[00:07:06] He went to bed a couple of hours ago [00:09:21] cartoons of pokeman gladiators fighting mohammed? [00:16:11] heh! [00:16:52] i went and saw the first pokemon movie in the theater, actually, and walked out thinking, "this movie teaches children that enslaving creatures and forcing them to fight in gladiatoral combat is a good thing!" [00:19:12] You mean, it isn't? :S [19:50:23] I have to add this to the quotes on Code Review [19:50:39] "I see your point. However... I personally prefer consistency over sanity" [20:01:18] if ye can't deal with insanity, you chose the wrong software to work on :p [20:29:59] robla: My fixmes are 1) a rev that has been taken out of 1.17 and 2) a very old one that was part of the 1.16 deploy and marked fixme because it outputs a number in the API that's off in rare cases [20:33:06] then fix it :) [20:33:56] heh [20:33:58] If only I knew how [20:34:15] Tim gave a vague answer along the lines of "hmm, maybe this should be refactored slightly" and it's all parser internals, so... :) [20:35:04] hexmode: can you figure out how to dispatch Roan's revs? [20:35:36] I suspect the answer may be for 1) mark as "nodeploy", and 2) file a bug in BZ, then mark as "resolved" in code review [20:35:38] "dispatch" ?? [20:35:44] review? [20:36:03] Do we care? I mean I don't believe we should hit fixme-zero, just have assurance that each outstanding fixme is relatively harmless [20:36:36] RoanKattouw: having fixmes laying around in the list that we don't care about causes us to miss ones we do [20:37:03] Hm [20:37:32] I guess you're right about these fixmes, though, regarding the action to take [20:38:55] r52393 is ancient history. If Roan hasn't managed to fix it by now and Tim has done several deployments in the interim, I'd say lets mark it "DONT CARE" [20:39:09] or "old" [20:39:12] Filing a bug is a good idea [20:39:18] It technically is a bug [20:39:27] right, make the bugmeister bug Roan [20:39:36] haha [20:39:41] since he obviously ignores CR ;) [20:43:10] Whee, got the 1.17 merge queue down to two revs [20:43:27] One fixme and one FR rev that Chad had problems with due to merge conflicts [20:47:20] awjr: this comment was the info we discussed over lunch: https://bugzilla.wikimedia.org/show_bug.cgi?id=21860#c22 [20:48:17] robla: awesome, thanks [20:49:43] RoanKattouw: excellent! [20:50:00] Moving on to the revert1.17 queue... [20:50:03] RoanKattouw: what are the two revs? [20:50:07] Which fortunately isn't very large [20:50:25] hexmode: http://www.mediawiki.org/wiki/Special:Code/MediaWiki/tag/1.17 [20:51:59] ah, actually that fixme is only there for the CLI installer [20:52:13] let me quickly test pg and then I'll try to fix that one. [20:53:22] *robla makes quick mod to stats rev report script to filter out revs marked "1.18" [20:56:12] done: http://www.mediawiki.org/wiki/MediaWiki_roadmap/1.17/Revision_report [21:04:45] http://www.mediawiki.org/wiki/Special:Code/MediaWiki/70608 [21:04:51] Not sure how you want to tag that one [21:05:16] The problems on non MSSQL stuff have been fixed, but the core problem remains. It's not usable, so not really worth trying to revert out [21:26:08] Is there anybody here from the foundation here? I have a request. [21:26:30] There are foundation people here, yes. Just state your request in the channel and someone will answer [21:27:22] Well, there is a request to use a bot to use google's geocode api to find some coordinates for english wikipedia articles. [21:27:44] However, to do this would be a violation of thier ToS. Plus they throttle requests to 2000 per day. [21:28:01] The discussion is here, by the way. http://en.wikipedia.org/wiki/Wikipedia:Bot_requests#Category:Articles_needing_coordinates [21:28:44] I'd like someone from the foundation to send google an email asking them if we can use the API. [21:28:59] ...Outside of the ToS. [21:29:12] Right [21:29:51] tomaszf: Could you help Tim1357 find the right person to talk to? I don't really know and I'm very tired [21:30:34] greetings Tim1357 [21:30:44] Hey tomaszf. [21:31:56] i know there was some talk about google actually releasing that exact data set for import [21:32:08] kaldari@wikimedia was discussing it with them [21:32:24] he's on vacation right now but would be a good person to ask about progress [21:32:40] Ok. [21:32:42] if you message me your email then i can get that discussion going [21:35:11] tomaszf: Thanks a lot for your help. [21:36:00] glad to help [21:46:05] robla, my fixmes don't block deployment [21:46:24] actually, i'd consider r66749 resolved [21:46:36] Platonides: could you comment in your fixme? [21:47:01] I already commented "I don't remember the original reason. ialex replaced it in r80438" [21:47:28] I assume Roan was busy with other things [21:47:49] That's usually a good assumption [21:48:14] that's not the same as saying "my fixmes don't block deployment" [21:48:39] well, "shouldn't block" :P [21:48:58] if one is fixed (but not marked as such) and the other on a extension not in wikimedia... [21:49:39] I see the response on r66749. I'll look at it in the morning, as I was the one to mark it fixme in the first place [21:50:26] Platonides: maybe mark it "new" if you're not the owner anymore [21:50:57] I have also seen some revisions marked as nondeploy, that could be appropiate for r74159 [21:51:52] it's something I should eventually fix, anyway [21:52:02] but don't let it make you nervous