[15:49:16] (03CR) 10Brion VIBBER: [C: 032] "I'm just gonna self-merge this since nobody's active right now and it's using the existing defs." [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/291921 (https://phabricator.wikimedia.org/T108172) (owner: 10Brion VIBBER) [15:49:58] (03CR) 10Brion VIBBER: [C: 032] "I'm just going to self-merge this for now since nobody's active right now in TMH." [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/289900 (https://phabricator.wikimedia.org/T135581) (owner: 10Brion VIBBER) [15:51:21] (03Abandoned) 10Brion VIBBER: WIP: very basic TimedMediaHandler frontend for mobile [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/217485 (owner: 10Brion VIBBER) [15:53:46] (03PS4) 10Brion VIBBER: Update ogv.js to 1.1.2-alpha.7 [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/292114 (owner: 10Paladox) [15:57:58] (03Merged) 10jenkins-bot: Enable 720p and 1080p .ogv Ogg Theora transcodes [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/291921 (https://phabricator.wikimedia.org/T108172) (owner: 10Brion VIBBER) [15:58:34] (03Merged) 10jenkins-bot: Don't resample audio with ffmpeg2theora, it's unreliable [extensions/TimedMediaHandler] - 10https://gerrit.wikimedia.org/r/289900 (https://phabricator.wikimedia.org/T135581) (owner: 10Brion VIBBER) [20:29:26] 06Multimedia, 10Analytics, 06Commons, 10Tabular-Data, and 3 others: Allow structured datasets on a central repository (CSV, TSV, JSON, GeoJSON, XML, ...) - https://phabricator.wikimedia.org/T120452#2397626 (10Yurik) [20:31:53] 06Multimedia, 10Analytics, 06Commons, 10Tabular-Data, and 3 others: Allow structured datasets on a central repository (CSV, TSV, JSON, GeoJSON, XML, ...) - https://phabricator.wikimedia.org/T120452#2397636 (10Yurik) [20:31:59] 06Multimedia, 10Analytics, 06Commons, 10Tabular-Data, and 4 others: Review shared data namespace (tabular data) implementation - https://phabricator.wikimedia.org/T134426#2397639 (10Yurik) [22:11:51] 06Multimedia, 10MediaWiki-Uploading, 13Patch-For-Review: Create check hook before really uploading file with infos about description page and file - https://phabricator.wikimedia.org/T89302#2397917 (10matmarex) Some of the changes I'm making will change the action=upload responses for non-hook-related errors... [22:52:31] MatmaRex, prtksxna, busy today? [22:56:30] marktraceur: i don't really have anything to say. been testing the abusefilter patches, and slacking off [22:56:37] OK. [22:56:38] i can show up if we have something to discuss though [22:56:42] Not really [22:57:24] MatmaRex, prtksxna, I was going to ask what you thought about file annotations...it seems like we can have a poorly structured but flexible format, or a well-structured but rigid format. I'm not sure which I prefer. [22:58:57] what exactly do you mean? [23:02:30] MatmaRex: My two thoughts right now are 'wikitext' which is actually poorly formed XML, or JSON with no wikitext support [23:05:33] marktraceur: hm. you're going with a separate namespace for annotations, right? i think JSON has already became pretty much a standard for things like this [23:06:11] marktraceur: why couldn't we have wikitext in JSON, though? i guess you'd need to parse every annotation separately, but surely that can be done somehow. [23:06:27] It can, but it would be slow [23:06:29] (a standard in MediaWiki extensions) [23:06:35] hmm. [23:06:36] I guess I could do that in a new API module. [23:06:37] Hmmmm [23:06:53] it's not like we expect to have hundreds upon hundreds of annotations? or do we? [23:07:03] i guess if we do timed media, maybe… [23:07:14] is calling the parser repeatedly really that slow? [23:07:41] Probably not in PHP, I was thinking I'd do it on the client via the API, but having a new module is probably easier [23:08:22] I'll do some profiling with large numbers of annotations maybe. [23:08:26] Thanks! [23:08:54] ah. yeah, having a separate API module that would parse them all sounds like a good idea [23:09:21] doing separate action=parse API calls for each annotations less so :) [23:10:54] Yeah that's why I was unhappy with that as a solution but this seems way nicer