[00:01:23] TimStarling: ping [00:01:35] (03Merged) 10jenkins-bot: Update OOjs UI to v0.1.0-pre (eaa1b7f06d) [core] - 10https://gerrit.wikimedia.org/r/123807 (owner: 10Jforrester) [00:01:59] Error: 503, Service Unavailable at Fri, 04 Apr 2014 00:00:19 GMT [00:02:06] yes? [00:02:30] (03CR) 10SG: "Fixed build issues. Notes:" [core] - 10https://gerrit.wikimedia.org/r/123804 (owner: 10SG) [00:03:20] (03CR) 10Krinkle: mediawiki.api: Refactor getToken and postWithToken methods (031 comment) [core] - 10https://gerrit.wikimedia.org/r/111636 (owner: 10Krinkle) [00:03:32] TimStarling, pm [00:04:20] (03CR) 10Krinkle: Replace deprecated/removed $wgExtensionAliasesFiles (031 comment) [extensions/UpdateMediaWiki] - 10https://gerrit.wikimedia.org/r/123550 (owner: 10Siebrand) [00:04:24] (03CR) 10jenkins-bot: [V: 04-1] mw-ui-button: Revamp the LESS to be simpler, and improve cross-browser support [core] - 10https://gerrit.wikimedia.org/r/123804 (owner: 10SG) [00:04:49] (03PS7) 10Jforrester: Split /resources into /resources/lib and /resources/src [core] - 10https://gerrit.wikimedia.org/r/123782 [00:04:52] TimStarling, so I'm not sure how far your discussion on this is / what your plans are [00:04:59] well, we already have a service boundary for text storage [00:05:02] (03CR) 10Jforrester: "PS7 is a rebase." [core] - 10https://gerrit.wikimedia.org/r/123782 (owner: 10Jforrester) [00:05:08] i.e. ExternalStore [00:05:35] what I am imagining is a service boundary for the things that are currently in the revision table [00:05:51] yeah, that's what rashomon partly provides [00:05:54] and possibly also recentchanges and its linked tables [00:06:31] I think it makes sense to separate out RC [00:07:08] as in could be external, but should be a different entry point [00:07:13] you know that Special:Contributions is provided by the revision table [00:07:49] that's a backend issue [00:08:00] in an interface you'd want to separate the concerns [00:08:32] so what do you imagine rashomon doing, and how do you imagine contributions would be provided? [00:09:04] for now Rashomon stores revisions, lets you list and retrieve them by timestamp etc [00:09:24] (03CR) 10Krinkle: Deprecate mw.util.$content (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [00:09:29] the data model is not built to provide something like contributions [00:09:48] in a distributed storage world you'd probably optimize something like contributions separately [00:09:59] and provide a different entry point for it [00:10:39] there is support for eventually consistent 'atomic' batch operations that can be used to update several (potentially denormalized) tables at once [00:10:43] in Cassandra [00:10:54] other backends likely have similar mechanisms [00:11:13] all without locking [00:12:07] we'd have to look at the details, but I think that's one way we could add efficient contribution tracking in Rashomon [00:12:40] and when you say that it stores revisions, you mean that it stores text? [00:12:50] text and metadata? [00:12:51] text, json, html etc [00:13:08] each revision is a key-value object [00:13:42] revision ids are an issue in a distributed store [00:13:53] just wondering about locality [00:14:05] locality is by page and type [00:14:22] so all html for main page ends up concatenated on disk, by timestamp [00:14:24] yeah, but if you want to display a history page then you only need the metadata, not the text [00:14:37] similar all metadata etc [00:14:42] so if you have the text stored with the metadata, then you end up with poor cache efficiency [00:14:49] you could have a json blob called 'history' [00:14:50] and presumably inefficient I/O as well [00:14:56] reading that is very efficient [00:15:15] no, each type is stored concatenated [00:15:36] where type means what? [00:15:36] so reading all 'history' json entries is reading a contiguous chunk (also compressed) on disk [00:15:51] html, wikitext, 'history' (json), etc [00:17:06] we should create a list of access patterns and see if those are efficiently covered [00:17:30] re ids, in cassandra we are using timestamp uuids [00:17:50] we also maintain a mapping from mediawiki revision id to timestampuuid [00:18:37] and support queries like 'give me the first html render for this mediawiki revision id' [00:20:08] https://github.com/gwicke/rashomon/blob/master/cassandra-revisions.cql#L7 [00:20:20] that's the current schema [00:20:32] (03PS2) 10Ori.livneh: Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 [00:21:03] it needs a clean-up before deploy, hope to get to that after my vacation [00:21:38] the interesting thing is the primary key at https://github.com/gwicke/rashomon/blob/master/cassandra-revisions.cql#L15 [00:22:12] the first component hashes the page name, and uses that to map storage to a specific node [00:22:27] (03PS3) 10Ori.livneh: Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 [00:22:50] remaining components (prop and timestamp uuid) determine the hierarchical clustering / ordering [00:23:20] so all entries with prop = 'html' end up together on disk, ordered by timestamp [00:25:17] note that there are no destructive renames here [00:25:36] I'd like to hear springle's view on the idea of using cassandra for revision storage [00:25:39] a rename is just a new revision at both the old & new location [00:25:43] it's certainly not the only feasible solution [00:26:14] I looked at Riak too, but decided to go with Cassandra in the end [00:26:49] that's pretty much it in the space I'm interested in [00:26:53] IIRC springle was excited about [00:26:58] (03PS17) 10Bsitu: Closing & Summarizing topic frontend changes [extensions/Flow] - 10https://gerrit.wikimedia.org/r/119640 [00:27:23] so do you consider this to definitely be the best option for revision storage in MediaWiki? [00:27:39] or is it still experimental? [00:28:06] (03CR) 10jenkins-bot: [V: 04-1] Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 (owner: 10Ori.livneh) [00:28:08] it is not proven in production yet obviously, so not calling it experimental would be wrong [00:28:18] I did a lot of load testing though [00:28:55] I think that something like Cassandra is currently the best backend option for this application [00:29:21] the goal of Rashomon is to abstract the backend though, so if something better comes along we can take advantage of that [00:29:45] also Rashomon can use several backends [00:29:54] abstraction as a service? [00:30:02] yes, with a REST API [00:30:05] fair enough, I guess [00:30:05] (03PS4) 10Ori.livneh: Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 [00:30:18] can it use the existing revision table in MySQL? [00:31:08] probably yes [00:31:39] but you would have to add a method for HTML storage [00:32:10] since the revision table is not that generic [00:32:11] well, the difficulty of doing that is what got us to start Rashomon [00:32:26] the text table is certainly that generic, it is used for storing all sorts of structured data unrelated to revisions [00:32:40] http://phprestsql.sourceforge.net/ [00:32:51] yeah, but maintaining the mapping of revision to wikitext, html, json, ... is hard [00:33:52] TimStarling, I think if we wanted to provide a light-weight alternative or mock, then designing a different schema would be better [00:35:13] or at least extending the revision table with the ability to maintain a list of any number of associated objects and the ability to look those up by name & timestamp [00:36:09] so re the PHP side, Aaron has been working on the virtual REST interface bindings [00:36:13] timestamps are not unique [00:36:35] when constructing a timestampuuid you can seed the uuid [00:36:51] a UUID is unique, by definition [00:36:53] normally entropy is used, but you can also use the page title or something else [00:37:00] maybe you are just abbreviating? [00:37:22] oh, you'd want a fully unique timeuuid [00:37:31] but with the ability to look up by timestamp [00:37:43] so slightly changed byte order [00:37:53] cassandra does that automatically [00:38:37] so re PHP interface, https://www.mediawiki.org/wiki/Requests_for_comment/PHP_Virtual_REST_Service describes that mostly [00:39:05] Aaron's WIP is at https://gerrit.wikimedia.org/r/#/c/110129/ [00:40:01] parsoid HTML storage is not especially critical -- if it goes away, we can rebuild it [00:40:25] yeah, so a good first use case [00:40:28] but moving the primary revision storage for MediaWiki to cassandra is something different [00:40:40] definitely [00:40:50] we'd probably want to run in parallel for a while [00:41:32] (03PS1) 10Werdna: Fix a few bugs where Flow was not working on pristine pages. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123812 [00:41:34] I have been talking about starting a project to shard the revision table [00:41:57] as in manual sharding? [00:42:13] the options are open [00:42:23] presumably you would prefer to see it use rashomon? [00:42:29] how pressing is the problem? [00:43:08] I don't think that manual sharding is the best long-term solution, but if something is needed that will definitely work next week then it might be a good short-term solution [00:48:11] it is not a pressing performance problem, maybe it will be in a year [00:48:11] the urgency comes from a desire to make MW look nicer and be more abstract and service oriented, so that it doesn't get left behind [00:48:11] but maybe that is not convincing to everyone as a project proposal, so it is nice to have performance improvements tied to that [00:48:12] one thing that rashomon can speed up is access to content [00:48:12] basically load the page content directly from Rashomon, and insert it with ESI [00:48:12] for both anons and logged-in users [00:48:12] well, we could write to both backends, rashomon and mysql, on page save [00:48:13] re timing, we should pick up speed towards the summer with the service team [00:48:13] like what we did during the migration to swift [00:48:37] yup, and maybe also send read requests to it without using the data at first [00:49:11] we'll use it for Parsoid pretty soon [00:49:42] have been saying that for a while, I know; I blame HTML templating for getting in the way [00:49:55] swift is an ominous precedent, you know [00:50:21] yeah, I know [00:50:23] we should hope that it will be easier than swift/ceph for the ops team [00:51:07] I have been able to break early cassandra 2.0 versions [00:51:16] with really heavy write loads [00:51:39] (03CR) 10MaxSem: [C: 032] Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 (owner: 10Ori.livneh) [00:52:07] as in more than 1000 revisions / second [00:52:39] stabilized around 2.0.5 [00:54:20] the cassandra policy is to release .0 early, and stabilize in the first dot releases; generally production is not recommended before .6 [00:54:52] 2.6 or 2.0.6? [00:54:55] (03Merged) 10jenkins-bot: Callers of ResourceLoader::getModule should check for null return [core] - 10https://gerrit.wikimedia.org/r/123802 (owner: 10Ori.livneh) [00:55:04] x.x.6 [00:55:20] so in the current cycle 2.0.6, which is now out [00:55:37] (03PS1) 10Jforrester: Update OOjs UI to v0.1.0-pre (eaa1b7f06d) [core] (wmf/1.23wmf21) - 10https://gerrit.wikimedia.org/r/123814 [01:02:05] (03CR) 10Krinkle: [C: 032] Update OOjs UI to v0.1.0-pre (eaa1b7f06d) [core] (wmf/1.23wmf21) - 10https://gerrit.wikimedia.org/r/123814 (owner: 10Jforrester) [01:02:30] (03PS1) 10Jdlrobson: Port thumbnail styling from VectorBeta into Vector [core] - 10https://gerrit.wikimedia.org/r/123815 [01:02:32] (03PS1) 10Jdlrobson: Component common css [core] - 10https://gerrit.wikimedia.org/r/123816 [01:02:36] (03PS1) 10Jdlrobson: Vector: Simply style external links [core] - 10https://gerrit.wikimedia.org/r/123817 [01:07:12] (03PS1) 10Robmoen: Create base UI module. [extensions/UploadWizard] (oojsify) - 10https://gerrit.wikimedia.org/r/123818 [01:07:23] (03Merged) 10jenkins-bot: Update OOjs UI to v0.1.0-pre (eaa1b7f06d) [core] (wmf/1.23wmf21) - 10https://gerrit.wikimedia.org/r/123814 (owner: 10Jforrester) [01:08:11] (03CR) 10Robmoen: [C: 04-1] "WIP" [extensions/UploadWizard] (oojsify) - 10https://gerrit.wikimedia.org/r/123818 (owner: 10Robmoen) [01:13:05] (03CR) 10jenkins-bot: [V: 04-1] Component common css [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [01:32:09] (03PS4) 10Werdna: Implement multiPut method. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/120734 [01:32:12] (03PS28) 10Werdna: Extract wiki and external links, file and template usages from text. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/110090 [01:32:46] (03CR) 10jenkins-bot: [V: 04-1] Extract wiki and external links, file and template usages from text. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/110090 (owner: 10Werdna) [01:32:52] * werdna glares at jenkins-bot [01:34:25] (03CR) 10Werdna: "PS4 addresses all feedback, except the feedback about superfluous count() checks, I don't see how they're superfluous – where else is it c" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/120734 (owner: 10Werdna) [01:38:20] (03PS1) 10Krinkle: mw.hook: Make hook.fire actually chainable [core] - 10https://gerrit.wikimedia.org/r/123820 [01:38:22] (03PS1) 10Krinkle: mw.hook: Implement hook.once [core] - 10https://gerrit.wikimedia.org/r/123821 [01:40:35] (03CR) 10Werdna: "Tests all pass on my local box, and I can't figure out why the HTML is being output?" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/110090 (owner: 10Werdna) [01:41:48] (03CR) 10Krinkle: "@DanielFriesen: Beware though that mw.util.$content cannot be globally accessed either, it is null until mw.util.init runs, which has no c" [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [01:43:02] (03PS3) 10Krinkle: Deprecate mw.util.$content [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [01:44:11] (03CR) 10Werdna: "PS28 is a rebase, btw." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/110090 (owner: 10Werdna) [01:50:57] (03PS1) 10Krinkle: resourceloader: Don't add superfluous line breaks and semicolons [core] - 10https://gerrit.wikimedia.org/r/123822 [01:57:31] (03PS8) 10Ori.livneh: resourceloader: Refactor module links output [core] - 10https://gerrit.wikimedia.org/r/95463 (owner: 10Krinkle) [01:57:33] (03CR) 10jenkins-bot: [V: 04-1] resourceloader: Refactor module links output [core] - 10https://gerrit.wikimedia.org/r/95463 (owner: 10Krinkle) [01:57:39] (03CR) 10Ori.livneh: [C: 032] resourceloader: Refactor module links output [core] - 10https://gerrit.wikimedia.org/r/95463 (owner: 10Krinkle) [01:58:35] (03PS4) 10Werdna: Create test for bug 59637 [extensions/Flow] - 10https://gerrit.wikimedia.org/r/108879 [01:58:42] ori: I'll rease [01:58:43] rebase [01:58:54] (03CR) 10Werdna: "PS4: Fix indenting issue :)" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/108879 (owner: 10Werdna) [01:59:18] (03Abandoned) 10Werdna: Add basic experimental hotkey navigation. [extensions/Flow] - 10https://gerrit.wikimedia.org/r/91348 (owner: 10Werdna) [02:00:58] (03Merged) 10jenkins-bot: resourceloader: Refactor module links output [core] - 10https://gerrit.wikimedia.org/r/95463 (owner: 10Krinkle) [02:01:11] Awesome [02:01:41] Krinkle: what happened there? [02:01:53] not sure [02:02:04] Might be a race condition with gerrit slave stuff [02:02:08] replag [02:03:09] the oddness starts with the fact that the 'unable to rebase' message was given in response to a successful rebase [02:04:06] Hi, I need a little help with Javascript in LiquidThreads extension [02:04:56] zeek: just ask [02:05:21] (03CR) 10Werdna: [C: 032] Serialize UUID into something more compact [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123689 (owner: 10Matthias Mullie) [02:06:37] (03PS2) 10Werdna: Warn when __toString is needed on UUID [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123692 (owner: 10Matthias Mullie) [02:06:53] (03CR) 10jenkins-bot: [V: 04-1] Warn when __toString is needed on UUID [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123692 (owner: 10Matthias Mullie) [02:07:28] I want to pass a new parametre with every form submit. I was able to generate the parametre as a hidden field, but when I pass it in the function, and check it in APIThreadAction.php, it shows empty. Currently I am concerened with Reply, so I added a parametre manually to doReply function in lqt.js , and checking it in the actionReply in APIThreadAction. But it shows empty. I am pretty sure that these are the places to do it but [02:09:49] (03CR) 10Jforrester: [C: 04-1] resourceloader: Don't add superfluous line breaks and semicolons (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123822 (owner: 10Krinkle) [02:10:37] ori: Krinkle: http://www.fpaste.org/91549/57740013/ If you can see this please.. [02:12:17] zeek: When you look in the Network tab if your browser dev tools, does it actually get submitted? [02:12:25] (be it malformed maybe) [02:12:30] or not sent at all? [02:13:04] debug the http request in Chrome (or other browser if it can and know how) and look at the post payload [02:13:24] I am not sure, but things do work if I do not check for that empty(params['formToken']) [02:13:36] Ok let me see.. I use Firefox though [02:13:36] alternatively, put a debugger; statement in the function's first line and inspect the parameters and verify they are actually containing the expected values [02:13:41] Don't use empty() [02:14:10] zeek: Is the parameter registered in the API module? [02:14:17] you can't pass arbitrary query parameters. [02:14:35] I dont think it is, I just added it in my code [02:14:46] How can I do that? [02:15:02] empty() does too many things, for one, it also does isset() which means if you mistype the any part of the variable name it silences an important error [02:15:30] zeek: Search for occurences for some of the other params[] keys in the file and you'll find a method returning an array of query parameter names and their signature [02:15:36] built-in validation [02:15:58] I used empty() because the similar was used to check other params[] [02:16:02] (03CR) 10Daniel Friesen: [C: 04-1] "- This is going to break skins using common/commonContent.css, we still don't have a stable release that actually has the modules, much le" (032 comments) [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [02:16:11] empty() also tolerates way too many values (just like php boolean cast). Be more explicit (for example !== "") [02:16:53] All right will do so, still the current problem is that the value of params['formToken'] is not available in the API [02:16:57] empty() is sufficiently crappy that I'd say even "consistency" in old code is not a rationale. It's a small function, not a big deal or pattern, just avoid it in new code. [02:17:25] right, so add it in that parameter registry of that api module file [02:17:42] Let me see.. [02:17:51] https://github.com/wikimedia/mediawiki-extensions-LiquidThreads/blob/master/api/ApiThreadAction.php#L911 [02:18:08] Looks like it's lacking information for most values, but at least it needs to be listed. [02:18:12] Ugh [02:22:34] (03CR) 10Daniel Friesen: "@Krinkle, the issue I was thinking of was more about things loaded after all those modules are already loaded. For example in a module loa" [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [02:23:25] Krinkle|detached: Added, still fails. :( [02:26:36] Krinkle|detached: Works... needed a ctrl+F5 :) Thanks a lott.... [02:53:38] (03PS14) 10Wctaiwan: Change UI of special page and support Flow Thanks [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/119879 [03:00:22] hmm, who's about? [03:00:50] siebrand: poke poke? [03:02:27] (03PS7) 10Wctaiwan: Implement a redirector at Special:Flow [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121306 [03:04:09] (03Abandoned) 10Bencmq: Add mw-ui-light button style [core] - 10https://gerrit.wikimedia.org/r/121101 (owner: 10Bencmq) [03:05:22] (03PS2) 10Bencmq: Clone thanked label before inserting [extensions/Thanks] - 10https://gerrit.wikimedia.org/r/122116 [03:19:57] TimStarling: cassandra for revision text, whether as ES backend or rashomon, +1. moving revision metadata to cassandra too, lock, stock, and barrel... needs thought [03:21:01] I don't think rashomon is really meant to be an ES backend [03:21:24] since ES is a dumb key/value store and rashomon is smarter [03:21:32] I suspect a hybrid approach would be needed. mariadb 10 cassandra engine could allow a single interface for writes, pushed out elsewhere for reads [03:21:36] right [03:21:52] cassandra itself could presumably be an ES backend, without much effort [03:23:34] what are your concerns about moving metadata to cassandra? [03:23:35] regarding revision table, we're using partitioning on some slaves now to speed up some queries. the size of the table isn't such an issue that way [03:31:54] maintaining consistency between two storage backends for both transactions and backups. identifying the access methods we currently use and deciding whether a single cassandra schema could servive them (including both OLTP and OLAP). unknowns. [03:32:29] none of them reasons not to try, of course [03:32:45] ok [03:33:21] but if it is easy to push information out to cassandra without extensive modifications to mediawiki, we surely have a cheap way to trial a hybrid approach, and a safe fallback position [03:49:47] (03PS1) 10Yurik: Set proper SSH URL for all GIT repos [vagrant] - 10https://gerrit.wikimedia.org/r/123830 [04:09:57] TimStarling: what other options are you considering in your project to shard revision? [04:10:06] * springle reads backlog, curious [04:10:50] nothing really seriously considered [04:11:17] but the options are basically 1) leave it how it is, 2) use multiple MySQL servers, 3) use some other thing [04:11:54] 4) trial another mariadb storage engine, like tokudb [04:12:29] if you think the performance is fine for another year or two, then that will lend weight to option 1, I guess [04:15:42] yeah, revision isn't really an insurmountable pain point for performance. it's very slow to do maintenance like schema changes, but proper online ALTER TABLE is already in mysql 5.6 and will flow through to mariadb [04:16:11] the true pain is in special page queries, mainly anything under IndexPager: they could all use a rethink [04:20:11] tokudb engine is almost as exciting as cassandra. revision, logging, the *links tables, anything with more than a couple indexes, could all be candidates for tokudb [04:29:14] what's wrong with IndexPager? [04:36:39] (03PS8) 10Werdna: Implement a redirector at Special:Flow [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121306 (owner: 10Wctaiwan) [04:43:03] (03CR) 10Werdna: "Should probably rebase on top of change 121855 and use the functions that it does." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121306 (owner: 10Wctaiwan) [04:44:59] (03CR) 10Werdna: [C: 032] API: Set 'missing': '', instead of 'missing': 'missing' [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123178 (owner: 10Legoktm) [04:47:07] (03CR) 10Werdna: [C: 031] "OK but giving Lego a chance to address Bernie's feedback." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123174 (owner: 10Legoktm) [04:50:30] (03PS7) 1001tonythomas: Added username as parameter to MediaWiki:Comments-reply-to [extensions/Comments] - 10https://gerrit.wikimedia.org/r/118957 [04:51:29] (03CR) 10Werdna: [C: 031] "Pretty much good." (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123114 (owner: 10EBernhardson) [04:52:17] (03CR) 10Werdna: [C: 032] Standards compliant reset of min/max-width [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123138 (owner: 10EBernhardson) [04:52:31] (03Merged) 10jenkins-bot: Standards compliant reset of min/max-width [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123138 (owner: 10EBernhardson) [04:57:42] (03CR) 10Werdna: [C: 032] Remove unused class variable $dbFactory [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123470 (owner: 10Bsitu) [04:57:55] (03Merged) 10jenkins-bot: Remove unused class variable $dbFactory [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123470 (owner: 10Bsitu) [05:09:30] (03PS2) 10Umherirrender: Allow access to revision deleted wikitext on action=edit [core] - 10https://gerrit.wikimedia.org/r/120858 [05:15:13] (03PS8) 1001tonythomas: Added username as parameter to MediaWiki:Comments-reply-to [extensions/Comments] - 10https://gerrit.wikimedia.org/r/118957 [06:03:42] (03CR) 10Fomafix: "$.client.profile() do not need domready. There is no reason to initialize tooltipAccessKeyPrefix with a static value and overwrite it on d" [core] - 10https://gerrit.wikimedia.org/r/123780 (owner: 10Gerrit Patch Uploader) [06:41:19] (03PS1) 10Werdna: Add a hover state for topic viewing filter buttons [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123832 [06:48:23] (03CR) 10Nemo bis: [C: 04-1] "Thanks for submitting. Lacks rationale in commit message and bug report for user story and discussion." [core] - 10https://gerrit.wikimedia.org/r/123817 (owner: 10Jdlrobson) [06:48:40] (03CR) 10Nemo bis: [C: 04-1] "Thanks for submitting. Lacks rationale in commit message and bug report for user story and discussion." [core] - 10https://gerrit.wikimedia.org/r/123815 (owner: 10Jdlrobson) [07:25:49] (03CR) 10PleaseStand: Vector: Simply style external links (032 comments) [core] - 10https://gerrit.wikimedia.org/r/123817 (owner: 10Jdlrobson) [07:26:27] (03PS12) 10Mattflaschen: WIP: Refactor and add non-linear tours [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/116228 [07:26:29] (03PS2) 10Mattflaschen: Update firsteditve [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/123171 [07:26:54] (03CR) 10jenkins-bot: [V: 04-1] WIP: Refactor and add non-linear tours [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/116228 (owner: 10Mattflaschen) [07:27:14] (03CR) 10jenkins-bot: [V: 04-1] Update firsteditve [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/123171 (owner: 10Mattflaschen) [07:27:39] (03CR) 10Mattflaschen: "Besides the rebase, this changes how hiding is handled somewhat." [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/123171 (owner: 10Mattflaschen) [07:31:46] (03CR) 10Siebrand: [C: 032 V: 032] Added username as parameter to MediaWiki:Comments-reply-to [extensions/Comments] - 10https://gerrit.wikimedia.org/r/118957 (owner: 1001tonythomas) [07:37:47] (03CR) 10Mattflaschen: "It works (try firsteditve, chained from here)" [extensions/GuidedTour] - 10https://gerrit.wikimedia.org/r/116228 (owner: 10Mattflaschen) [07:40:33] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121306 (owner: 10Wctaiwan) [07:42:57] (03CR) 10Siebrand: [C: 031] "L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/119640 (owner: 10Bsitu) [07:44:24] (03CR) 10Siebrand: Replace deprecated/removed $wgExtensionAliasesFiles (031 comment) [extensions/UpdateMediaWiki] - 10https://gerrit.wikimedia.org/r/123550 (owner: 10Siebrand) [07:46:14] (03CR) 10Siebrand: [C: 04-1] "This patch set should be rebased on the updated i18n structure." [extensions/SpamBlacklist] - 10https://gerrit.wikimedia.org/r/118636 (owner: 10Jackmcbarn) [07:46:43] (03CR) 10Siebrand: [C: 04-1] "This patch set should be rebased on the updated i18n structure." [extensions/InputBox] - 10https://gerrit.wikimedia.org/r/97559 (owner: 10Jackmcbarn) [07:54:31] (03CR) 10Siebrand: [C: 032] WikiEditor: Remove unmaintained template editor modules [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123647 (owner: 10TheDJ) [07:55:19] (03CR) 10Siebrand: [C: 032] WikiEditor: Remove unmaintained beta iframe mode [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123649 (owner: 10TheDJ) [07:55:58] (03CR) 10Siebrand: [C: 032] WikiEditor: Remove unmaintained highlight plugin [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123648 (owner: 10TheDJ) [07:56:20] (03CR) 10Siebrand: [C: 032] WikiEditor: Remove old unmaintained ToC module [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123645 (owner: 10TheDJ) [07:56:25] (03Merged) 10jenkins-bot: WikiEditor: Remove old unmaintained ToC module [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123645 (owner: 10TheDJ) [07:56:27] (03Merged) 10jenkins-bot: WikiEditor: Remove unmaintained template editor modules [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123647 (owner: 10TheDJ) [07:56:29] (03Merged) 10jenkins-bot: WikiEditor: Remove unmaintained highlight plugin [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123648 (owner: 10TheDJ) [07:56:31] (03Merged) 10jenkins-bot: WikiEditor: Remove unmaintained beta iframe mode [extensions/WikiEditor] - 10https://gerrit.wikimedia.org/r/123649 (owner: 10TheDJ) [07:56:35] (03PS1) 10Adi.iiita: LQT should not accept duplicate post [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 [08:01:42] (03PS2) 10Adi.iiita: LQT should not accept duplicate post [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 [08:03:12] Hate trailing spaces. How do you guys avoid accidently putting trailing spaces or tabs while writing code? :| [08:05:55] (03CR) 10Siebrand: [C: 04-1] "This patch set should be rebased on the updated i18n structure." [extensions/GlobalCssJs] - 10https://gerrit.wikimedia.org/r/120333 (owner: 10Legoktm) [08:07:04] (03CR) 10Siebrand: "@matmarex Not adding dots saves keyboard strikes? ;)" [core] - 10https://gerrit.wikimedia.org/r/123641 (owner: 10Gerrit Patch Uploader) [08:08:30] (03CR) 10Siebrand: [C: 032] Add a dot at the end of the deprecate messages [core] - 10https://gerrit.wikimedia.org/r/123641 (owner: 10Gerrit Patch Uploader) [08:10:02] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/MassMessage] - 10https://gerrit.wikimedia.org/r/120334 (owner: 10Wctaiwan) [08:11:46] (03Merged) 10jenkins-bot: Add a dot at the end of the deprecate messages [core] - 10https://gerrit.wikimedia.org/r/123641 (owner: 10Gerrit Patch Uploader) [08:13:25] zeek: we use an editor which removes or highlights them [08:13:57] Which? and can it be done in sublime... [08:14:11] no idea about sublime [08:14:56] Which one you use then? [08:16:53] zeek: kate [08:17:42] Nikerabbit: Ok, thanks.. will look into it. :) [08:18:42] zeek: it's part of KDE though [08:19:11] (03CR) 10Siebrand: [C: 031] "i18n/L10n reviewed." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/118399 (owner: 10Bsitu) [08:20:48] Nikerabbit: Found something interesting... :) https://github.com/SublimeText/TrailingSpaces [08:24:49] zeek: nice... [08:25:10] blegh, i need something to keep track of the different sublime modules i need... [08:30:16] (03CR) 10Siebrand: "Needs rebase." [extensions/MediaWikiChat] - 10https://gerrit.wikimedia.org/r/122878 (owner: 10GeorgeBarnick) [08:36:59] (03CR) 10Siebrand: [C: 04-1] "i18n/L10n reviewed." (032 comments) [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/84625 (owner: 10MarkTraceur) [08:37:04] thedj: something like a wiki page? :P [08:37:29] zeek: there is https://www.mediawiki.org/wiki/CC#Trailing_whitespace too [08:38:42] (03PS1) 10Konarak: MediaWiki:Watchnologintext: Add returnto parameter [core] - 10https://gerrit.wikimedia.org/r/123837 [08:39:02] thedj: use this to keep track of sublime modules https://sublime.wbond.net [08:40:10] (03Abandoned) 10Konarak: MediaWiki:Watchnologintext: Add returnto parameter [core] - 10https://gerrit.wikimedia.org/r/122089 (owner: 10Konarak) [08:44:57] If someone is not busy, please review my code on LiquidThreads https://gerrit.wikimedia.org/r/123836 . Thanks [08:46:46] (03CR) 10Siebrand: "Other than that this will generate some 1M+ lines of diff to translation files that are rarely ever edited manually, I have no objections." [core] - 10https://gerrit.wikimedia.org/r/123518 (owner: 10PleaseStand) [08:47:31] (03CR) 10Siebrand: "That is: ... will remain having *spaces* as indentation ..." [core] - 10https://gerrit.wikimedia.org/r/123518 (owner: 10PleaseStand) [09:14:24] what is happening with https://git.wikimedia.org/? [09:14:43] getting timeouts constantly [09:16:13] Tobi_WMDE: gitblit keeps breaking and then people restart the service, fixes it, then it happens again [09:16:20] that's all i know so far [09:16:51] mutante: thx. hope that will stop anytime soon.. [09:16:52] * mutante restarts it yet another time [09:17:33] Tobi_WMDE: try now [09:18:31] mutante: works again. it's just annoying when you're relying on something like e.g. https://git.wikimedia.org/zip/?r=mediawiki/core.git&format=gz&h=master [09:18:47] as our jenkins does.. [09:19:38] nods [09:25:36] gitblit faq says it's just designed for "small workgroups" to be fair. i think ^demon|away has looked before. so far all i have done is restart the service a few times. it used to happen before but not that often [09:26:17] and unsure how to properly debug the java app ...hrm [09:40:23] (03PS1) 10Raimond Spekking: Fix invalid JSON. Follow up to I9cfd71bf7fba134c2d1e22640600bcc6e209e06b [extensions/Comments] - 10https://gerrit.wikimedia.org/r/123839 [09:40:39] (03CR) 10Raimond Spekking: [C: 032 V: 032] Fix invalid JSON. Follow up to I9cfd71bf7fba134c2d1e22640600bcc6e209e06b [extensions/Comments] - 10https://gerrit.wikimedia.org/r/123839 (owner: 10Raimond Spekking) [09:50:44] (03CR) 10Raimond Spekking: Added username as parameter to MediaWiki:Comments-reply-to (031 comment) [extensions/Comments] - 10https://gerrit.wikimedia.org/r/118957 (owner: 1001tonythomas) [10:12:34] (03PS1) 10Siebrand: Migrate to JSON i18n [extensions/WikiObjectModel] - 10https://gerrit.wikimedia.org/r/123841 [10:34:47] (03PS1) 10Siebrand: Extension has moved to GitHub. Make that clearer [extensions/ThumbParser] - 10https://gerrit.wikimedia.org/r/123846 [10:36:31] (03PS2) 10Siebrand: Extension has moved to GitHub. Make that clearer [extensions/ThumbParser] - 10https://gerrit.wikimedia.org/r/123846 [10:41:59] (03PS1) 10Siebrand: Remove extension code because of pervasive security issues [extensions/Poll] - 10https://gerrit.wikimedia.org/r/123847 [10:46:47] (03PS1) 10Siebrand: Extension has moved to GitHub. Make that clearer [extensions/FacebookOpenGraph] - 10https://gerrit.wikimedia.org/r/123849 [10:56:23] (03PS2) 10Matthias Mullie: Remove couple of methods by using built-in sort [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123688 [11:00:45] (03PS1) 10Siebrand: Migrate to JSON i18n [extensions/SocialProfile] - 10https://gerrit.wikimedia.org/r/123850 [11:01:10] (03CR) 10MaxSem: [C: 04-1] "Had an attempt been made to verify that the extension is still insecure? From what I see, the security issues were fixed in I7249b80e9c246" [extensions/Poll] - 10https://gerrit.wikimedia.org/r/123847 (owner: 10Siebrand) [11:01:39] sigh [11:01:52] happy now NickServ? OK. [11:02:40] Anyway - hi. What's the etiquette for asking about the status of a bug? I'd like to nudge something forward but don't want to overstep any mark or be annoying. [11:03:13] (03CR) 10Raimond Spekking: [C: 032 V: 032] Migrate to JSON i18n [extensions/WikiObjectModel] - 10https://gerrit.wikimedia.org/r/123841 (owner: 10Siebrand) [11:04:41] MaxSem: Thanks for looking into that poll thing. Ive restored the docs at https://www.mediawiki.org/wiki/Extension:Poll and removed the XSS template. [11:05:09] (03Abandoned) 10Siebrand: Remove extension code because of pervasive security issues [extensions/Poll] - 10https://gerrit.wikimedia.org/r/123847 (owner: 10Siebrand) [11:15:18] Tobi_WMDE: we tried a fix .. https://gerrit.wikimedia.org/r/#/c/123848/ [11:15:42] (03CR) 10Bartosz Dziewoński: [C: 04-1] MediaWiki:Watchnologintext: Add returnto parameter (032 comments) [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [11:16:24] mutante: I see. Thanks for the update! [11:17:57] (03CR) 10Bartosz Dziewoński: "Looks good, I left some comments inline. Thanks for working on this!" [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [11:26:14] (03CR) 10Siebrand: [C: 04-1] "i18n/L10n reviewed." (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [11:34:42] (03CR) 10Bartosz Dziewoński: MediaWiki:Watchnologintext: Add returnto parameter (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [11:36:25] (03CR) 10Siebrand: MediaWiki:Watchnologintext: Add returnto parameter (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [11:41:59] (03CR) 10Bartosz Dziewoński: "I'm not a huge fan of this per what Daniel said." [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [12:21:28] (03CR) 10MarkAHershberger: [C: 04-1] "Please address the comments I left so that we can merge this." (034 comments) [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 (owner: 10Adi.iiita) [12:24:56] (03PS2) 10UltrasonicNXT: Adding documentation of action messages [extensions/MediaWikiChat] - 10https://gerrit.wikimedia.org/r/122878 (owner: 10GeorgeBarnick) [12:25:10] (03CR) 10UltrasonicNXT: [C: 032 V: 032] Adding documentation of action messages [extensions/MediaWikiChat] - 10https://gerrit.wikimedia.org/r/122878 (owner: 10GeorgeBarnick) [12:45:54] (03CR) 10Tim Landscheidt: [C: 031] FormatJson: Make it possible to change the indent string [core] - 10https://gerrit.wikimedia.org/r/123518 (owner: 10PleaseStand) [12:49:26] (03PS1) 10Gerrit Patch Uploader: Use hasOwnProperty to allow hooks with name of predefined methods [core] - 10https://gerrit.wikimedia.org/r/123856 [12:49:28] (03CR) 10Gerrit Patch Uploader: "This commit was uploaded using the Gerrit Patch Uploader [1]." [core] - 10https://gerrit.wikimedia.org/r/123856 (owner: 10Gerrit Patch Uploader) [13:00:13] (03PS1) 10Manybubbles: Update Elasticsearch installation instructions [search/highlighter] - 10https://gerrit.wikimedia.org/r/123858 [13:01:44] (03PS2) 10Manybubbles: Update Elasticsearch installation instructions [search/highlighter] - 10https://gerrit.wikimedia.org/r/123858 [13:01:46] (03CR) 10jenkins-bot: [V: 04-1] Update Elasticsearch installation instructions [search/highlighter] - 10https://gerrit.wikimedia.org/r/123858 (owner: 10Manybubbles) [13:02:25] (03PS3) 10Manybubbles: Update Elasticsearch installation instructions [search/highlighter] - 10https://gerrit.wikimedia.org/r/123858 [13:02:36] (03CR) 10Manybubbles: [C: 032] Update Elasticsearch installation instructions [search/highlighter] - 10https://gerrit.wikimedia.org/r/123858 (owner: 10Manybubbles) [13:28:58] (03CR) 10Jdlrobson: "I think it useful in reviews to suggest alternative ways to achieve the same end result." [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [13:31:12] (03CR) 10Jdlrobson: "in nd yes I think it is necessary. Increases readability and resets inside skin CSS files suggest that those CSS in common are not sensibl" [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [14:01:45] (03Abandoned) 10Phuedx: [WIP] "Magnify" the Edit tab [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/118478 (owner: 10Phuedx) [14:32:35] (03CR) 10Adi.iiita: LQT should not accept duplicate post (034 comments) [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 (owner: 10Adi.iiita) [14:37:59] (03CR) 10MarkAHershberger: LQT should not accept duplicate post (032 comments) [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 (owner: 10Adi.iiita) [14:40:40] (03PS15) 10Tpt: Addition of a parser test for page= parameter of image inclusion [core] - 10https://gerrit.wikimedia.org/r/102475 [14:42:25] (03CR) 10Tpt: "PS 15: fix the deletion of LoremIpsum.djvu in old parser test runner" [core] - 10https://gerrit.wikimedia.org/r/102475 (owner: 10Tpt) [14:47:42] (03CR) 10jenkins-bot: [V: 04-1] Addition of a parser test for page= parameter of image inclusion [core] - 10https://gerrit.wikimedia.org/r/102475 (owner: 10Tpt) [15:09:29] (03PS1) 10Manybubbles: Fix boost_before for top_scoring [search/highlighter] - 10https://gerrit.wikimedia.org/r/123866 [15:09:31] (03PS1) 10Manybubbles: Formatting fixes [search/highlighter] - 10https://gerrit.wikimedia.org/r/123867 [15:10:59] (03CR) 10Manybubbles: [C: 032] Fix boost_before for top_scoring [search/highlighter] - 10https://gerrit.wikimedia.org/r/123866 (owner: 10Manybubbles) [15:13:05] (03CR) 10Anomie: "Code itself looks good, seems to work. On a purely technical level, I'd +2." [core] - 10https://gerrit.wikimedia.org/r/120858 (owner: 10Umherirrender) [15:16:17] (03CR) 10Manybubbles: [C: 032] Formatting fixes [search/highlighter] - 10https://gerrit.wikimedia.org/r/123867 (owner: 10Manybubbles) [15:16:29] (03CR) 10Siebrand: "I think Matmarex's suggestion should work just fine." [core] - 10https://gerrit.wikimedia.org/r/123837 (owner: 10Konarak) [15:16:45] (03PS3) 10Adi.iiita: LQT should not accept duplicate post [extensions/LiquidThreads] - 10https://gerrit.wikimedia.org/r/123836 [15:17:04] (03PS16) 10Tpt: Addition of a parser test for page= parameter of image inclusion [core] - 10https://gerrit.wikimedia.org/r/102475 [15:18:47] (03PS17) 10Tpt: Addition of a parser test for page= parameter of image inclusion [core] - 10https://gerrit.wikimedia.org/r/102475 [15:19:22] (03PS1) 10Manybubbles: README update [search/highlighter] - 10https://gerrit.wikimedia.org/r/123868 [15:19:36] (03CR) 10Manybubbles: [C: 032] README update [search/highlighter] - 10https://gerrit.wikimedia.org/r/123868 (owner: 10Manybubbles) [15:23:42] (03PS2) 10Jdlrobson: Vector: Simply style external links [core] - 10https://gerrit.wikimedia.org/r/123817 [15:23:59] (03CR) 10Jdlrobson: Vector: Simply style external links (032 comments) [core] - 10https://gerrit.wikimedia.org/r/123817 (owner: 10Jdlrobson) [15:24:45] (03CR) 10jenkins-bot: [V: 04-1] Addition of a parser test for page= parameter of image inclusion [core] - 10https://gerrit.wikimedia.org/r/102475 (owner: 10Tpt) [15:25:31] (03PS1) 10Manybubbles: [maven-release-plugin] prepare release experimental-0.0.2 [search/highlighter] - 10https://gerrit.wikimedia.org/r/123869 [15:25:33] (03PS1) 10Manybubbles: [maven-release-plugin] prepare for next development iteration [search/highlighter] - 10https://gerrit.wikimedia.org/r/123870 [15:25:35] (03PS1) 10Manybubbles: Update readme for release [search/highlighter] - 10https://gerrit.wikimedia.org/r/123871 [15:25:48] (03CR) 10Manybubbles: [C: 032 V: 032] [maven-release-plugin] prepare release experimental-0.0.2 [search/highlighter] - 10https://gerrit.wikimedia.org/r/123869 (owner: 10Manybubbles) [15:25:58] (03CR) 10Manybubbles: [C: 032 V: 032] [maven-release-plugin] prepare for next development iteration [search/highlighter] - 10https://gerrit.wikimedia.org/r/123870 (owner: 10Manybubbles) [15:26:09] (03CR) 10Manybubbles: [C: 032 V: 032] Update readme for release [search/highlighter] - 10https://gerrit.wikimedia.org/r/123871 (owner: 10Manybubbles) [15:28:47] (03PS1) 10Zfilipin: Added files needed for Selenium tests [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/123873 [15:29:13] (03PS2) 10Zfilipin: Added files needed for Selenium tests [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/123873 [15:30:39] (03PS1) 10Bencmq: [WIP] Browser tests for paging nagivation [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123874 [15:33:31] (03PS2) 10Bencmq: [WIP] Browser tests for paging nagivation [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123874 [15:34:24] (03PS1) 10Alex Monk: Show pending changes notice about unstable transclusions in VE [extensions/FlaggedRevs] - 10https://gerrit.wikimedia.org/r/123875 [15:41:59] (03CR) 10Bencmq: "There are two attempts this patch tries to make:" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123874 (owner: 10Bencmq) [15:42:17] (03PS3) 10Zfilipin: WIP User sees call to action after registration [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/123873 [15:43:33] (03PS4) 10Zfilipin: WIP User sees call to action after registration [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/123873 [15:44:44] (03CR) 10Cmcmahon: "Let us know if editable_article looks like a reasonable description of a first browser test (we were just guessing). If it is, we can take" [extensions/GettingStarted] - 10https://gerrit.wikimedia.org/r/123873 (owner: 10Zfilipin) [15:46:41] (03CR) 10Anomie: [C: 04-1] "This is a lot of code duplication. At the PHP level, you could easily add a $expanded parameter (and rename the functions), they're identi" [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/123672 (owner: 10Jackmcbarn) [15:50:08] (03CR) 10Bencmq: [WIP] Browser tests for paging nagivation (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123874 (owner: 10Bencmq) [16:24:36] (03CR) 10Kelson: [C: 032 V: 032] "OK to me. Thx." [qrpedia] - 10https://gerrit.wikimedia.org/r/121661 (owner: 10Addshore) [16:43:39] (03PS1) 10Chad: Strip citation links (like [1], [2], etc) from HTML [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 [16:47:12] (03CR) 10Daniel Friesen: ""Are you telling me skins reference the CSS file commonContent.CSS rather than the module? That is so totally totally broken."" [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [16:52:04] (03CR) 10Jdlrobson: ""That is how the feature was introduced several releases ago. In hindsight it wasn't the best idea, but it's how things are currently expe" [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [16:52:41] (03CR) 10Manybubbles: [C: 032] Strip citation links (like [1], [2], etc) from HTML [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 (owner: 10Chad) [16:57:17] (03PS6) 10Bsitu: Add topic/post summary diff view [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121128 [16:59:19] (03CR) 10Jforrester: Add script to send notifications for new features (032 comments) [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/84625 (owner: 10MarkTraceur) [16:59:21] (03CR) 10Chad: [C: 04-2] "Bleh, didn't test. Only affects body content, not section headers." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 (owner: 10Chad) [17:01:34] (03CR) 10Manybubbles: [C: 04-1] "Probably a good call. I can write some selenium tests for it later on if you'd like." [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 (owner: 10Chad) [17:02:43] (03CR) 10Krinkle: [C: 032] Update jquery.fullscreen to newer version (fix IE11 support) [core] - 10https://gerrit.wikimedia.org/r/122648 (owner: 10Gilles) [17:08:10] (03PS10) 10Bsitu: Add summarize/close/open topic to history/contribution [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123123 [17:09:49] (03PS2) 10Chad: Strip citation links (like [1], [2], etc) from HTML [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 [17:13:06] (03CR) 10Krinkle: [C: 04-1] Disable NavigationPopups (Gadget) if Popups (Extension) is enabled [extensions/Popups] - 10https://gerrit.wikimedia.org/r/120188 (owner: 10Prtksxna) [17:18:31] (03CR) 10Krinkle: "@TheDJ: There's a lot of different things we could do if we want it to be back compat like that. For one we could disable the user prefere" [extensions/Popups] - 10https://gerrit.wikimedia.org/r/120188 (owner: 10Prtksxna) [17:26:55] (03CR) 10Bsitu: [C: 032] Remove couple of methods by using built-in sort [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123688 (owner: 10Matthias Mullie) [17:29:11] (03CR) 10Krinkle: "@DanielFriesen: Yes, I got that. How is that a problem? This is a hook, not an event. Like domready, it has memory of the list firing. And" [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [17:43:38] (03PS1) 10Manybubbles: WIP: Add support for the experimental highlighter [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123885 [17:47:47] (03CR) 10Bsitu: [C: 032] "I will merge this for now since this will address the issue of escaping html links. Better implementation of Templating->getContent() can" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121385 (owner: 10Matthias Mullie) [17:55:07] (03CR) 10TheDJ: [C: 04-1] Use hasOwnProperty to allow hooks with name of predefined methods (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123856 (owner: 10Gerrit Patch Uploader) [17:58:21] (03CR) 10Manybubbles: [C: 04-1] "I'm verifying that I didn't break any thing when you leave those settings false. I don't expect things to work 100% the same when you set" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123885 (owner: 10Manybubbles) [18:02:45] (03CR) 10Bsitu: API: Remove 'flowaction' parameter from list=flow (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123174 (owner: 10Legoktm) [18:05:14] (03CR) 10Siebrand: Add script to send notifications for new features (031 comment) [extensions/BetaFeatures] - 10https://gerrit.wikimedia.org/r/84625 (owner: 10MarkTraceur) [18:09:06] Hello. https://bugzilla.wikimedia.org/show_bug.cgi?id=4676 has a patch needing review and I'm wondering who I should ask about that. [18:12:16] (03PS2) 10EBernhardson: Provide better URLs to IRC rcfeed [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123114 [18:12:38] Scott_WP: Hmm. https://gerrit.wikimedia.org/r/112920 has one +1 so far. Probably welcomes another review and then somebody merging it :-/ [18:13:36] andre__: Right... I'm not a dev, I was just involved in shepherding along the RfC getting approval for this. So I don't really know much about the process :-) [18:13:43] yeah :-/ [18:16:22] * Scott_WP aspires to find someone to help out but not be annoying in the process. [18:16:37] o/` highway to the danger zone o/` [18:17:45] Scott_WP: I added Reedy to the review for Iec1ee0a. This seems "shell bug" related and he's the best I know for handling that kind of review. [18:18:05] Thank you! [18:18:50] (03PS2) 10Konarak: MediaWiki:Watchnologintext: Add returnto parameter [core] - 10https://gerrit.wikimedia.org/r/123837 [18:20:47] (03CR) 10Krinkle: "Could be a coincedence but it seems that once again the test suite of mwext-CirrusSearch-phantomjs is clogging up Jenkins for over 35 minu" [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123885 (owner: 10Manybubbles) [18:22:28] Krinkle: might be that that suite is just too long comfort [18:22:46] i can/should reduce it [18:24:26] (03CR) 10Daniel Friesen: ""Well let's fix that now. How can we make this better? Surely with this approach, we simply need to identify those skins and make them ref" [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [18:26:37] (03PS2) 10Jackmcbarn: Add frame.uargs (unexpanded arguments) [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/123672 [18:27:24] (03CR) 10Siebrand: [C: 04-1] "i18n/L10n reviewed. This needs to be rebased, because of the JSON i18n migration." [core] - 10https://gerrit.wikimedia.org/r/118113 (owner: 10Anomie) [18:33:52] (03PS2) 10Ori.livneh: mw.hook: Make hook.fire actually chainable [core] - 10https://gerrit.wikimedia.org/r/123820 (owner: 10Krinkle) [18:34:43] paravoid: around? [18:35:40] (03CR) 10Ori.livneh: [C: 032] mw.hook: Make hook.fire actually chainable [core] - 10https://gerrit.wikimedia.org/r/123820 (owner: 10Krinkle) [18:37:28] (03CR) 10Daniel Friesen: "Now for alternatives. Given that the real core issue here is that Vector wants a completely different thumbnail style than other skins and" [core] - 10https://gerrit.wikimedia.org/r/123816 (owner: 10Jdlrobson) [18:41:36] (03CR) 10Umherirrender: "When visiting a old revision which is revision deleted the sysop has to click another link to get the unhide=1 url, than he will get the r" [core] - 10https://gerrit.wikimedia.org/r/120858 (owner: 10Umherirrender) [18:46:46] jackmcbarn: not much or for long, what's up? [18:47:15] paravoid: regarding https://gerrit.wikimedia.org/r/#/c/122283/ , what exactly was the issue, and is it something i can fix, or can those just never be categorized? [18:47:50] http://ganglia.wikimedia.org/latest/graph.php?r=month&z=xlarge&c=Memcached+eqiad&m=cpu_report&s=by+name&mc=2&g=network_report [18:48:11] (03PS1) 10Alex Monk: Fix POST queries [extensions/ApiSandbox] - 10https://gerrit.wikimedia.org/r/123897 [18:48:42] wow, i knew we'd find a lot of problems with the fonts change, but that it would be so much... [18:49:04] (03CR) 10Alex Monk: "Caused bug 63537, fixed in Iaf56d357" [extensions/ApiSandbox] - 10https://gerrit.wikimedia.org/r/112307 (owner: 10Alex Monk) [18:49:18] jackmcbarn: also, http://graphite.wikimedia.org/render/?width=586&height=308&_salt=1396244200.34&from=-2weeks&target=MediaWiki.SpamBlacklist.getRegex.count [18:49:30] also, nice case of how testing something in beta for 5 months basically makes shit difference in feedback. [18:49:45] thedj: context? [18:50:13] apparently truly nothing in the world other than wikipedia resembles wikipedia, not even it's beta testgroup :D [18:50:24] paravoid: https://www.mediawiki.org/wiki/Talk:Typography_refresh [18:50:36] jackmcbarn: so, I'm not intimately familiar with SpamBlacklist; you'd have to ask regular MW reviewers for that [18:50:39] users with not font smoothing enabled in windows [18:50:49] paravoid: kk, thanks [18:51:00] jackmcbarn: but this was causing problems as you see from the graphs above and I emergency-reverted it [18:51:36] basically, all the problems i feared, from past en.wp experiences with changing fonts for users :( [18:51:58] the web and OS'es are simply not ready to mess with fontstacks [18:52:06] other than in select cases [18:53:03] especially the feedback from polish wikt is interesting. [18:53:07] (03CR) 10MaxSem: [C: 032] Fix POST queries [extensions/ApiSandbox] - 10https://gerrit.wikimedia.org/r/123897 (owner: 10Alex Monk) [18:55:00] (03CR) 10EBernhardson: [C: 031] Add "cur" & "prev" difflinks in history, similar to article history [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123215 (owner: 10Matthias Mullie) [18:55:13] (03CR) 10Jforrester: "If this is going to happen, it should remove the types of the images from wikitext, and make these changes to core, not just one skin." [core] - 10https://gerrit.wikimedia.org/r/123815 (owner: 10Jdlrobson) [18:59:14] yikes, volunteers wrote a VE reference gadget, while the VE team was writing the same ??? [18:59:41] (03CR) 10EBernhardson: "also https://gerrit.wikimedia.org/r/#/c/121855 refactors most of the url generation, could these updates be rebased ontop of that?" (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123216 (owner: 10Matthias Mullie) [19:00:15] https://en.wikipedia.org/wiki/Wikipedia:Village_pump_(technical)#Adding_citations_in_VE [19:02:32] (03PS3) 10Jackmcbarn: Allow InputBoxes to be used to move pages [extensions/InputBox] - 10https://gerrit.wikimedia.org/r/97559 [19:03:25] (03CR) 10Daniel Friesen: [C: 04-1] "Ok thanks, I thought that hooks worked like our normal hooks or events, not something that memorizes the last run." [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [19:04:43] (03Abandoned) 10Jackmcbarn: Unlink/remove blacklisted links during parse [extensions/SpamBlacklist] - 10https://gerrit.wikimedia.org/r/118636 (owner: 10Jackmcbarn) [19:18:47] (03CR) 10EBernhardson: UUID revamp (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123691 (owner: 10Matthias Mullie) [19:21:05] What's wrong with Jenkins? [19:23:06] (03PS8) 10Jackmcbarn: Allow moving category pages [core] - 10https://gerrit.wikimedia.org/r/111096 [19:28:27] (03Merged) 10jenkins-bot: Fix POST queries [extensions/ApiSandbox] - 10https://gerrit.wikimedia.org/r/123897 (owner: 10Alex Monk) [19:28:34] there it is [19:32:38] (03Merged) 10jenkins-bot: Remove couple of methods by using built-in sort [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123688 (owner: 10Matthias Mullie) [19:32:57] (03Merged) 10jenkins-bot: Don't escape the userlinks displayed in "This topic was suppressed by " [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121385 (owner: 10Matthias Mullie) [19:35:46] (03CR) 10jenkins-bot: [V: 04-1] Add frame.uargs (unexpanded arguments) [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/123672 (owner: 10Jackmcbarn) [19:39:50] I notice https://integration.wikimedia.org/dashboard/ shows a bunch of missing pmtpa entries [19:42:02] (03PS3) 10Jackmcbarn: Add frame.uargs (unexpanded arguments) [extensions/Scribunto] - 10https://gerrit.wikimedia.org/r/123672 [19:43:07] (03Merged) 10jenkins-bot: Update jquery.fullscreen to newer version (fix IE11 support) [core] - 10https://gerrit.wikimedia.org/r/122648 (owner: 10Gilles) [19:43:09] (03Merged) 10jenkins-bot: mw.hook: Make hook.fire actually chainable [core] - 10https://gerrit.wikimedia.org/r/123820 (owner: 10Krinkle) [19:43:36] (03PS2) 10Krinkle: mw.hook: Implement hook.once [core] - 10https://gerrit.wikimedia.org/r/123821 [19:43:42] (03CR) 10Krinkle: [C: 04-2] mw.hook: Implement hook.once [core] - 10https://gerrit.wikimedia.org/r/123821 (owner: 10Krinkle) [19:43:53] (03PS2) 10Krinkle: resourceloader: Don't add superfluous line breaks and semicolons [core] - 10https://gerrit.wikimedia.org/r/123822 [19:44:26] (03PS4) 10Krinkle: Deprecate mw.util.$content [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [19:44:28] (03PS1) 10Krinkle: [WIP] mediawiki.util: Clean up whitespace [core] - 10https://gerrit.wikimedia.org/r/123912 [19:45:55] (03PS3) 10Krinkle: resourceloader: Don't add superfluous line breaks and semicolons [core] - 10https://gerrit.wikimedia.org/r/123822 [19:45:59] (03CR) 10Krinkle: resourceloader: Don't add superfluous line breaks and semicolons (031 comment) [core] - 10https://gerrit.wikimedia.org/r/123822 (owner: 10Krinkle) [19:52:12] (03CR) 10Krinkle: [C: 04-1] "@DanielFriesen: Indeed, I've come to agree. We can discourage it for purposes where the user really just needs #mw-content-text (it'll be " [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [19:52:26] (03PS2) 10Krinkle: [WIP] mediawiki.util: Clean up whitespace [core] - 10https://gerrit.wikimedia.org/r/123912 [19:52:40] (03PS3) 10Krinkle: mediawiki.util: Clean up whitespace [core] - 10https://gerrit.wikimedia.org/r/123912 [20:01:22] (03CR) 10jenkins-bot: [V: 04-1] resourceloader: Don't add superfluous line breaks and semicolons [core] - 10https://gerrit.wikimedia.org/r/123822 (owner: 10Krinkle) [20:13:09] (03PS4) 10Krinkle: resourceloader: Don't add superfluous line breaks and semicolons [core] - 10https://gerrit.wikimedia.org/r/123822 [20:13:47] (03CR) 10Daniel Friesen: [C: 04-1] "@Krinkle, yup feel free to update $content explaining its difference from #mw-content-text and recommending use of the hook if the code do" [core] - 10https://gerrit.wikimedia.org/r/123559 (owner: 10Gerrit Patch Uploader) [20:17:55] where is -{foo|bar}- syntax documented at ? [20:19:41] That's language converter [20:20:50] Which is a black box nobody understands... [20:21:17] Well actually, liangent and cscott are probably the most likely to know about it. I'm not aware of any docs anywhere [20:22:07] yeah, I just accidentally found out that the app does variant conversion for you automatically if you are logged in. [20:22:10] very black boxey [20:23:12] Basically I prey I just never get asked to fix a bug in it [20:24:01] Also, if you're looking to test it, its much easier to test in serbian then it is to test in chinese [20:24:38] bawolff: oh, right. [20:24:47] bawolff: we don't really have full fledge support for it the first time, I think [20:25:37] Since to an english speaker, traditional Chinese characters look a lot like simplified Chinese, but latin and Cyrillic look quite different [20:25:59] right [20:26:21] I think for now the only major support is just that if you have it set in your user preferences in desktop then the app will respect that [20:33:57] i bet i can smuggle state between #invoke's with that [20:34:29] (03PS1) 10Raimond Spekking: Revert "Localisation updates from https://translatewiki.net." [core] - 10https://gerrit.wikimedia.org/r/124000 [20:37:50] (03PS2) 10Manybubbles: Add support for the experimental highlighter [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123885 [20:42:04] jackmcbarn: I think there's an open rfc to change it due to that reason [20:42:24] andre__: Hmm, you're not an admin at mediawiki.org? That seems odd [20:42:36] bawolff, why should I be? :) [20:43:18] Well you're a developer and all that [20:43:44] I just saw your comment at https://www.mediawiki.org/wiki/Thread:Project:Support_desk/Spelling_error_in_desc_of_a_gadget [20:43:59] why is that even in core and not an extension? [20:44:00] If you asked Gloria, he would probably give you admin rights on the spot [20:58:54] Hi all [20:59:06] csteipp: yt? [21:09:34] (03PS2) 10Brian Wolff: Add two Commons templates to GWToolset [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/123248 (owner: 10Dan-nl) [21:09:41] (03CR) 10Brian Wolff: [C: 032] Add two Commons templates to GWToolset [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/123248 (owner: 10Dan-nl) [21:09:46] (03Merged) 10jenkins-bot: Add two Commons templates to GWToolset [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/123248 (owner: 10Dan-nl) [21:11:10] (03PS1) 10MaxSem: Optional coordinates compression mode [extensions/GeoData] - 10https://gerrit.wikimedia.org/r/124008 [21:17:51] (03CR) 10Bsitu: [C: 031] "Looks good to me but not sure how to test" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123114 (owner: 10EBernhardson) [21:17:59] (03PS1) 10Manybubbles: Cirrus now needs the Cite extension for testing [integration/jenkins-job-builder-config] - 10https://gerrit.wikimedia.org/r/124011 [21:18:39] (03CR) 10EBernhardson: "I wasn't really sure either, which is why i wrote the HookTest.php. Better than nothing i suppose." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123114 (owner: 10EBernhardson) [21:20:17] (03PS2) 10MaxSem: Optional coordinates compression mode [extensions/GeoData] - 10https://gerrit.wikimedia.org/r/124008 [21:21:04] (03PS1) 10Brian Wolff: Fix some minor HTML issues [extensions/GWToolset] - 10https://gerrit.wikimedia.org/r/124012 [21:23:27] (03CR) 10EBernhardson: Fix a few bugs where Flow was not working on pristine pages. (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123812 (owner: 10Werdna) [21:36:18] (03PS2) 10Bsitu: Serialize UUID into something more compact [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123689 (owner: 10Matthias Mullie) [21:36:29] (03CR) 10EBernhardson: Implement multiPut method. (032 comments) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/120734 (owner: 10Werdna) [21:36:41] (03CR) 10JGonera: [C: 04-1] "1. This breaks the style guide (make kss while in resources/ directory)." [core] - 10https://gerrit.wikimedia.org/r/123804 (owner: 10SG) [21:39:58] (03CR) 10Bsitu: [C: 032] Serialize UUID into something more compact [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123689 (owner: 10Matthias Mullie) [21:40:12] (03Merged) 10jenkins-bot: Serialize UUID into something more compact [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123689 (owner: 10Matthias Mullie) [21:41:31] (03CR) 10EBernhardson: [C: 032] Create test for bug 59637 [extensions/Flow] - 10https://gerrit.wikimedia.org/r/108879 (owner: 10Werdna) [21:41:47] (03Merged) 10jenkins-bot: Create test for bug 59637 [extensions/Flow] - 10https://gerrit.wikimedia.org/r/108879 (owner: 10Werdna) [21:41:57] (03PS1) 10BryanDavis: Remove use of deprecated --extended flag with mwversionsinuse [tools/scap] - 10https://gerrit.wikimedia.org/r/124015 [21:47:08] (03PS3) 10EBernhardson: Warn when __toString is needed on UUID [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123692 (owner: 10Matthias Mullie) [21:47:15] (03PS1) 10BryanDavis: Flush logger before exit [tools/scap] - 10https://gerrit.wikimedia.org/r/124016 [21:47:27] (03CR) 10jenkins-bot: [V: 04-1] Warn when __toString is needed on UUID [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123692 (owner: 10Matthias Mullie) [21:48:22] (03PS3) 10Manybubbles: Strip citation links (like [1], [2], etc) from HTML [extensions/CirrusSearch] - 10https://gerrit.wikimedia.org/r/123878 (owner: 10Chad) [21:49:20] (03CR) 10EBernhardson: [C: 04-1] "This causes errors in phpunit, semi out of our control:" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123692 (owner: 10Matthias Mullie) [21:52:11] (03CR) 10EBernhardson: API: Remove 'flowaction' parameter from list=flow (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123174 (owner: 10Legoktm) [21:55:46] (03PS1) 10MaxSem: Use indexed bounding box [extensions/GeoData] - 10https://gerrit.wikimedia.org/r/124017 [21:56:15] (03CR) 10EBernhardson: [C: 032] "I still don't like that its 3 images for what is basically 1 image at different levels of gray(couldn't that be done in css with one image" [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123832 (owner: 10Werdna) [21:56:32] (03Merged) 10jenkins-bot: Add a hover state for topic viewing filter buttons [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123832 (owner: 10Werdna) [22:08:16] renoirb: I'm back [22:08:22] Hey! [22:08:44] I’m trying to authenticate with the extension, I’m a bit confused. [22:09:38] I’m attempting to create a provider within oauth.io to authenticate with a MediaWiki installation [22:10:16] renoirb: As in https://oauth.io/ ? [22:10:26] csteipp: My question is when I make the call, what is oob in http://www.mediawiki.org/wiki/User:CSteipp/OAuth_demo_client [22:10:55] For MediaWiki, it must be the literal string 'oob' [22:10:59] yes, oauth.io but a local deployment. [22:11:06] what is that string litteral?? [22:11:14] (required by the oauth spec for "out of band" negotiated url) [22:11:18] i searched the code, I thought it was something else [22:11:24] oh!!! Out Of Band!! [22:11:26] :) [22:11:49] So, in my consumer, I have to set that explicitly [22:11:53] ok. [22:11:56] Yeah, since you put in the url when you register the consumer, it's "oob" [22:13:01] aka as "boo spelt backwards" [22:13:05] And the base url has to be “/w/index.php?title=Special:Oauth” [22:13:24] otherwise it adds it, i think and it breaks signatures (i’m rephrasing what I read) [22:13:54] renoirb: Yeah, you need to account for "title=Special:OAuth" as a parameter [22:13:56] So, for a token request, I guess if we append ‘/initiate&format=json&oauth_callback=oob’, right? [22:14:51] :wq [22:14:56] oops :/ [22:16:02] Yeah, to start you want that. Then you send users off to /authorize. Then you get the authorized token from /token [22:18:17] So, what is the verification code, that is asked in the example file? [22:18:23] http://www.mediawiki.org/wiki/User:CSteipp/OAuth_demo_client ^ [22:18:48] how do I generate that token? [22:19:17] After the user logs in to wikipedia (or your wiki), and clicks to approve the application, the wiki adds the verification code to the url that they redirect the user back to on your website. [22:19:56] ok. [22:19:59] So your Consumer needs to collect that, and send it back when you make the call to /token [22:21:02] But with that code OAuth_demo_client, its only a terminal php file. I do not see anywhere where to get the string there. [22:21:05] (03PS1) 10MaxSem: Add debug information [extensions/GeoData] - 10https://gerrit.wikimedia.org/r/124022 [22:21:11] And i’m even tailing the database read and writes :/ [22:22:02] Hi. I'm having trouble trouble with the Scibunto extension giving me Lua errors and I read that someone was able to figure it out here [22:22:12] Oh, yeah, if you run the demo, you'll see the verification code in the url that your user is redirected to, and then you can cut and paste that into the demo app [22:22:39] Oh, that’s what I thought. I read, but could not find. I should’ve dug deeper [22:22:53] let me read the urlencoded string again. [22:23:06] Yeah, the wiki will generate a new one for each authorization, so it will change each time you run it [22:24:00] Is it the string here, in ‘ OAuthSignatureMethod_HMAC_SHA1::build_signature: HMAC Key:’ [22:24:35] hmm? [22:24:39] Not sure I follow [22:25:07] I’m reading the echoed strings I get when I execute the script [22:26:46] oh, i think I get it. [22:27:47] When I call the script, what I get is a 404 from my web server. [22:28:21] Hmm.. do you have extension installed? [22:28:21] I guess that if I had openssl installed, i would have had something to read from instead of the 404 page html [22:28:51] A few. But roughly basic WMF version [22:28:55] openssl shouldn't be an issue, if you set $wgMWOAuthSecureTokenTransfer = false [22:29:14] I thought I had it configured already [22:29:17] let me double check [22:29:47] Well, if you registered your consumer, then it would be installed [22:30:08] But I'm not sure why Special:OAuth would give you a 404 [22:31:23] strange. [22:31:33] Even tough the script says “Calling: http://sandbox.webplatform.org/w/index.php?title=Special:OAuth/initiate?format=json&oauth_callback=oob” [22:31:49] The line after is I have: Returned: …. a 404 error page HTML [22:33:36] tsss, the code has curl_setopt with ssl [22:39:38] It worked! [22:40:13] I figured out that I tampered with the options in the script you provided [22:50:28] (03CR) 10Bsitu: Standardize URL generation (034 comments) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/121855 (owner: 10EBernhardson) [22:51:00] jackmcbarn: Do you need adminship? [23:08:58] Are you still there csteipp ? [23:09:23] When I call the Special:OAuth/token, I get {"error":"mwoauthdatastore-bad-token"} [23:09:55] renoirb: yeah [23:10:11] (I commented the curl_setopt to not use SSL in my case. [23:10:13] ) [23:10:33] But, when I read the response, I wonder where it broke. [23:18:11] renoirb: Yeah, that's not an issue with SSL. I'm looking up what that one is... [23:18:52] renoirb: So that's when you call /token, right? [23:23:28] (03CR) 10Gergő Tisza: [C: 032] Improving zoom of embedded map, on address search [extensions/UploadWizard] (osm) - 10https://gerrit.wikimedia.org/r/118978 (owner: 10Inchikutty) [23:23:41] (03Merged) 10jenkins-bot: Improving zoom of embedded map, on address search [extensions/UploadWizard] (osm) - 10https://gerrit.wikimedia.org/r/118978 (owner: 10Inchikutty) [23:33:48] (03CR) 10Werdna: "Hmph, this is actually missing RTL, which I didn't mention in the commit message. Will need to fix that up." [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123832 (owner: 10Werdna) [23:35:13] (03CR) 10Werdna: Fix a few bugs where Flow was not working on pristine pages. (031 comment) [extensions/Flow] - 10https://gerrit.wikimedia.org/r/123812 (owner: 10Werdna) [23:36:14] no, authorize [23:36:21] csteipp: I just created an issue https://github.com/oauth-io/oauthd/issues/70 [23:39:51] Oh renoirb: /authorize is where you send the user's browser-- so that won't be json formatted [23:40:09] sure. [23:40:11] wait a sec [23:40:27] And it should contain the request/temp token [23:40:28] I got detailed in /oauth-io/oauthd/issues/70 describes what I get [23:41:16] Wait a sec csteipp, it is at Special:OAuth/token [23:41:26] Ok, that makes more sense :) [23:41:46] Gloria: what, where? [23:41:48] Sure, i was creating the ticket, my brain might have reboted [23:41:52] That error means the verification code you sent wasn't valid for your consumer [23:42:15] jackmcbarn: :-) [23:42:21] i don't think so atm [23:42:35] i think bawolff was talking about andre__ not having it [23:42:48] What I got before “Enter verification code” * Connection #0 to host sandbox.webplatform.org left intact [23:42:49] Returned: {"key":"IM_PASTING_THIS ","secret”:”…”,”oauth_callback_confirmed":"true"} [23:43:15] Cool, that's the right response for /initialize [23:43:37] THen you send the user to /authorize with that key that you got from /initialize [23:43:40] And I think that ‘IM_PASTING_THIS’ is what I should have sent, right? [23:44:28] Yeah, that would be your oauth_token parameter. [23:44:38] Your oauth_consuer_key is your app's key [23:45:11] And then in the web browser, you click "Accept", and that sends you to the callback url [23:45:15] that key, in the code, seems to be oauth_verifier. [23:45:17] that’s confusing [23:46:14] oauth_verifier is the verification code, yeah. [23:46:27] yeah, at line 72 [23:47:21] So yeah, when you then call /token, it's either got the wrong verification code, or the wrong consumer key [23:47:37] Both will be in your wiki's debug log-- that would be a good place to start [23:47:52] Look for "Getting new access token for token , consumer . " [23:48:02] oh, when I use the browser, I get Invalid format state must be present [23:48:22] That's odd.. [23:48:30] oh, right. [23:48:36] let me see how I configured that consumer in MW [23:48:57] I have to run, might be back on in a bit. Anomie can also help out if you see him around.. [23:49:06] let me see something [23:50:21] jackmcbarn: Ah, thanks. [23:50:42] Strange, it sent to https://ssl.webplatform.org/oauth … even though i’m following the link the shell script gave me.