21:01] * marktraceur fails at reading things [19:21:08] for wgNamespaceProtection I have everything set to editallpages for ukwikimedia [19:21:16] except for NS_USER_TALK [19:21:22] I dunno if that is the right approach though [19:21:55] <^d> apergos: Hard to tell without code in front of us. Push for review :) [19:21:55] I can just push it for review if that's easier than discussing it without seeing it [19:21:58] <^d> Yeah [19:21:58] sure :-D [19:23:05] anomie: I think I'm going to keep the helper...seems more understandable [19:23:25] marktraceur: ok [19:24:12] Actually, no, now that I look at it again, never mind [19:24:15] * ^d throws his patch into the channel again [19:24:16] <^d> https://gerrit.wikimedia.org/r/#/c/86151/ [19:24:27] <^d> Maybe someone will trip over it :) [19:24:27] (03PS1) 10Bartosz Dziewoński: HTMLForm: Deprecate lack of $context in constructor and #setTitle [core] - 10https://gerrit.wikimedia.org/r/86304 [19:24:28] (03PS1) 10Bartosz Dziewoński: HTMLForm: Disallow lack of $context in constructor, remove #setTitle [core] - 10https://gerrit.wikimedia.org/r/86305 [19:24:30] 'cause I don't use the props array other than to filter, so I can flip it and be done [19:24:46] https://gerrit.wikimedia.org/r/#/c/86303/ [19:24:56] now you'll see what I mean and be able to tell me how it should be [19:25:17] I left it with crat edit anywhere because I think they want to put up a giant notice [19:25:20] 'We're moving' etc [19:25:32] (03PS7) 10MarkTraceur: Add meta=filerepoinfo for getting file repo info [core] - 10https://gerrit.wikimedia.org/r/85344 [19:26:10] <^d> apergos: lgtm, minus some weird indentation on the 'importupload' => false, lines. [19:26:22] <^d> I'll let Reedy +2 though [19:26:25] woops [19:26:49] <^d> And on that note, lunch :) [19:27:37] (03CR) 10Amire80: [V: 032] Remove sortable as it makes the table not functional [extensions/InviteSignup] - 10https://gerrit.wikimedia.org/r/86300 (owner: 10Nikerabbit) [19:27:49] fixing the tabs [19:28:22] (03PS2) 10Reedy: Fix up noise in wfArrayDiff2_cmp due to messages being wrapped in an array [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/86296 [19:28:30] Woops [19:28:35] anomie: Ignore that patch for a sec [19:28:44] done [19:28:49] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/85161 (owner: 10Krinkle) [19:29:05] (03PS8) 10MarkTraceur: Add meta=filerepoinfo for getting file repo info [core] - 10https://gerrit.wikimedia.org/r/85344 [19:29:08] anomie: Turns out scoping is hard [19:30:17] Reedy: https://gerrit.wikimedia.org/r/86305 [19:34:25] (03CR) 10Parent5446: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [19:35:30] (03CR) 10Bartosz Dziewoński: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [19:35:54] MatmaRex: I presume that isn't to be merged now? [19:35:56] (03PS2) 10Anomie: Use HTTPS for Special:MWOAuth/token [extensions/OAuth] - 10https://gerrit.wikimedia.org/r/85218 [19:36:19] 8 useages of setTitle() in core [19:36:24] Another 73 across extensions [19:36:58] Reedy: …what's a couple of warning in production between friends? [19:37:30] it was supposed to be killed two years ago, eh [19:37:45] (03CR) 10Parent5446: [C: 031] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [19:38:15] MatmaRex: You might want to find a stick to poke Ryan_Lane... 45 of them are in OSM [19:39:11] (03CR) 10Reedy: [C: 04-1] "8 useages of setTitle in core need to go too before this can be merged" [core] - 10https://gerrit.wikimedia.org/r/86305 (owner: 10Bartosz Dziewoński) [19:39:35] (03CR) 10Anomie: [C: 04-1] "(3 comments)" [core] - 10https://gerrit.wikimedia.org/r/85344 (owner: 10MarkTraceur) [19:45:21] MatmaRex, Reedy: what is? [19:45:59] Ryan_Lane: HTMLForm#setTitle [19:46:37] Reedy: the removal thing i'm hoping to get in 1.23, as promised in the dependency which deprecates them [19:46:37] what am I supposed to use otherwise? [19:46:47] Ryan_Lane: pass context to constructor as second parameter [19:46:49] (03CR) 10Krinkle: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [19:47:14] context? [19:47:17] which context? [19:47:33] see, the problem here is that there's relatively no documentation for html form at all [19:47:50] is this documented, like anywhere at all? [19:48:22] Heh, context stuff arrived in 1.18 [19:48:22] :D [19:48:23] (03CR) 10Parent5446: [C: 031] "Haven't tested." [core] - 10https://gerrit.wikimedia.org/r/83047 (owner: 10Krinkle) [19:49:31] I have to guess at how to use this crap by looking at the code and other extensions [19:49:42] Ryan_Lane: on mw.org, a little [19:49:52] 1 line you can just remove [19:49:54] https://www.mediawiki.org/wiki/HTMLForm [19:50:00] the constructor line would change [19:50:03] (03CR) 10Krinkle: [C: 04-2] "I assume this is supposed to be blocked until there actually is 1.23alpha and we branch REL1_22 (also, the releaese notes file heading rea" [core] - 10https://gerrit.wikimedia.org/r/86305 (owner: 10Bartosz Dziewoński) [19:50:07] hahaha [19:50:10] The "hardest" thing is going to be working out where to get the context object from [19:50:11] there's actually quite a lot of docs for the particular options [19:50:15] the example uses setTitle [19:50:32] Reedy: yes, that's what I'd like to know :) [19:50:42] $htmlForm->setTitle( $this->getTitle() ); [19:50:45] :D [19:50:48] killing me [19:51:05] <^d> Reedy: Getting a context is easy! [19:51:11] <^d> RequestContext::getMain() [19:51:12] <^d> Done [19:51:25] (03CR) 10Bartosz Dziewoński: [C: 04-2] "Yes, this is waiting for 1.23. I just added the release notes file as a note to self, I'll fix it before this is merged." [core] - 10https://gerrit.wikimedia.org/r/86305 (owner: 10Bartosz Dziewoński) [19:52:29] can we just put a requirement into review that features have to be documented to be accepted? [19:52:41] :) [19:53:00] Ryan_Lane: this was supposed to happen in 1.19, y'know [19:53:04] the removal stuff [19:53:09] i just found the comment now [19:53:36] I'm fine with it being reviewed. but the htmlform example linked actually does it incorrectly [19:53:47] err [19:53:52] with it being removed [19:54:19] {{sofixit}}? <3 [19:54:21] I'm more than happy to fix my code for this, assuming there's an example of what I'm supposed to do [19:54:24] eh, i might even fix it myself, maybe [19:54:28] MatmaRex: dude. you assume I know how. [19:54:55] that example from ye olde days, it seems [19:55:15] I'd be happy to fix it if I knew what I needed to do ;) [19:55:42] Ryan_Lane: fixed the example [19:55:45] thanks [19:55:56] i don't think it'll fatal :P [19:56:05] hahahaha [19:56:06] well, 'fixed' [19:56:11] that uses $wgOut [19:56:11] stabby stabby stab [19:56:12] :) [19:56:15] why does it even do that [19:56:20] if it's a SpecialPage [19:56:21] wtf [19:57:28] fixed [19:58:05] argh, there are more examples [19:58:12] who wrote all that useless documentation! [19:58:43] oh Reedy... got a few minutes now for uk wikimedia? [19:59:09] (03PS1) 10Amire80: Adding initial scenarios for Main Page in source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86309 [20:02:11] ori-l: \o/ Joy! [20:03:32] apergos: Where's the "medlem" group come from? [20:04:16] hmm se.wikimedia where I cribbed that from :-D [20:04:37] lemme lose that! [20:05:42] try that [20:05:59] note that I have no idea if I'm covering all the bases here or too many or too few [20:08:23] Reedy: [20:12:03] (03PS2) 10Bartosz Dziewoński: HTMLForm: Deprecate lack of $context in constructor and #setTitle [core] - 10https://gerrit.wikimedia.org/r/86304 [20:13:43] (03CR) 10Rillke: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/74391 (owner: 10TheDJ) [20:13:50] (03PS1) 10Spage: Trivial jshint cleanup [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86311 [20:15:30] (03PS2) 10Spage: Explanatory comment for gerrit 80501 [extensions/Campaigns] - 10https://gerrit.wikimedia.org/r/82443 [20:15:33] apergos: Looks like it's a bit verbose, but should be fine [20:16:40] you wanna merge it and I'll shovel it around? (sync-file still works right?) [20:19:53] (03PS3) 10Spage: Explanatory comment for gerrit 80501 [extensions/Campaigns] - 10https://gerrit.wikimedia.org/r/82443 [20:20:18] (03CR) 10Spage: "Sure, sorry about the delay." [extensions/Campaigns] - 10https://gerrit.wikimedia.org/r/82443 (owner: 10Spage) [20:21:59] Reedy: ok what did you change in the last patchset? I couldn't tell [20:22:26] removed a trailing tab [20:22:34] ah thx [20:22:46] so sync-file will do the job? [20:23:22] yup [20:27:16] anyone know how to reset the cache that holds rendered recentchanges ? Doing some work on custom recentchanges stuff, but the OldChangesListRecentChangesLine hook only runs the first reload. Bouncing memcached doesn't do it, so not sure where its cached [20:29:21] ebernhardson: as last resort, php maintenance/update.php should reset all caches [20:29:52] my turn now – how do i dup/clone a RequestContext? [20:30:28] DerivativeContext I think [20:30:43] There's some mechanism by which you can tell it what bits should and shouldn't be cloned I think [20:30:50] ah [20:30:56] let's see, thanks [20:31:16] heh, even after maintenance/update.php it still doesn't call the hook :S [20:31:26] (03CR) 10CSteipp: "Sorry, I meant line 439, so something like" [extensions/CleanChanges] - 10https://gerrit.wikimedia.org/r/86080 (owner: 10Siebrand) [20:31:28] but it gets called if i do something that creates a new recentchanges line [20:31:50] ebernhardson: what exactly do you need? [20:31:52] or when? [20:32:06] just for development? [20:32:17] aude: basically, i want to change some code and reload Special:RecentChanges and see the change [20:32:20] aude: yes developemnt [20:32:29] (03CR) 10Catrope: [C: 032] CSSJanus: Support text-shadow and box-shadow flipping [core] - 10https://gerrit.wikimedia.org/r/52611 (owner: 10Bartosz Dziewoński) [20:32:45] ok [20:33:16] but its only calling my hook when i reload recentchanges if i've inserted a new recentchanges row since the last time it was reloaded [20:33:18] i haven't come up with a better way but can request https://test.wikidata.org/w/index.php?title=Special:RecentChanges&limit=50 [20:33:21] then https://test.wikidata.org/w/index.php?title=Special:RecentChanges&limit=49 [20:33:26] ahh :) [20:33:30] or something different each time :) [20:34:03] to see how it renders based on what i am coding [20:34:18] aude: that does the trick, thanks [20:34:21] yay! [20:35:18] (03Merged) 10jenkins-bot: CSSJanus: Support text-shadow and box-shadow flipping [core] - 10https://gerrit.wikimedia.org/r/52611 (owner: 10Bartosz Dziewoński) [20:40:02] anomie: For some reason array_unique appears to have nil effect. [20:42:52] (03PS1) 10CSteipp: Redirect to https for signup [core] - 10https://gerrit.wikimedia.org/r/86314 [20:43:50] (03CR) 10Bartosz Dziewoński: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [20:45:07] marktraceur: Is it being cached in memcache? [20:45:25] I turned memcached off just now [20:45:33] Still returning bad values [20:48:46] (03CR) 10PleaseStand: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/86246 (owner: 10Krinkle) [20:51:01] Oh, hah, I had enabled it twice [20:51:03] Now it's off [20:51:09] Better [20:51:27] anomie: Is there a reason for wrapping array_unique with array_values? [20:52:44] marktraceur: array_unique( array( "a", "a", "b", "a", "c" ) ) => array( 0 => "a", 2 => "b", 4 => "c" ), which sometimes confused things. Maybe it's not necessary to regularize that in this case though. [20:52:53] Elsie: the documentation is wrong, its (by default) User:MyUserName/Echo-whitelist [20:53:02] Elsie: (feel free to fix) [20:53:10] although i suppose might be worth checking whatever production config is for the $wgEchoPerUserWhitelistFormat vairable [20:53:14] anomie: Ah, can't hurt [20:53:44] (03PS9) 10MarkTraceur: Add meta=filerepoinfo for getting file repo info [core] - 10https://gerrit.wikimedia.org/r/85344 [20:53:45] I'm so over this patch. :P [20:54:28] maaaan [20:54:38] why do we have eight hits for $form->setTitle in special pages [20:54:52] why don't people just use FormSpecialPage [20:55:14] now i'm going to replace a nasty hack with another nasty hack in eight places [20:56:14] Yay [20:56:15] MatmaRex: I recall looking at FormSpecialPage once, and found it insufficiently flexible to do what I needed to do. Details escape me at the moment, though. [20:56:17] Nasty hacks [20:56:43] MatmaRex: so, how about not writing nasty hacks? :) [20:56:50] MatmaRex: replace 8 occurances with something elegant [20:57:14] ebernhardson: You've mistaken us for Haskell programmers [20:57:17] This is PHP. [20:58:04] marktraceur: i think you will find there is a ton of elegant php out there, just not in mediawiki [20:58:08] ebernhardson: i think SpecialPage already has enough methods :P [20:58:13] marktraceur: which is to say, you all could write elegant php if you cared enough :) [20:58:18] MatmaRex: thats not elegant [20:58:19] anomie: hmm, yeah, it might be a little inflexible, but it's workable [20:58:25] MatmaRex: instead of continuing hacks, invent something actually elegant :) [20:58:34] ebernhardson: that's the elegantest i can do without rewriting seven classes [20:58:38] :) [20:58:47] MatmaRex: sounds like a cop out [20:58:56] rewrites take a lot of time to get merged [20:59:00] "elegantest" I think MatmaRex wins the Fake Word of the Day contest [20:59:01] and i need this done in time for 1.22 [20:59:27] (and they also take a lot of time to write, and academic year starts this monday) [20:59:59] MatmaRex: Aw, will you be around much less? [21:00:08] shrug ... just as long as your clear that the reason the code sucks is because you dont care enough to fix it :) [21:00:37] (i fall into the same boat regularly, not hatin ) [21:02:37] ebernhardson: shall https://gerrit.wikimedia.org/r/#/c/86282/ be my atonement [21:02:58] (feel free to review :P) [21:03:04] Jorm still the guy for echo/notifications? [21:03:13] (i haven't really tested if it works, though) [21:03:26] T13|needsCoffee: depends on what you mean by 'the guy'. for most definitions, no [21:03:57] T13|needsCoffee: jorm is a designer, not sure if thats what your looking for :) [21:03:58] T13|needsCoffee: echo is run by fabrice florin and primarily programmed by benny situ [21:04:26] Fabrice is off Echo; Maryana is now the product owner. [21:04:35] I'm wondering if [[Special:MyPage/Echo-blacklist]] was technically feasible. [21:04:40] The PM is dead; long live the PM. [21:04:53] that's a benny situ question. [21:04:57] (almost worth bashing that) [21:05:35] <^d> RoanKattouw: https://gerrit.wikimedia.org/r/#/c/86151/ is the change :) [21:05:52] Already looking at it :) [21:06:11] T13|needsCoffee: a) i think the docuemntation is wrong, because echo whitelists are at User:MyUserName/Echo-whitelist [21:06:15] <^d> RoanKattouw: I can show you before/after with hhvm if you want :) [21:06:19] (03PS1) 10Bartosz Dziewoński: ttlesadssafda [core] - 10https://gerrit.wikimedia.org/r/86318 [21:06:20] (03PS1) 10Bartosz Dziewoński: Remove calls to deprecated HTMLForm#setTitle [core] - 10https://gerrit.wikimedia.org/r/86319 [21:06:25] wooot [21:06:26] And he's not on irc so I'd have to create a ticket on Bugzilla for a feature I don't particularly care about but would solve a question on en:VPT [21:06:35] ttlesadssafda was not supposed to go there. [21:06:38] T13|needsCoffee: b) should be quite easy to reuse the classes i created for whitelisting to do black listing (i wrote that part of echo) [21:06:56] T13|needsCoffee: also it already has blacklisting, but the feature requirements were global blacklist and per-user whitelist [21:07:28] (03PS2) 10Bartosz Dziewoński: Remove calls to deprecated HTMLForm#setTitle [core] - 10https://gerrit.wikimedia.org/r/86319 [21:07:44] ebernhardson: that's the "nasty" patch ^ [21:07:57] ebernhardson: there is currently a [[MediaWiki:Echo-blacklist]] and a [[Special:MyPage/Echo-whitelist]] but no per user [[Special:MyPage/Echo-blacklist]]... [21:08:02] (03Abandoned) 10Bartosz Dziewoński: ttlesadssafda [core] - 10https://gerrit.wikimedia.org/r/86318 (owner: 10Bartosz Dziewoński) [21:09:04] T13|needsCoffee: correct, as said the feature requirement at the time was global blacklist and per-user whitelist [21:09:39] T13|needsCoffee: but since i tried to write that as a non-hack, I'm 99% that you can reuse the existing classes that implement white/blacklist to get a per-user blacklist (i suppose its not clear, were you looking to impelment the change? i was thinking yes because this is -dev) [21:09:44] This came about from a question about suppressing notifications from Elsie EdwardsBot when it posts on user talk pages. [21:10:23] ebernhardson: I was the ine that wrote the ticket formally requesting it iirc.. [21:10:57] Not yet to answer your question. [21:11:14] I'd rather see the bot on the global blacklist. [21:11:17] (03CR) 10Bartosz Dziewoński: "Removing 8 calls to deprecated #setTitle method in https://gerrit.wikimedia.org/r/#/c/86319/ ." [core] - 10https://gerrit.wikimedia.org/r/86304 (owner: 10Bartosz Dziewoński) [21:11:31] (03CR) 10Bartosz Dziewoński: "Reedy: https://gerrit.wikimedia.org/r/#/c/86319/" [core] - 10https://gerrit.wikimedia.org/r/86305 (owner: 10Bartosz Dziewoński) [21:12:06] T13|needsCoffee: getting it on the global blacklist doesn't invlove us though :) [21:12:28] (03CR) 10Nikerabbit: [C: 032] Adding initial scenarios for Main Page in source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86309 (owner: 10Amire80) [21:12:37] (03Merged) 10jenkins-bot: Adding initial scenarios for Main Page in source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86309 (owner: 10Amire80) [21:12:58] Right. I just wanted to know how much of an issue it would be to offer that if the community refuses to put it on global blacklist. ;) [21:13:06] T13|needsCoffee: not hard at all [21:13:19] That's all I was wondering. :) [21:13:52] If I end up having to create a ticket I'll CC you and ben situ.. :) [21:14:14] T13|needsCoffee: basically, the same 12 lines as whitelist( https://github.com/wikimedia/mediawiki-extensions-Echo/blob/master/controller/NotificationController.php#L255 ) but with a ! and rename white to black [21:14:34] T13|needsCoffee: i have a feeling the no per-user blacklist might have been a product decision though, it certainly wasn't a technical limitation [21:15:33] Looking. [21:17:35] (03CR) 10Reedy: [C: 032] Fix up noise in wfArrayDiff2_cmp due to messages being wrapped in an array [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/86296 (owner: 10Reedy) [21:17:38] (03Merged) 10jenkins-bot: Fix up noise in wfArrayDiff2_cmp due to messages being wrapped in an array [extensions/WikimediaMessages] - 10https://gerrit.wikimedia.org/r/86296 (owner: 10Reedy) [21:19:49] (03PS1) 10Reedy: Update WikimediaMessages to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86321 [21:20:07] Looks straightforward enough. [21:20:10] :) [21:20:29] T13|needsCoffee: thats what happens when people build reusable code instead of piling hacks ontop of hacks ) [21:21:03] * ebernhardson piles the hacks on regularly though :P [21:21:27] No comment. lol Must be some WFM internal discussion I've no interest in knowing more about. :p [21:22:13] heh, it was the convo right before you poped in this channel, a few people implied the only way to write php was hacks ontop of hacks [21:22:20] S/WFM/WMF/ [21:22:51] (03CR) 10Reedy: [C: 032 V: 032] Update WikimediaMessages to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86321 (owner: 10Reedy) [21:22:59] Only if you work for Microsoft are hacks on hacks on hacks acceptable.... [21:23:07] * T13|needsCoffee wanders off... [21:23:10] grrrit-wm: Why didn't you announce my new patchset...weird. [21:23:40] <^d> He doesn't like you. [21:24:14] Prolly. [21:26:04] (03PS2) 10Mgrover: Changed from str to message [tools/release] - 10https://gerrit.wikimedia.org/r/86210 [21:42:38] (03CR) 10Hashar: [C: 04-1] "We don't bother cleaning up the files between request, I am not sure why we should do it between tests. That adds a bunch of overhead." [core] - 10https://gerrit.wikimedia.org/r/86218 (owner: 10Aaron Schulz) [21:42:43] (03PS1) 10Ori.livneh: Update NavigationTiming to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86323 [21:43:29] (03PS1) 10Ori.livneh: Update NavigationTiming to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86324 [21:44:26] (03CR) 10Ori.livneh: [C: 032] Update NavigationTiming to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86323 (owner: 10Ori.livneh) [21:44:34] (03CR) 10Ori.livneh: [C: 032] Update NavigationTiming to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86324 (owner: 10Ori.livneh) [21:51:16] (03Merged) 10jenkins-bot: Update NavigationTiming to master [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86323 (owner: 10Ori.livneh) [21:51:39] (03Merged) 10jenkins-bot: Update NavigationTiming to master [core] (wmf/1.22wmf18) - 10https://gerrit.wikimedia.org/r/86324 (owner: 10Ori.livneh) [21:51:39] (03CR) 10Ori.livneh: [C: 032] less: Add '@since 1.22' comments [core] - 10https://gerrit.wikimedia.org/r/86248 (owner: 10Krinkle) [21:52:50] (03CR) 10Ori.livneh: "Except for the commit SHA1, the top comment block should be remain the same. Its format was specified by RobLa as a requirement for includ" [core] - 10https://gerrit.wikimedia.org/r/86246 (owner: 10Krinkle) [21:54:50] (03PS1) 10Nikerabbit: Improve html structure [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86325 [22:01:11] RoanKattouw: th ve 2 chr dir thng is rly wtf, i h8 it b/c i can nvr rmmbr where nethng is [22:01:44] y u no use wrds? [22:01:48] * brion_ pipes ori-l through gunzip [22:02:54] $ ls [22:02:55] ce dm init test ui [22:03:13] ಠ_ಠ [22:05:42] (03CR) 10Chad: [C: 032] "Roan says this looks good, so if it isn't, BLAME HIM!" [core] - 10https://gerrit.wikimedia.org/r/86151 (owner: 10Chad) [22:05:44] ori-l++ [22:05:56] <^d> See it's not a self-merge because I asked Roan ;-) [22:07:21] <^d> ori-l: Once ^ lands you can actually get through phpunit on hhvm with less than 25 pages of failures :) [22:08:03] nice! [22:08:26] (03Merged) 10jenkins-bot: Clean up language test cases objection construction [core] - 10https://gerrit.wikimedia.org/r/86151 (owner: 10Chad) [22:08:36] <^d> Next step: intl, dba_* and xdiff support. [22:08:51] <^d> Well, the latter two. I might ignore intl and hope someone else does it. [22:19:05] (03CR) 10Ori.livneh: [C: 032] Update cherry-picked LESSFunctions.php to final ResourceLoaderFunctions.php [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86198 (owner: 10Brion VIBBER) [22:22:04] (03Merged) 10jenkins-bot: Update cherry-picked LESSFunctions.php to final ResourceLoaderFunctions.php [core] (wmf/1.22wmf19) - 10https://gerrit.wikimedia.org/r/86198 (owner: 10Brion VIBBER) [22:23:15] (03CR) 10Amire80: [C: 032] Improve html structure [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86325 (owner: 10Nikerabbit) [22:23:26] (03Merged) 10jenkins-bot: Improve html structure [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86325 (owner: 10Nikerabbit) [22:26:31] (03PS6) 10Mwalker: Allow MW and extensions to declare included components [core] - 10https://gerrit.wikimedia.org/r/66054 [22:27:50] (03CR) 10Mwalker: [C: 04-1] "I've addressed Aude's comments -- but I realized I'm still deficient -- jslicense-labels needs to include information from extensions if a" [core] - 10https://gerrit.wikimedia.org/r/66054 (owner: 10Mwalker) [22:28:03] (03PS1) 10Nikerabbit: First two source language tests [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86328 [22:30:01] (03CR) 10Awjrichards: [C: 04-1] "(1 comment)" [tools/release] - 10https://gerrit.wikimedia.org/r/86210 (owner: 10Mgrover) [22:44:48] (03PS11) 10Mattflaschen: Towards OB6: show a popup CTA to test onboarders [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/77087 (owner: 10Spage) [22:45:05] (03CR) 10jenkins-bot: [V: 04-1] Towards OB6: show a popup CTA to test onboarders [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/77087 (owner: 10Spage) [22:47:49] AaronSchulz, MaxSem: https://www.mediawiki.org/wiki/User:GWicke/Notes/Storage#Key.2Fvalue_store_without_versioning [22:49:06] (03CR) 10Aaron Schulz: "How does the singleton get destroyed between tests?" [core] - 10https://gerrit.wikimedia.org/r/86218 (owner: 10Aaron Schulz) [22:49:27] (03PS7) 10Bsitu: Add primary/secondary link, email, bundling to post reply [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86064 [22:52:12] (03CR) 10Chad: [C: 032] Hide filebackend listing exceptions for thumbnail purges [core] - 10https://gerrit.wikimedia.org/r/86286 (owner: 10Aaron Schulz) [22:54:58] (03Merged) 10jenkins-bot: Hide filebackend listing exceptions for thumbnail purges [core] - 10https://gerrit.wikimedia.org/r/86286 (owner: 10Aaron Schulz) [22:58:10] ori-l: Perhaps Roan is unaware of tab completion. ;-) [22:59:38] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86064 (owner: 10Bsitu) [23:00:23] it's possible [23:02:11] Hi Elsie. :) [23:03:21] greg-g: I'm going to place a fix for CentralNotice into monday's LD slot [23:03:32] mwalker: cool [23:03:51] you know the drill :) [23:05:30] Hi T13|needsCoffee. [23:05:45] greg-g: Thanks for scheduling MassMessage. :-) [23:06:07] I had just bothered legoktm about doing that when he pointed out that you already had. [23:06:50] greg-g: yep yep; I feel like I'm the only one who uses them [23:06:56] but surely that cannot be the case [23:07:01] Elsie: :) [23:07:07] Elsie: anything for you, my dear [23:07:24] mwalker: 'tis not, you and yurik and ^d [23:07:27] legoktm: If you can be around during the deployment on Monday, that would be great. [23:07:35] I probably won't be. [23:07:42] I'll be here in spirit, of course. [23:07:50] ah; solidarity [23:08:35] I make my own lightning outside of the regular timeslots. [23:08:45] Time slots? [23:09:37] yes; you do seem to wander around holding a lighting rod [23:10:06] I've got the magic in me. [23:13:03] i hate php's bug tracker [23:13:04] https://bugs.php.net/bug.php?id=10203 [23:13:11] can somebody tell me what version is this fixed in? [23:15:48] lokes like 5.4.0+ [23:16:17] (03PS1) 10Nikerabbit: Test for project card language selector [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86333 [23:16:18] but uh [23:16:22] who knows :) [23:16:58] MatmaRex: PHP 5.5. http://3v4l.org/jqoKH [23:20:42] (03PS2) 10Nikerabbit: First three source language tests [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86328 [23:21:09] PleaseStand: heh. okay, thank you [23:21:17] (03Abandoned) 10Nikerabbit: Test for project card language selector [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86333 (owner: 10Nikerabbit) [23:23:52] (03CR) 10Bsitu: [C: 032] Trivial jshint cleanup [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86311 (owner: 10Spage) [23:24:01] (03Merged) 10jenkins-bot: Trivial jshint cleanup [extensions/Flow] - 10https://gerrit.wikimedia.org/r/86311 (owner: 10Spage) [23:26:05] !b 54647 [23:26:05] https://bugzilla.wikimedia.org/54647 [23:28:48] marktraceur: That bug was an adventure in learning new stuff for me. [23:28:57] bd808: Cool [23:29:12] Elsie: If it's after 1pm PDT, I'll be around [23:29:24] (03PS3) 10Nikerabbit: All source language tests implemented [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/86328 [23:35:39] (03PS3) 10MarkTraceur: Pull data from API, load title [extensions/MultimediaViewer] - 10https://gerrit.wikimedia.org/r/86063 [23:36:55] hmmmmmm [23:37:16] is it possible that the user_options table might contain options which will not validate with current code? [23:37:54] or, is that something we want to support (transparently restore defaults, i guess), or something that - if it actually does happen - the user could be asked to deal with on the preferences page? [23:38:15] because there is some code that supports that and i hate it [23:39:34] (03PS1) 10Yurik: Fix settings and documentation for remote_debug role [vagrant] - 10https://gerrit.wikimedia.org/r/86336 [23:55:37] (03PS5) 10Bartosz Dziewoński: Rip out hacky user preferences code [core] - 10https://gerrit.wikimedia.org/r/86282 [23:56:36] (03PS1) 10BryanDavis: Set $wgServer using WebRequest::detectServer(). [vagrant] - 10https://gerrit.wikimedia.org/r/86338 [23:58:35] (03PS1) 10PleaseStand: Deprecate wfDebugDieBacktrace [core] - 10https://gerrit.wikimedia.org/r/86339 [23:58:47] (03CR) 10Mattflaschen: [C: 032] Explanatory comment for gerrit 80501 [extensions/Campaigns] - 10https://gerrit.wikimedia.org/r/82443 (owner: 10Spage) [23:59:22] (03Merged) 10jenkins-bot: Explanatory comment for gerrit 80501 [extensions/Campaigns] - 10https://gerrit.wikimedia.org/r/82443 (owner: 10Spage)