[01:19:36] (03PS3) 10Dan-nl: addressing brian wolff’s comments [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/83657 [03:27:02] (03CR) 10Mwalker: [C: 032] Fix some bugs with the indentation patch. [extensions/BookManagerv2] - 10https://gerrit.wikimedia.org/r/84093 (owner: 10Mollywhite) [03:27:05] (03Merged) 10jenkins-bot: Fix some bugs with the indentation patch. [extensions/BookManagerv2] - 10https://gerrit.wikimedia.org/r/84093 (owner: 10Mollywhite) [03:28:00] (03CR) 10Parent5446: [C: 04-1] "(3 comments)" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [03:31:51] (03CR) 10Parent5446: [C: 04-1] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [03:49:52] (03PS1) 10Amire80: Show "For all languages" and no stats bar for source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84107 [03:50:23] (03PS2) 10Amire80: Show "For all languages" and no stats bar for source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84107 [04:13:10] (03PS3) 10Amire80: Show "For all languages" and no stats bar for source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84107 [04:28:06] http://www.dungeonsanddevelopers.com/ for anyone who enjoy skilltrees [04:28:25] http://www.analogmuseum.org/english/impressions/eai_pacer_1.jpg for anyone who appreciates fine design [04:41:51] Hi quiddity. [04:42:32] heyo. just doing the blog dive rounds... [04:43:04] thanks for the template at meta explaining the change, [04:43:24] No problem. [04:53:20] (03PS1) 10Ori.livneh: Consistently follow conventions for documenting parameters [core] - 10https://gerrit.wikimedia.org/r/84108 [04:54:58] (03CR) 10Ori.livneh: "Tyler: if this looks OK, go ahead and merge it, and I'll rebase Id052a04dd on top and make the comment tweaks you suggested." [core] - 10https://gerrit.wikimedia.org/r/84108 (owner: 10Ori.livneh) [04:59:40] (03CR) 10Ori.livneh: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [05:04:29] i hope he's mollified by that [05:08:55] ori-l: have some time to help me with a javascript issue? [05:09:12] maybe, what is it? [05:09:57] so in my user script, part of my code looks like https://dpaste.de/Eo5W2/raw/ [05:10:16] i can't get the autocomplete module to trigger [05:10:30] its never reaching the console.log part [05:10:56] so im not sure if i'm using the mw.loader.using thing right, and even with that it wasnt working either [05:12:22] well, console.log debugging is javascript's version of printf debugging, and that's an ancient and noble practice, don't let anyone tell you otherwise [05:12:48] so first, stick a console.log at the top of the callback you're passing to mw.using to find out if it succeeds [05:13:06] if it doesn't, you can do the following things: [05:13:59] 1) look up the source for 'using' in mediawiki.js and see if you're using it right. (does it take an extra parameter you could use, perhaps?) [05:14:50] 2) inspect the module state manually by calling mw.loader.getState('jquery.ui.autocomplete') [05:14:59] ok, it didnt work. [05:15:18] using has a third option, which is a callback upon error, let me try and see if that is the issue [05:16:28] hm, that didnt work either... [05:16:52] ok [05:17:06] so when i type it directly in my console, it works... [05:17:07] i wonder if [05:21:15] getState returns "ready" on a fresh reload [05:21:53] ori-l: so if i type that snippet in manually, it works fine, just not if it gets loaded along with the rest of the script... [05:22:16] what page are you executing it on? [05:22:34] https://www.wikidata.org/wiki/User:Legoktm/ADB.js is the full script [05:22:42] i'm running it on https://www.wikidata.org/wiki/User:Legoktm/Test?debug=true [05:24:33] works for me [05:25:09] http://i.imgur.com/DhNvSDr.png [05:25:34] er, select add claims, hit continue [05:25:47] then there will be another form, one of those is "Value". that box should have autocomplete enabled [05:25:54] thats the part that isnt working [05:26:18] what's something i could type into that form to test that? [05:26:23] Barack Obama [05:27:30] $('#qid-value').autocomplete({ [05:27:36] does '#qid-value' exist at that point? [05:27:59] uh, no it wouldn't [05:29:07] * legoktm rearranges [05:29:13] so move that whole block so that it's just below make_form( buttons, '#form2' ); [05:29:42] woot. [05:29:44] that was it :D [05:29:47] thanks! [05:30:20] np [05:33:07] (03PS4) 10Amire80: Show "For all languages" and no stats bar for source language [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84107 [05:33:08] (03PS1) 10Amire80: WIP: Show stats for all languages [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84109 [05:38:35] (03CR) 10Anomie: "This change appears to have exposed a bug in Chrome, see [[en:Wikipedia:Village pump (technical)#Scrolling past the bottom of the page...]" [extensions/Cite] - 10https://gerrit.wikimedia.org/r/79472 (owner: 10Hoo man) [06:32:39] (03PS1) 10Rfaulk: mod - path to private key + dormant code to process ini config. [sartoris] - 10https://gerrit.wikimedia.org/r/84110 [06:32:40] (03PS1) 10Rfaulk: mod - paramiko scp/ssh deps utilize private keys for auth. [sartoris] - 10https://gerrit.wikimedia.org/r/84111 [06:33:31] (03CR) 10Rfaulk: [C: 032 V: 032] mod - path to private key + dormant code to process ini config. [sartoris] - 10https://gerrit.wikimedia.org/r/84110 (owner: 10Rfaulk) [06:33:56] (03CR) 10Rfaulk: [C: 032 V: 032] mod - paramiko scp/ssh deps utilize private keys for auth. [sartoris] - 10https://gerrit.wikimedia.org/r/84111 (owner: 10Rfaulk) [12:14:16] (03CR) 10Helder.wiki: "FYI: this fixed bug 34743." [core] - 10https://gerrit.wikimedia.org/r/83834 (owner: 10Kaldari) [13:49:25] (03CR) 10Parent5446: [C: 031] "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [13:50:48] (03CR) 10Parent5446: [C: 032] Consistently follow conventions for documenting parameters [core] - 10https://gerrit.wikimedia.org/r/84108 (owner: 10Ori.livneh) [13:54:08] (03Merged) 10jenkins-bot: Consistently follow conventions for documenting parameters [core] - 10https://gerrit.wikimedia.org/r/84108 (owner: 10Ori.livneh) [14:19:31] (03CR) 10Bartosz Dziewoński: "(1 comment)" [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [14:36:31] (03PS2) 10Amire80: WIP: Show stats for all languages [extensions/TwnMainPage] - 10https://gerrit.wikimedia.org/r/84109 [14:38:30] (03PS4) 10Bartosz Dziewoński: Add a dropdown list for the tag selector [core] - 10https://gerrit.wikimedia.org/r/80781 [14:48:30] (03PS2) 10J: Add TitleMove hook [core] - 10https://gerrit.wikimedia.org/r/83995 [14:57:58] anyone online that knows how templatedata is working for Template:Artwork/doc on commons? i'm trying to set it up locally, but the template data section shows Syntax error in JSON with no further info on what may be wrong. i can create a test template with a section that works, but i can't find a section in Template:Artwork/doc, so it looks like it's using a different method for creating the section … ? [15:15:25] dan-nl: set what up? [15:15:36] dan-nl: can you paste source of a page that's failing to parse somewhere? [15:15:38] !pastebin [15:15:38] To avoid overflowing the channel with inane amounts of text, use https://dpaste.org/ or other awesome pastebin sites to share code, errors, and other large texts. [15:16:33] as for templatedata errors being extremely unhelpful, that's https://bugzilla.wikimedia.org/show_bug.cgi?id=50511 :/ [15:16:34] MatmaRex: you can see the error here http://gwtoolset.wmflabs.org/wiki/Template:Artwork/doc [15:18:02] dan-nl: huh… that's be easier to debug if everything wasn't magically autogenerated [15:18:30] * MatmaRex is digging [15:18:35] MatmaRex: :) i can turn on debugging … will that help? [15:19:19] dan-nl: no idea [15:19:30] dan-nl: what would help is installing the expandtemplates extension :) [15:19:40] !e ExpandTemplates [15:19:40] https://www.mediawiki.org/wiki/Extension:ExpandTemplates [15:19:51] MatmaRex: ah, ok … doing that now [15:21:15] (03Abandoned) 10Daniel Werner: Introduction of Content::getActualContent [core] - 10https://gerrit.wikimedia.org/r/83647 (owner: 10Daniel Werner) [15:23:41] MatmaRex: k, its installed [15:25:30] dan-nl: well, then it's simple, the generated templatedata tag is empty [15:25:34] don't ask me why [15:25:37] commons template magic, eh [15:25:49] dan-nl: (just paste the source into http://gwtoolset.wmflabs.org/wiki/Special:ExpandTemplates ) [15:30:45] MatmaRex: very strange indeed ... [15:31:26] MatmaRex: who would have set it up on commons? i must be missing some template, special page or other configuration(s) ... [15:31:51] (03CR) 10J: "(5 comments)" [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/78235 (owner: 10J) [15:32:04] (03PS2) 10J: Parse Ogg Opus files and enable playback if supported [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/78235 [15:32:20] dan-nl: i have no idea, sorry [15:32:30] MatmaRex: ok, thanks for trying [15:36:04] (03PS1) 10Brion VIBBER: Retarget to Windows 8.1 [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84128 [15:37:41] (03PS1) 10Brion VIBBER: quick fix for weird image failure mode [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84129 [15:37:42] (03PS1) 10Brion VIBBER: Tidy up margins in snapped mode [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84130 [15:37:43] (03PS1) 10Brion VIBBER: search pane -> search box [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84131 [15:37:44] (03PS1) 10Brion VIBBER: add a bit to that about page [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84132 [15:37:45] (03PS1) 10Brion VIBBER: Retarget from Windows 8.1 Preview to Windows 8.1 RTM [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84133 [15:37:46] (03PS1) 10Brion VIBBER: Pull README.md to root dir; add SVG originals of icons & tiles to scratch dir [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84134 [15:37:47] whee merging [15:38:10] (03CR) 10Brion VIBBER: [C: 032 V: 032] "(Windows 8.1 preview)" [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84128 (owner: 10Brion VIBBER) [15:38:30] (03CR) 10Brion VIBBER: [C: 032 V: 032] quick fix for weird image failure mode [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84129 (owner: 10Brion VIBBER) [15:39:06] (03CR) 10Brion VIBBER: [C: 032 V: 032] "Windows 8.1 changes snapped mode so it's no longer fixed size (or quite as tiny) -- minimum size is now 500px by default instead of 320px," [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84130 (owner: 10Brion VIBBER) [15:39:42] (03CR) 10Brion VIBBER: [C: 032 V: 032] "Windows 8.1 no longer recommends using the search charm for in-app search; a SearchBox control is recommended which has a similar interfac" [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84131 (owner: 10Brion VIBBER) [15:40:14] (03CR) 10Brion VIBBER: [C: 032 V: 032] "Source link needs updating -- will do in separate commit." [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84132 (owner: 10Brion VIBBER) [15:40:23] (03CR) 10Brion VIBBER: [C: 032 V: 032] Retarget from Windows 8.1 Preview to Windows 8.1 RTM [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84133 (owner: 10Brion VIBBER) [15:40:33] (03CR) 10Brion VIBBER: [C: 032 V: 032] Pull README.md to root dir; add SVG originals of icons & tiles to scratch dir [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84134 (owner: 10Brion VIBBER) [15:42:09] (03PS1) 10Brion VIBBER: [Bug 37278] Set X-Analytics header on XHR reqs [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84135 [16:51:44] (03PS2) 10Ori.livneh: Provide a dynamic global scope of LESS variables [core] - 10https://gerrit.wikimedia.org/r/84101 [16:53:23] (03CR) 10Ori.livneh: "Bartosz, yep! The RFC has details ." [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [17:08:34] (03PS7) 10Ori.livneh: Support LESS stylesheets in ResourceLoader [core] - 10https://gerrit.wikimedia.org/r/78669 [17:10:55] (03CR) 10Ori.livneh: "PS7 makes the doc changes Tyler suggested." [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [17:21:38] (03CR) 10Ori.livneh: [C: 032] [Bug 37278] Set X-Analytics header on XHR reqs [apps/win8/wikipedia] - 10https://gerrit.wikimedia.org/r/84135 (owner: 10Brion VIBBER) [17:23:20] hey brion [17:23:35] yo [17:23:57] woo, finally figured out how to enable the keyboard shortcuts for os x 'mission control' [17:24:09] can finally switch desktops with the keyboard when docked and don't have the trackpad handy [17:25:01] ah, cool [17:25:42] thoughts re: https://www.mediawiki.org/wiki/Requests_for_comment/LESS ? timo, roan & trevor approve, AFAIK. [17:26:50] hmm [17:27:03] ori-l: what's the implementation do? shell out, or a PHP implementation of the preprocessor? [17:27:11] the latter [17:27:26] you didn't read it through :P [17:27:49] i skimmed it [17:27:58] took me several times through to notice a mention of a php implementation of less [17:28:05] but the rfc doesn't say anything about whether it uses it [17:28:14] and the gerrit link to a draft implementation fails [17:28:24] * ori-l blushes [17:28:25] 'The page you requested was not found, or you do not have permission to view this page.' [17:28:45] https://gerrit.wikimedia.org/r/#/c/78669/ & https://gerrit.wikimedia.org/r/#/c/84101/ [17:29:12] whee [17:29:20] can you fix the link in the info box to point at those? [17:29:25] * brion reads code \o/ [17:29:25] yes, doing so [17:29:33] thanks :D [17:29:51] the php implementation is really nice: good docs, lots of tests, actively developed, merges patches quickly http://leafo.net/lessphp/docs/ [17:30:06] awesome [17:32:01] ori-l: any security concerns with importing include files etc? [17:33:13] hm -- what do you have in mind? LESS imports are handled on the server [17:33:56] @import "/etc/passwd" for starters :) [17:34:59] currently LESS support is only provided for ResourceLoaderFileModule [17:35:21] so you'd have to actually commit a file containing that payload to core or to an extension and have it merged [17:35:34] ok that helps for now :D [17:35:59] yeah, so it doesn't seem like a vector, i mean you can var_dump( file_get_contents( '/etc/passwd' ) ) too [17:36:06] yup [17:36:07] !htmllogs [17:36:07] http://tools.wmflabs.org/wm-bot/logs/index.php?display=%23mediawiki [17:36:18] it'll be something to consider if/when adding to gadgets/site scripts/etc [17:37:17] yeah. that should probably be noted somewhere. dunno where tho [17:39:36] you can configure the PHP compiler to disable imports (https://github.com/leafo/lessphp/blob/master/lessc.inc.php#L53) so we'd probably use an instance so configured in those cases [17:39:41] nice [17:40:02] ori-l: https://www.mediawiki.org/wiki/Requests_for_comment/LESS#.40import_and_ResourceLoader_dependencies ? hmm [17:40:18] do the less files have to explicitly hook up their deps internally? [17:41:16] yes, I think so. the second patch adds a global LESS variable scope, but otherwise the environment is not shared across different invocations of the LESS compiler [17:41:26] i guess https://gerrit.wikimedia.org/r/#/c/84101 tries to cover that sorta [17:41:28] yeah :) [17:41:47] that doesn't cover mixins, though [17:41:59] and not all variables should be declared globally [17:43:47] but i don't see a great solution for that.. if someone explicitly declares a module with styles => array( 'a.less', 'b.less' ) they might be unpleasantly surprised if the environment was shared [17:43:56] yeah [17:44:06] i'm just looking over the MobileFrontend less files to check what they're doing... [17:44:16] for mixing they're explicitly @import'ing [17:44:18] *mixins [17:44:26] damn you autocorrect [17:45:07] right, they have to, because RL doesn't have LESS support presently. it's a manual build step atm. [17:45:46] (03CR) 10Brion VIBBER: [C: 031] "Looks workable, haven't tested yet." [core] - 10https://gerrit.wikimedia.org/r/84101 (owner: 10Ori.livneh) [17:46:16] well even without the build step i think they'd need the @imports [17:46:18] i… think :D [17:46:43] right [17:47:44] (03CR) 10Brion VIBBER: [C: 031] "Looks workable, haven't tested yet." [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [17:47:55] ori-l: i might give a whirl at adapting MobileFrontend to run on this to test it [17:48:49] ori, you had a handy site for testing cookies that I only rememebered to bookmark after I forgot the URL. Do you still have it from the top of your head? [17:49:07] MartijnH: http://httpbin.org/ probably [17:49:18] yup, thanks [17:49:23] you're a live saver [17:49:58] :) [17:50:05] brion: that would be cool. you'll run into a small snag, btw [17:50:24] one of the mixins in mf-mixins.less uses inline JS [17:51:40] https://github.com/wikimedia/mediawiki-extensions-MobileFrontend/blob/master/less/mixins.less#L58 [17:51:53] IIRC jon was just going to get rid of it [17:52:08] heh [17:52:52] lesscss.org: "JavaScript expressions can be evaluated as values inside .less files. We recommend using caution with this feature as the LESS will not be compilable by ports and it makes the LESS harder to maintain. If possible, try to think of a function that can be added to achieve the same purpose and ask for it on github." [17:54:55] it's the only such instance though [17:55:42] oh, even better [17:55:44] it's redundant [17:55:50] it was copy-pasted from http://www.toekneestuck.com/blog/2012/05/15/less-css-arguments-variable/ [17:55:58] "As commenter Matthew points out, as of the 1.3.2 release, this solution is no longer needed. LESS now has built-in support for comma-separated values, via semicolon (;) delimiters. " [17:56:08] yeah it looks unnecessary [17:56:39] (03PS1) 10Umherirrender: update.php clears caches twice [core] - 10https://gerrit.wikimedia.org/r/84137 [17:57:05] hmm except that transform->-webkit-transform. lemme double-check if that can be replaced with something else [17:59:03] it should just have its own mixin, IMO [17:59:33] like https://github.com/twbs/bootstrap/blob/master/less/mixins.less#L150 [17:59:52] that seems much nicer [18:01:15] errr, that should have been L143 [18:01:17] ori-l: yeah and that's used in a couple separate places that don't need transitions [18:01:45] problem is that the -webkit-transition lines also need transformation of transform to -webkit-transform in their contents [18:02:15] hmm does .transition (.transform @duration, opacity @duration ); work ? [18:02:16] * ori-l squints at that sentence a little [18:02:32] i don't think i had enough coffee, give me a sec :P [18:02:38] heh [18:04:12] SyntaxError: expected ')' got '.' in less/modules/watchstar.less:12:15 [18:04:13] nope [18:04:57] i think that's probably sane [18:05:39] not supporting nested mixin invocation [18:05:42] could ask to add a string replace function i guess :P [18:06:42] http://leafo.net/lessphp/docs/#custom_functions [18:07:26] excellent :D [18:20:22] (03CR) 10Brion VIBBER: [C: 04-1] "Potential security issue: filesystem paths are exposed if there's an error in less processing:" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [18:20:45] actually… that's probably not a problem again [18:22:10] (03CR) 10Brion VIBBER: [C: 031] "eh that's probably not a problem, don't use broken files" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [18:27:46] brion: this seems to work: https://dpaste.de/6337N/raw/ [18:28:11] ori-l: right now i'm just trying to get basic styles working [18:28:13] no dice yet [18:28:20] not sure what's wrong [18:28:51] you modified the resource loader module definitions to refer to the less files rather than css files, right? [18:29:03] yup [18:29:26] hmm [18:30:41] i'm not sure they're all backed by less files [18:30:41] sighhhhh [18:30:58] what do you mean? [18:31:10] there are not less files for every .css file [18:33:44] brion: just references and filepage [18:33:47] it looks like [18:33:54] yeah [18:33:56] (03PS1) 10Umherirrender: Use always content language for localized ContentModel [core] - 10https://gerrit.wikimedia.org/r/84138 [18:33:57] that shouldn't break it…. [18:34:46] lemme try copying em over [18:35:12] p stylesheets/modules/references.css less/modules/references.less ; mkdir less/file ; cp stylesheets/file/filepage.css less/file/filepage.less [18:35:26] that should start with 'cp', not 'p' [18:35:44] ok as expected that doesn't make a difference :( [18:36:06] lemme push a temp pull req [18:36:14] blargh, i'm sorry. are you getting an error of some sort? [18:36:23] none that i can see [18:36:27] just lots of missing styles [18:39:08] * ori-l tries [18:39:12] hmm maybe it's failing to fetch these images [18:39:16] lemme move em [18:41:10] \o/ [18:41:13] ok that's working much better [18:41:25] weeee, awesome [18:42:50] ok lemme do a little more cleanup and upload a patchset [18:45:35] we should have an
in the DOM in debug mode [18:45:55] then in CSSmin / LESS errors, instead of making the exception message a comment, [18:46:17] Warning: array_keys() expects parameter 1 to be array, null given in /Library/WebServer/Documents/core/includes/resourceloader/ResourceLoaderFileModule.php on line 731
[18:46:18]
[18:46:18] #mw-debug: before { content: "error in line ..., etc."; } [18:46:18] Fatal error: Unsupported operand types in /Library/WebServer/Documents/core/includes/resourceloader/ResourceLoaderFileModule.php on line 731
[18:46:36] something seems awry there lemme try to track it down [18:48:44] that's very odd [18:49:16] looks like cachedCompile() is returning null [18:50:08] what are you passing in? [18:50:58] something dying in mobile.styles, don't know what yet [18:51:08] it was working not long ago…. [18:51:11] i don't understand [18:52:07] the value passed to cachedCompile has to be a string or an array with (at least) a 'root' key [18:52:30] ResourceLoaderFileModule should probably check for 'root' key in line 711: if ( !is_array( $source ) ) { [18:52:33] ori-l: here's my work in progress https://gerrit.wikimedia.org/r/84139 [18:52:34] but i'm not sure why it wouldn't be set [18:52:47] trying it [18:53:15] oh, wait [18:53:32] no, nevermind [18:54:10] ori-l: https://gist.github.com/brion/6554604 <-$source [18:54:44] so i've got one with root null [18:54:47] seems odd? [18:55:24] and why does it have compiled info on output? eh confusing [18:55:25] ah there's a less error [18:58:08] the styles look broken but it's compiling for me [18:58:30] * brion grumbles [18:58:33] seems like some of the variables are undefined? [18:58:35] body{font-family:;font-size:;line-height:/ 3 * 4;color:#252525 [18:59:36] that's defined in less/variables.less [19:01:08] oh you know what, i bet i mixed up versions of files [19:01:10] lemme adjust this [19:02:00] (03CR) 10Parent5446: [C: 031] "Haven't tested yet. Will test and merge if I have time and somebody doesn't beat me to it." [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [19:03:29] ugh i don't know what's going on, will poke it after lunch [19:04:01] (03CR) 10Ori.livneh: "@Parent5446: it's an open RFC, so Brion, Tim or Mark have to OK it first." [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [19:04:29] brion: thanks. I think there's at least one obvious improvement to error checking i can make in resourceloaderfilemodule [19:04:40] i'll make that now, bbiab too [19:04:54] thanks for your time and effort! [19:05:22] :D [19:05:25] yay [19:15:51] (03PS1) 10saper: Improve markdown and workaround bug 54050 [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/84142 [19:15:54] (03PS1) 10Umherirrender: Always pass user to LogPage::addEntry [core] - 10https://gerrit.wikimedia.org/r/84143 [19:54:10] (03PS8) 10Ori.livneh: Support LESS stylesheets in ResourceLoader [core] - 10https://gerrit.wikimedia.org/r/78669 [19:57:49] (03CR) 10Ori.livneh: "PS8:" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [19:58:05] ^ brion [19:59:32] hey guys is the developer of the questy captcha around? [19:59:37] emufarmers? [19:59:54] according to the media wiki extension page for it [19:59:57] not at the moment; he leaves the channel when he's away [20:00:02] but he's often here [20:00:18] his nickname is 'Emufarmers', same as you'd expect. [20:00:50] what's the question? [20:01:21] ori-l: yep :) i woudl like to discuss somethings and possibly get some mentoring on getting questy to detect a users browser language that way then questy can work with localized questions depending on the users browser [20:01:38] hm [20:01:53] so it would work with the UniversalLanguageSelector? [20:02:09] Nemo_bis: i havent delved into its inner workings [20:02:13] in terms of questy [20:02:19] hence why i have come seeking its creator [20:03:07] Nemo_bis: how would you recommend i go about doing it? cuz the document foundation use it as it seems to be better then alot of the others even recaptcha [20:03:08] as long as you leave the language detection to ULS, it may not be hard [20:03:31] yes, questycaptcha is vastly superiorior for normal wikis if you keep an eye on it [20:03:50] ya TDF did some statistical comparisons its a million times better [20:04:12] the reason im tryign to contact the plugin creator is so i dont end up duplicating the same work for nothign if its somethign he has in the pipeline [20:04:25] sure [20:04:36] you can also ask on his talk page or file a bug :) [20:04:46] Nemo_bis: i was going to email the mailing list [20:05:47] anyway, I guess it's not so hard, you'd only have to change the format of the configuration and add a language parameter to the questions; then ULS, if enabled for anons, would tell MediaWiki the language and you'd serve a question chosen based on that [20:06:09] humm that is much easier then the way i was planning it [20:06:18] having it detect the browser language [20:06:20] the problem is that configuration files don't scale, so this will/would work better if the question setting was on a special page [20:06:41] And *this* is what Emufarmers plans to work on. Sometime. [20:07:12] well that is why i want to make contact with him maybe i can help him out [20:07:14] was he in here? [20:08:49] he often is [20:09:48] Hm, maybe not so often lately, and surely more on #mediawiki [20:10:59] im there but havent seen him :( [20:14:52] !cms [20:14:53] Wikis are designed for openness, to be readable and editable by all. If you want a forum, a blog, a web authoring toolkit or corporate content management system, perhaps don't use wiki software. There is a nice overview of free tools available at including the possibility to try each system. For ways to restrict access in MediaWiki, see !access. [20:19:51] eagles051387: he's online but doesn't lurk in the channels. if you pm him, he should respond [20:20:55] thanks legoktm :) [20:27:41] (03CR) 10Physikerwelt: "...in theorie code review sounds like a good idea... in practice it is sometimes a little bit annoying..." [extensions/Math] - 10https://gerrit.wikimedia.org/r/75828 (owner: 10Physikerwelt) [20:30:01] grrrit-wm: its a cool concept forsure. the document foundation use it and it helps before they implemented gerrit they used to have everyone send in patches via email and sometimes patches slipped through the cracks [20:30:41] er, you're talking to a bot [20:30:57] doh [20:30:58] lol [20:31:08] <--noob [21:36:15] (03PS1) 10Jackmcbarn: Fix SQLite being incompatible with AbuseFilter [extensions/AbuseFilter] - 10https://gerrit.wikimedia.org/r/84185 [21:48:43] (03PS3) 10Ori.livneh: Provide a dynamic global scope of LESS variables [core] - 10https://gerrit.wikimedia.org/r/84101 [21:49:02] legoktm: got a hold of emu :) we are going to touch base over the weekend :) [21:50:57] :D [21:51:21] (03PS9) 10Ori.livneh: Support LESS stylesheets in ResourceLoader [core] - 10https://gerrit.wikimedia.org/r/78669 [21:54:23] (03PS4) 10Ori.livneh: Provide a dynamic global scope of LESS variables [core] - 10https://gerrit.wikimedia.org/r/84101 [21:56:55] (03CR) 10Ori.livneh: "PS9: Don't reuse lessc instance across compilations, because stray bits of the parser environment are preserved and can mess up subsequent" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [21:57:21] brion: ^ your mobilefrontend patch works with this [21:57:38] oho [21:57:54] awesome! [21:58:30] i don't know why i thought it'd be safe to reuse the compiler instance, I was probably just aping ResourceLoaderModule::jsParser without thinking about it [21:59:37] one of the side-effects of compilation is that the compiler instance's protected 'allParsedFiles' property is an array populated with all the files that it already encountered [21:59:58] which it uses as a heuristic to avoid recompiling files that are referenced multiple times [22:00:40] like variables.less :) [22:03:23] hmmmmm [22:03:51] it's almost working [22:05:47] what isn't? [22:06:11] images that should be embedded aren't, and some of the styles seem wrong or broken [22:09:52] sonofa [22:09:58] reloaded and the broken ones are gone [22:10:01] i'm not sure what was up with that [22:10:04] lemme double-check the embedding [22:10:41] oh no wait, that's cause i went back to master… of course it works ;) [22:13:23] MobileFrontend is preventing the embedding somehow [22:13:31] it happens with current master of core + MFE [22:13:38] and it's like that in prod too: https://bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&lang=en-gb&modules=mobile.pagelist.styles%7Cmobile.styles%7Cmobile.styles.page%7Cmobile.zero.styles&only=styles&skin=mobile&target=mobile&* [22:14:30] i see a bunch of data uris in there [22:14:46] not every image but a bunch of em [22:14:46] oh, hrm [22:14:50] like the menu icons [22:16:55] i can only assume something was broken in the caching re: the other breakages…. it seems happy now except for the embedding [22:17:00] * brion stabs caching [22:18:41] Is this correct image syntax? [[File:Foobar.svg|thumb|width=200|lang=de]] [22:18:44] passing width in that way? [22:19:19] |200px| ? [22:23:46] (03CR) 10Brion VIBBER: "Yay, it's mostly working!" [core] - 10https://gerrit.wikimedia.org/r/78669 (owner: 10Ori.livneh) [22:24:11] arghhh what is going on with @embeds [22:24:33] i think i know [22:44:29] oh man, this is going to be awful