[02:36:47] why does jenkins make my browser lock up when I click on the "more" link on build history? [02:37:45] UIs made this decade really suck [02:38:27] I want Java Web Start back [02:41:51] it's because of getElementOverflowParams in hudson-behavior.js [02:41:57] it forces several thousand layouts [02:43:09] wow, this is some awful javascript [02:46:17] this is the highlight: [02:46:21] // First we force it to wrap so we can get those dimension. [02:46:21] // Then we force it to "nowrap", so we can get those dimension. [02:46:21] // We can then compare the two sets, which will indicate if [02:46:23] // wrapping is potentially happening, or not. [02:48:57] this check is performed for each cell, and the checks queries the node's .clientWidth / .clientHeight, which forces the browser to do a full layout [04:01:47] 6MediaWiki-Core-Team, 10MediaWiki-User-blocking, 7Performance: ipb_range_start and ipb_range_end should be blank for single-IP blocks - https://phabricator.wikimedia.org/T51504#1099494 (10tstarling) a:5tstarling>3None [05:53:32] 6MediaWiki-Core-Team: CSS validator survey and design work - https://phabricator.wikimedia.org/T989#1099526 (10tstarling) I reviewed HTML Purifier, which uses [[http://csstidy.sourceforge.net/|CSSTidy]] for its CSS tokenizer. * CSSTidy's style is quite dated, it seems to be written for PHP 4. It has hardly been... [08:04:17] 6MediaWiki-Core-Team, 10Analytics, 10Wikimedia-Site-requests: Ran out of captcha images - https://phabricator.wikimedia.org/T91760#1099579 (10matthiasmullie) 5Open>3Resolved a:3matthiasmullie Seems like someone has already fixed this. Special:UserLogin&type=signup works again and the exception is no lo... [11:06:40] NoScript is the solution. [11:06:54] A website which requires JavaScript to be read is rarely worth visiting anyway [16:28:39] anomie: I'm in the weeds on creating the diagram(s) I said I would create for today :/ [16:28:50] bd808: Anything I can do to help? [16:29:16] I'm not sure what I though I was going to diagram exactly. [16:29:44] I started making a sequence diagram for SpecialUserLogin talking to AuthManager [16:29:54] and that mostly led to a lot of questions [16:30:34] Then I thought maybe to try listing the authentication states that AuthManager needs to track [16:31:19] no session; empty session; provider selected; user found; ... [16:31:48] I think I'll keep going with that actually. I may make what else is needed more clear [16:33:26] bd808: If it helps, https://gerrit.wikimedia.org/r/195297 [16:33:49] cool. That may help for sure [16:34:16] I failed to resist the temptation to actually write code for the "AuthnSession" object... [16:34:42] (which is the interesting part of https://www.mediawiki.org/wiki/Requests_for_comment/AuthManager#Session_management) [16:35:39] csteipp, legoktm: https://gerrit.wikimedia.org/r/195297 is up, if you're bored enough to want to look at it before the meeting. [16:35:44] * anomie goes to heat lunch [16:39:18] <_joe_> bd808: I added a few people as reviewers to https://gerrit.wikimedia.org/r/#/c/194830/, do you think I am missing someone who should take a look (you included)? [16:40:58] _joe_: Looks like you have the heavy hitters on there. I can take a look later today too on a just in case basis [16:44:46] _joe_: I see you have an extra newline between the "Bug:" footer and the "Change-Id:" footer in the commit message, which makes it not post status on the bug [16:45:13] <_joe_> anomie: will fix it [16:45:21] <_joe_> thanks [17:35:10] <_joe_> SMalyshev, bd808 http://blog.blazegraph.com/?p=826 [17:35:14] <_joe_> did you know that? [17:37:27] I ... no [17:38:21] well I knew that manybubbles had decided that. I didn't know there was a matching blog post from them [17:38:35] see https://lists.wikimedia.org/pipermail/wikidata-tech/2015-March/000740.html [17:41:08] such marketing [17:41:09] much PR [17:42:04] <_joe_> yeah I know of Nik's email of course [17:42:14] <_joe_> I just didn't know they blogged about it [17:42:52] It was announced on a public list so there's no reason to embargo them from talking about it [17:43:27] It's a net positive thing I think for them to be excited and invested in our use of their software [20:16:06] bd808: Is MWLoggerFactory::getInstance() the right way to get a logger, or is there an even newer way? [20:16:29] MWLoggerFactory::getInstance() is the right way [20:16:50] MWLogger::getInstance() is the way I'm wanting to kill off [20:48:18] 6MediaWiki-Core-Team, 7Epic: MediaWiki multi-datacenter investigation and work - https://phabricator.wikimedia.org/T88445#1101374 (10aaron) [20:50:56] 6MediaWiki-Core-Team, 7Epic: Investigate a better alternative for making user_newtalk talk updates - https://phabricator.wikimedia.org/T92007#1101397 (10aaron) 3NEW a:3aaron [20:58:33] 6MediaWiki-Core-Team, 10SUL-Finalization, 6operations: db1068 (s4/commonswiki slave) is missing data about at least 6 users - https://phabricator.wikimedia.org/T91920#1101449 (10Legoktm) >>! In T91920#1099367, @Springle wrote: > Binary logs don't go back that far. Starting a sync check to see how large the p... [21:03:37] 6MediaWiki-Core-Team, 7Epic: MediaWiki multi-datacenter investigation and work - https://phabricator.wikimedia.org/T88445#1101468 (10Joe) I have made a patch that is somewhat related to this in https://gerrit.wikimedia.org/r/#/c/194830/ [22:04:21] 6MediaWiki-Core-Team: Investigate a better alternative for making user_newtalk talk updates - https://phabricator.wikimedia.org/T92007#1101941 (10bd808) [22:24:27] 6MediaWiki-Core-Team, 6CA-team, 10SUL-Finalization, 6operations: db1068 (s4/commonswiki slave) is missing data about at least 6 users - https://phabricator.wikimedia.org/T91920#1102019 (10Philippe-WMF)