[04:25:06] (03CR) 10Bene: "And why do we need css class names on repo? I think they aren't needed here." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123441 (owner: 10Bene) [04:34:53] (03CR) 10Bene: "Ok, I get your point that you want to have it as configurable as possible. However, I think the prefix is necessary in this situation to p" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [06:35:05] is it somehow possible to use the LUA debug console for pages not connected to the module? Problem: A working Wikidata module stopped working, with "preview page" you get not debug output, with the debug consolse you cannot check how the module would behave in a different article. [08:16:45] [13WikibaseQuery] 15adrianlang 04force-pushed 06htmlForm from 14bdc970c to 14ea7c7e7: 02http://git.io/eLeGRA [08:16:45] 13WikibaseQuery/06htmlForm 14ea7c7e7 15Adrian Lang: Rewrite SimpleQuery special page's search form using HTMLForm [08:19:00] [travis-ci] wmde/WikibaseQuery/htmlForm/ea7c7e7 : Adrian Lang The build was fixed. http://travis-ci.org/wmde/WikibaseQuery/builds/22426144 [08:30:29] (03PS4) 10Tobias Gritschacher: Add option in travis setup to run client or repo tests only, or both [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:30:31] (03PS2) 10Tobias Gritschacher: Have travis test variety of setups (client-only, repo-only and both) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:36:54] (03CR) 10Tobias Gritschacher: "PS2 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:37:03] (03CR) 10Tobias Gritschacher: "PS4 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:37:05] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:40:26] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:43:48] (03PS5) 10Tobias Gritschacher: Add option in travis setup to run client or repo tests only, or both [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:43:50] (03PS3) 10Tobias Gritschacher: Have travis test variety of setups (client-only, repo-only and both) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:44:05] (03CR) 10Tobias Gritschacher: "PS3 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:44:19] (03CR) 10Tobias Gritschacher: "PS5 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:45:08] (03PS2) 10Tobias Gritschacher: (bug 62644) snak validation in ChangeOpMainSnak. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [08:45:10] (03PS3) 10Tobias Gritschacher: (bug 62644) snak validation in more change ops [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [08:45:12] (03PS14) 10Tobias Gritschacher: bug 62644) move snak validation into ChangeOps [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [08:45:27] (03CR) 10Tobias Gritschacher: "PS3 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [08:45:35] (03CR) 10Tobias Gritschacher: "PS2 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [08:45:46] (03CR) 10Tobias Gritschacher: "PS14 is a rebase" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [08:49:42] [13ValueView] 15adrianlang 04deleted 06enterkey at 14855e265: 02http://git.io/wv1kMQ [08:50:40] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [08:51:54] (03CR) 10Tobias Gritschacher: [C: 04-1] "I'm not able to add constraints. The loading-spinner in the property input field spins endless but no suggestions are shown." [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123563 (owner: 10Henning Snater) [08:53:40] [13WikibaseQuery] 15adrianlang created 06snakview (+1 new commit): 02http://git.io/4SzNEA [08:53:40] 13WikibaseQuery/06snakview 1458d4200 15Adrian Lang: Use Wikibase's snakview for the SimpleQuery special page [08:54:58] [13WikibaseQuery] 15adrianlang 04force-pushed 06snakview from 1458d4200 to 146a20a9c: 02http://git.io/29CD-Q [08:54:58] 13WikibaseQuery/06snakview 146a20a9c 15Adrian Lang: Use Wikibase's snakview for the SimpleQuery special page [08:56:01] [travis-ci] wmde/WikibaseQuery/snakview/58d4200 : Adrian Lang The build has errored. http://travis-ci.org/wmde/WikibaseQuery/builds/22428029 [08:57:11] [travis-ci] wmde/WikibaseQuery/snakview/6a20a9c : Adrian Lang The build passed. http://travis-ci.org/wmde/WikibaseQuery/builds/22428137 [08:57:26] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [08:57:55] [13WikibaseQuery] 15adrianlang opened pull request #13: Use Wikibase's snakview for the SimpleQuery special page (06master...06snakview) 02http://git.io/oIBG8A [09:01:25] [13ValueView] 15tobijat created 06updatevers (+1 new commit): 02http://git.io/fR0Pvw [09:01:25] 13ValueView/06updatevers 14c331e55 15Tobias Gritschacher: Update Version and Readme [09:02:27] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [09:05:42] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [09:05:55] [13ValueView] 15addshore closed pull request #46: Update Version and Readme (06master...06updatevers) 02http://git.io/ql4Ghg [09:08:05] [13WikibaseDatabase] 15addshore comment on commit 14713b20e: But is it valid syntax to have PRIMARY KEY in sql?... 02http://git.io/G8jhzQ [09:09:02] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [09:14:45] (03PS1) 10WikidataBuilder: New Wikidata Build - 07/04/2014 09:14 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124295 [09:15:31] new wikidata build at 9:14? :O [09:15:32] hehe [09:15:46] addshore: I just manually ran the script [09:15:51] :P [09:16:00] it did not do automatically [09:16:17] it still has anoth45 45 mins till it would run on the cron :P [09:16:22] 845 mins [09:16:30] oh my whats wrong with my keyboard... [09:16:31] what? [09:16:33] 45 [09:16:34] why? [09:16:41] oh, time changed [09:16:42] it doesnt run until 10:00 localtime [09:16:45] :P [09:16:55] ok [09:16:59] do we want that changed? ;p [09:17:05] hm. [09:17:09] (03PS6) 10Aude: Add option in travis setup to run client or repo tests only, or both [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 [09:17:10] don't know [09:17:11] (03PS4) 10Aude: Have travis test variety of setups (client-only, repo-only and both) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 [09:18:52] probably runs an hour off because of summer time [09:19:35] wooo! summertime! [09:25:33] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [09:27:07] (03PS3) 10Adrian Lang: (bug 57693) supply id for claim group [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121361 (owner: 10Daniel Kinzler) [09:29:57] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124295 (owner: 10WikidataBuilder) [09:31:31] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [09:34:52] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121361 (owner: 10Daniel Kinzler) [09:35:35] (03CR) 10Adrian Lang: "Added the necessary code in the frontend, removed duplicate space, rebased. I'm still not that happy with having such a generic DOM ID cla" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121361 (owner: 10Daniel Kinzler) [09:37:36] (03CR) 10Adrian Lang: [C: 032] Clean up Client Api Module [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121705 (owner: 10Addshore) [09:38:02] (03Merged) 10jenkins-bot: Clean up Client Api Module [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121705 (owner: 10Addshore) [09:44:45] DanielK_WMDE: we're in the hangout if you want to join [09:49:34] :O [09:56:43] DanielK_WMDE: In case you didn't joing the hangout... got a few moments? [09:59:44] Lydia_WMDE: Already saw my "In conclusion it looks like everything works as it should except of our beloved AbuseFilter." ? ;) [10:00:23] (03PS1) 10WikidataBuilder: New Wikidata Build - 07/04/2014 10:00 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124300 [10:00:44] hehe Tobi_WMDE ^^ ;p [10:06:43] (03Abandoned) 10Addshore: New Wikidata Build - 07/04/2014 09:14 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124295 (owner: 10WikidataBuilder) [10:08:34] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124300 (owner: 10WikidataBuilder) [10:10:01] DanielK_WMDE: JeroenDeDauw I guess we should decide on how to store time before I start writing stuff then [10:10:08] hmpf. my box just crashed. hard. third time in a week. hope it's just a flaky kernel, not the hardware :/ [10:10:51] addshore: I prefer int, but the question is really how hard it is to get that right [10:11:02] to get this *exactly* right is pretty hard [10:11:07] I guess we have three options, 1). seconds teh Epoche, 2). seconds from 0 i.e. 1/1/1 00:00:00 3). some longer string format, maybe just whats in time value now? [10:11:24] addshore: yes, we definitly should decide that [10:11:39] 1 and 2 are eqally hard to calculate, and i don't see the advantage of 2 over 1. [10:12:00] I will be happy when the performance is reasonable and when we do not end up writing a ton of code for this - implementing things with leap seconds and whatnot seems very suspicious to me [10:12:03] I would say 2 would probably be more consistent with itself [10:12:29] imho, the int index value doesn't have to be *exactly* right beyond the range of -9999 and +9999. Overing -9999 to -1 already needs some hacks in PHP it seems. [10:13:02] addshore: with itself, yea, but with nothing else. epüoch is the standard, option 2 opffers no computational advantages over it. [10:13:20] it'S just an arbitrary offset [10:13:37] so we go for 1? similar implementation to whats in the pull request now just tidier and tested and in TImeValue [10:14:03] havn't looked at the pull request [10:14:34] i'd suggest to use built in libs for -9999 to +9999, and beyond that, just add the year multiplied by some constant, and screw the leap years. [10:14:56] i don't think days or e3ven years are significant beyond that range. [10:15:05] okay, I'll go do that :) [10:15:05] at least not for indexing/searching [10:15:40] JeroenDeDauw: what do you think? SMW is using a padded string representation, right? [10:15:47] how many digits does it support for the years? [10:16:16] actually, it's not hard to support alternative DV handlers, is it? BAsically, how Time is indexed could be a config option [10:17:23] addshore: please double check and triple test that the build in libs work right in the BC range. quick testing implies that they don't. [10:18:04] hahaha, omg. it'S worse than i thought. [10:18:08] addshore: https://stackoverflow.com/questions/7198539/php-strtotime-returning-false-for-dates-less-than-1900 [10:19:02] hm, gregoriantojd? [10:19:13] bah, this sucks :) calendars are hard. [10:19:34] i think the only half way decent calendar support that i have seen yet was in java. and that sucked too. [10:20:36] DanielK_WMDE: I hate time :P [10:22:09] DanielK_WMDE: using build in libs for the "normal" range and being less accurate outside sounds reasonable to me [10:22:12] "The valid range of a timestamp is typically from Fri, 13 Dec 1901 20:45:54 UTC to Tue, 19 Jan 2038 03:14:07 UTC." [10:22:12] idk what SMW uses [10:22:16] http://www.php.net/manual/en/function.strtotime.php [10:22:24] >_< [10:22:34] thats a nice small range :P [10:22:47] And yes, its easy nto have multiple DVHandler implementations, though we only need one, so writing and maintaing multiple does not make much sense [10:22:51] " dates prior to Jan 1, 1970 will not work on Windows, some Linux distributions, and a few other operating systems" [10:22:54] * DanielK_WMDE dies [10:23:05] what is this, the 80s? [10:23:25] * JeroenDeDauw records the time on which DanielK_WMDE died, with some random precision [10:23:47] DanielK_WMDE: you do realize there is no such thing as common time to begin with right? [10:24:17] Lydia_WMDE: it seems like we have to hold general releativity and quantum mechanics into account for our software [10:24:25] :/ [10:24:37] And come up with the theory of everything to not be wrong when it comes to black holes and whatnot [10:26:42] JeroenDeDauw: and we need to make sure we get the direction right. time is defined by causality, and causality in physics works both ways - except, pretty much, for entropy. [10:27:05] so, entropy defines time. and entropy is defined using time. can we go home now? [10:27:32] ahh DanielK_WMDE my theroy about 2 being eaiser than 1, when calulcating for - years the process can then be exactly the same and just flip the value to a minus [10:27:51] when actually using number of seconds reletive to epoch this cant exactly be done [10:28:15] addshore: true. but to make this work with the standard epoch, all you need to do is add a constant. [10:28:40] indeed. [10:28:43] but frankly, i don't care much. the numbers in the database are arbitrary. we just need then to compare correctly. [10:28:48] :P [10:29:11] well, I guess I'll do a calculation based to 0 then and add the epoch ;p [10:29:28] whatever floats your rubber ducky. [10:29:38] my rubber ducky has holes in it :/ [10:29:49] ...too much information... [10:29:57] xD [10:30:11] addshore: no. :p [10:31:19] (03CR) 10Thiemo Mättig (WMDE): "Wait, I'm afraid I got confused in my first review. This is not about the ItemId, this is about all the PropertyIds. This will work for no" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121361 (owner: 10Daniel Kinzler) [10:31:52] addshore: actually, "my rubber ducky has holes in it" makes for a good bugzilla quip [10:32:49] xD [10:32:57] I'll let you add that ;p [10:34:37] hah [10:35:13] DanielK_WMDE: around? [10:35:44] review: https://github.com/DataValues/Common/pull/7 ! [10:35:58] it's rotting [10:36:13] aude: Any news on https://gerrit.wikimedia.org/r/120535 btw? [10:36:51] i am reviewing https://gerrit.wikimedia.org/r/#/c/122124/ [10:37:04] \o/ [10:38:59] hoo: back in 10 minutes or so. just write [10:39:14] hoo: for rebuild entity per page, i think DanielK_WMDE and adrian found issues with saves happening and then something going wrong and secondary tables not updated [10:39:26] [13Time] 15addshore pushed 1 new commit to 06master: 02http://git.io/eFUVdw [10:39:26] 13Time/06master 14c60f292 15addshore: Fix duplicate array keys in tests [10:39:42] aude: We don't want to update the secondary tables over there, only entity per page [10:39:46] right [10:39:52] but it's symptom of a problem or bug [10:39:59] DanielK_WMDE: Going to be away then :P Let's talk later on... [10:40:21] aude: Maybe... but last time I looked there were no recent cases, only problems from up to mid 2013 [10:40:25] i don't see issue with running the script but would be good to find a way to handle the source of the prolbem [10:40:28] ok [10:40:37] Lydia_WMDE: The channel topic is just OCD me again >.> [10:40:43] addshore: i have a really, really hard time following irc discussions. what's the purpose of that "epoche" thing? what's the base? which definition are we using? [10:40:45] stopped in like May 2013 AFAIR [10:40:53] when we run the script, what does the log say? [10:41:06] Thiemo_WMDE: I am sure it will make sense in my next patchset to DV/Time [10:41:17] aude: Like if we ran it now? [10:41:22] does it make a log of entities "rebuilt " in the table? [10:41:24] Tobi_WMDE: aude: JeroenDeDauw: might have found a way to get composer installed on the Wikimedia Jenkins slaves https://gerrit.wikimedia.org/r/#/c/124304/ :D [10:41:27] aude: Yep [10:41:31] hashar: \o/ [10:41:34] hashar: :D [10:41:38] last time it matched exactly the number of broken entities [10:41:45] basically: installs composer using .... composer! [10:41:59] as it essentially is using the same query I made up for checking inconsistencies [10:42:08] hoo: ok, so we can look at it and see if there continue to be missing entities [10:42:21] addshore: is there a document somewhere where i can read about what this is supposed to mean? [10:42:27] aude: We should make it log on terbium somewhere or so [10:42:46] looks like your puppet patch does that [10:42:46] don't think that's much of an issue, the rebuild items per site will be more awry, I guess [10:42:51] Thiemo_WMDE: dont think so, as we have only just decided what it is going to be ;p [10:43:00] runs onces a month? [10:43:06] Yep [10:43:11] If I didn't screw [10:43:13] so what was decided? i'm afraid i don't get it. [10:43:14] i think it can be more often [10:43:22] Sure, also possible [10:43:25] if there are broken entities, a month is a long time to wait [10:43:28] like weekly or daily or whatever [10:43:28] hashar: that's good news! [10:43:46] the initial query took maybe 2 minutes, but that's ok for a one-off thing [10:43:47] not like it's a demanding script [10:43:58] maybe once a week or something is better [10:44:00] on labs, probably faster in prod. [10:44:01] (03PS3) 10Adrian Lang: Apply claimview instead of statementview to property page output [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123562 (owner: 10Henning Snater) [10:44:06] Got to leave now... cu later [10:44:09] ok [10:45:35] (03CR) 10Tobias Gritschacher: [C: 032] New Wikidata Build - 07/04/2014 10:00 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124300 (owner: 10WikidataBuilder) [10:45:38] addshore: what's the use of this "epoche" thing at all? i thought we are using strings like '+1970-01-01T00:00:00Z' to store time stamps? why do we introduce a second format for the exact same thing? [10:46:19] Thiemo_WMDE: https://www.mediawiki.org/wiki/Extension:Wikibase_QueryEngine/SQL_schema_proposal [10:46:48] Thiemo_WMDE: because we would need to pad the year to 20 digits or so to get proper sorting. thats makes for a big and slow index. we'll have millions of dates... [10:47:11] and now it is available at /srv/deployment/integration/composer/vendor/bin/composer [10:47:16] doing that is an option, but not an option i like. [10:47:19] what? databases do have datetime fields, you know? [10:47:30] hashar: :D so.. osoooo... we can run our tests again maybe? ;p [10:47:39] yeah that would be nice :-] [10:47:47] Thiemo_WMDE: that actually work correctly over billions of years? [10:47:50] * addshore grins [10:47:54] hashar: would be wonderful [10:47:56] is this something like the scary string "hashes" that aren't hashes? [10:48:11] it's a terrible waste of time to maintain separate jenkins [10:48:15] Thiemo_WMDE: kind of, yea :P [10:48:16] indeed [10:48:23] * aude rather any time we invest in CI is to help wmf jenkins improve [10:48:26] but you have a point. let me read up on datetime support in mysql [10:48:32] (03Merged) 10jenkins-bot: New Wikidata Build - 07/04/2014 10:00 [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124300 (owner: 10WikidataBuilder) [10:48:44] mysql supports up to 9999, i'm afraid. [10:48:59] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123562 (owner: 10Henning Snater) [10:49:01] aude: Tobi_WMDE as all our stuff is in jenkins job builder with a few small tweaks we should be able to get it running stright away ;p [10:49:11] Thiemo_WMDE: there is our problem :P [10:49:31] Thiemo_WMDE: "The supported range is '1000-01-01' to '9999-12-31'. " [10:49:33] yay. not. [10:50:07] " TIMESTAMP has a range of '1970-01-01 00:00:01' UTC to '2038-01-19 03:14:07' UTC. " [10:50:34] TIMESTAMP is updating itself, you should never use it except thats what you want. [10:50:54] Thiemo_WMDE: yea, just saying that DATE and TIMESTAMPS support different ranges, more fun [10:51:48] does not make the hacks we are introducing in a core database scheme less scary. [10:51:52] DanielK_WMDE: what about DATE etc. in sqlite? postgresql? [10:52:00] if we want to make this cross-database compatible :) [10:52:16] aude: and in elastic. yea. fun. [10:52:19] better make our own. [10:52:20] yeah! [10:52:29] i'm fine with a seconds based timestamp if it does have a clear definition. [10:52:35] i don't think mysql people know what they are doing when it comes to time [10:52:36] ... and is tested. [10:52:43] only for basics [10:53:08] i trust mysql more than a "year * secondsperyear" calculation. [10:53:19] Thiemo_WMDE: remember that this is just and index for queries. we always store the "actual" date in the json. [10:53:45] if the queries return wrong or no results it's as bad as storing wrong data. [10:53:57] Thiemo_WMDE: but mysql doesn't support the range we need. it doesn't suzpport BC at all, for one thing [10:54:09] addshore: "straight away" is a bit too optimistic I fear [10:54:20] Tobi_WMDE: within a week or 2 ;p [10:54:30] yeah that's more accurate [10:54:41] Thiemo_WMDE: "wrong" is pretty hard to define for "far away" dates, since the notion of a year is not well defined. [10:55:07] it currently is wrong for dates in 2014 [10:55:59] i still don't have an answer for the basic question what the base of that epoche thing is? [10:56:18] unix timestamp's 1970? i'm totally fine with that. [10:56:45] yes, "epoch" is defined as the unix timestamp [10:57:03] ok. thanks. [10:57:13] https://en.wikipedia.org/wiki/Unix_epoch [10:57:34] which, of course, is "wrong", since POSIX doesn't account for leap seconds at all, afaik [10:57:56] but it's good enough for our purpose, even if it's off by days for "far off" dates [10:58:11] it's what people expect from a php based system. [10:58:18] :P [10:58:28] "being wrong"?... [10:58:33] xD [10:59:19] sorry, never heard of a problem with leap seconds and unix timestamps. [10:59:54] from wikipedia: [10:59:57] When a leap second occurs, so that the UTC day is not exactly 86400 seconds long, a discontinuity occurs in the Unix time number. The Unix time number increases by exactly 86400 each day, regardless of how long the day is. [11:00:04] [unix timestamps are] what people expect from a php based system. [11:00:33] it's an internal index value, not part of the public interface [11:00:39] people shouldn't know or care what we do there [11:00:50] even better. [11:00:51] ;) [11:01:49] is this needed to do range queries? [11:02:05] ... oh, it is. found it. [11:02:05] yes [11:02:18] well, all our queries are range queries, really [11:02:44] ...for time, geo and quantity, at least [11:02:44] even strings? [11:02:58] no, not for strings, ids and such [11:03:03] for stuff that has a precision [11:05:56] Lydia_WMDE: https://bugzilla.wikimedia.org/show_bug.cgi?id=45277 [11:06:40] Lydia_WMDE: the original description isn't quite what I have in mind right now. i'll add a comment. want a new bug? [11:06:54] comment and description change is fine :) [11:08:46] [13WikibaseDatabase] 15thiemowmde comment on commit 14713b20e: > But is it valid syntax to have `PRIMARY KEY` in sql?... 02http://git.io/_aNMBw [11:09:17] hoo|away: hehe yeah i did - sounds good :) [11:11:21] [13WikibaseDatabase] 15addshore comment on commit 14713b20e: space matters in lots of things, hence why I said... 02http://git.io/HnEs5w [11:12:29] (03CR) 10Aude: [C: 032] "looks good and works ok" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/122124 (owner: 10Hoo man) [11:12:42] Lydia_WMDE: commented. the headline is still fine. [11:12:45] (03Merged) 10jenkins-bot: Fix edit restriction handling [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/122124 (owner: 10Hoo man) [11:12:50] cool [11:12:51] thx [11:12:52] [13WikibaseDatabase] 15JeroenDeDauw comment on pull request #36 144603061: testGivenInvalidTypeName_constructorThrowsException... 02http://git.io/LVj_Lw [11:17:04] right, ill be back in a bit [11:20:30] [13WikibaseQuery] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/mHwz1A [11:20:30] 13WikibaseQuery/06master 1400742bf 15Jeroen De Dauw: Update .travis.yml [11:28:47] (03PS1) 10Aude: Fix issue of blocked user seeing page protection instead of block message [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124308 [11:29:13] is it somehow possible to use the LUA debug console for pages not connected to the module? Problem: A formerly working Wikidata-based module shows script errors now. With "preview page with this template" you get no debug output while with the debug console you cannot check how the module would behave in a different article. [11:29:35] reimann: ask hoo|away when he is back [11:29:42] thx [11:33:37] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124308 (owner: 10Aude) [11:34:15] (03PS1) 10Aude: Use SimpleSiteLink (for now) in UpdateRepo [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 [11:34:54] (03PS1) 10Aude: Fix edit restriction handling [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 [11:35:25] (03PS1) 10Aude: Set font of the wb-firstHeading toolbar back to sans-serif [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 [11:37:06] [13WikibaseQuery] 15JeroenDeDauw comment on pull request #8 14ea7c7e7: This is not a mock - it does not do any assertions. It actually is not any type of test double at all, since you return a real instance.... 02http://git.io/TpFSWw [11:38:14] aude: does jenkins still fail for client tests? [11:39:44] addshore: bypassing review now are we? good good, your journey to the dark side has now completed *evil laugh* [11:40:35] Adrian_WMDE: please see https://bugzilla.wikimedia.org/show_bug.cgi?id=63477 [11:41:21] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 (owner: 10Aude) [11:42:21] JeroenDeDauw: I've seen it, thanks. I/we/you should talk with lydia about that [11:45:57] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 (owner: 10Aude) [11:47:17] [travis-ci] wmde/WikibaseQuery/EntitiesByPropertyValue/0230562 : jeroendedauw The build passed. http://travis-ci.org/wmde/WikibaseQuery/builds/22437268 [11:49:17] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 (owner: 10Aude) [11:58:12] addshore: what's the deal with DM Serialization? [12:01:40] DanielK_WMDE: I am looking at OPT_ID_KEY_MODE which you introduced [12:01:47] Not sure we actually use this? [12:03:20] DanielK_WMDE: nvm [12:05:04] DanielK_WMDE: not supposed to [12:05:19] * aude looks at why they fail on jenkins for the branch [12:05:47] oh, that [12:06:29] (03PS1) 10Aude: Move WikibaseExceptionLocalizer messages to lib [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124314 [12:06:55] (03CR) 10WikidataJenkins: "Build Failed" [extensions/Wikidata] - 10https://gerrit.wikimedia.org/r/124300 (owner: 10WikidataBuilder) [12:11:04] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124314 (owner: 10Aude) [12:20:25] (03PS1) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 [12:21:24] (03PS2) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 [12:25:48] (03CR) 10Aude: [C: 032 V: 032] Move WikibaseExceptionLocalizer messages to lib [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124314 (owner: 10Aude) [12:26:34] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [12:26:48] orrr [12:27:15] is there a test wiki left which is connect to (live) Wikidata? Former test2 was a the only showcase we had so far... [12:27:26] reimann: no [12:27:29] :( [12:27:35] test2 is connected to test.wikidata now [12:27:36] thx [12:28:06] having test2 connected to wikidata has some undesired side effects [12:28:25] s/has/had/ [12:29:25] (03CR) 10Hoo man: "Jenkins failure looks unrelated... caused by core? We should investigate..." [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 (owner: 10Aude) [12:29:58] hoo|away: we just need to put patches in the correct order [12:30:06] fixed on master [12:30:07] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [12:31:01] (03PS3) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 [12:32:01] wo ist jenkins? :) [12:32:34] Lydia_WMDE: please see https://bugzilla.wikimedia.org/show_bug.cgi?id=63477 [12:32:54] test2 was the only place where you could show how a Wikidata backed infobox works [12:33:50] (03PS4) 10Daniel Kinzler: (bug 62644) snak validation in more change ops [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 [12:34:09] (03PS4) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 [12:34:32] i suppose have a copy of the properties on test.wikidata needed for an infobox? [12:35:18] Lydia_WMDE: apropos what reimann just said: we should really try to showcase a wikidata backed infobox on wikipedia at the hackathon. [12:35:36] perhaps ash hoo what needs to be done so we can do that [12:36:32] :) to cross-check all the corner cases of a data model (here: biological taxonomy) it is no fun to copy the data (and properties with corresponding numbers) to test.wikidata. Our bots are already running at full speed... [12:36:39] https://en.wikipedia.org/wiki/Module:Authority_control [12:36:49] https://en.wikipedia.org/wiki/Module:Infobox_road/map (and related ones) [12:37:07] reimann: hmmm [12:37:18] https://www.wikidata.org/wiki/Module:Taxobox [12:37:33] https://test2.wikipedia.org/wiki/Module:Taxobox [12:37:47] (1st one is a javascript-backed hack) [12:37:48] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [12:40:20] reimann: looking at test.wikipedia.org (instead of test2), it might be possible to connect that to wikidata [12:41:20] https://bugzilla.wikimedia.org/show_bug.cgi?id=61893 was the issue (test2 only has the javascript testing thing) [12:42:07] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [12:44:17] aude: k, this is of course not nice. but it is also increasingly complex to discuss details of models without test/showing how this would look like. [12:45:14] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [12:45:19] reimann: https://bugzilla.wikimedia.org/show_bug.cgi?id=63619 [12:45:45] aude: thx a lot [12:45:47] if no one sees an issue with that, i think we can do that [12:47:06] it might not have change dispatching, which means you need to purge a page on test.wikipedia [12:47:19] or maybe we can do change dispatching [12:50:42] would be not problem [12:50:42] (03PS15) 10Daniel Kinzler: bug 62644) move snak validation into ChangeOps [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 [12:50:58] (03PS3) 10Daniel Kinzler: (bug 62644) snak validation in ChangeOpMainSnak. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 [12:51:00] no [12:52:39] (03PS4) 10Daniel Kinzler: (bug 62644) snak validation in ChangeOpMainSnak. [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 [12:52:50] (03PS5) 10Daniel Kinzler: (bug 62644) snak validation in more change ops [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 [12:54:21] (03PS6) 10Daniel Kinzler: (bug 62644) snak validation in more change ops [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 [12:54:34] (03PS5) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 [12:57:28] DanielK_WMDE: advertisement: the WD taxobox is already able to show different hierarchies and the corresponding references (all autogenerated). See also:https://www.wikidata.org/wiki/User:FelixReimann/test :-) [12:57:32] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [12:58:16] reimann: https://gerrit.wikimedia.org/r/#/c/124321/ [12:58:24] hopefully that gets approved [12:58:39] great! thanks a lot. [12:58:48] happy to find a solution! [12:59:57] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [13:06:16] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [13:07:31] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [13:11:06] Where are the restrictions for label, alias and description length? [13:11:11] DanielK_WMDE: addshore aude ^ [13:11:32] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [13:12:10] multilang-limits setting in repo [13:14:42] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [13:15:16] why did jenkins start gate-and-submit on https://gerrit.wikimedia.org/r/#/c/124314/ almost hour ago and then not merge it? [13:16:05] (03PS2) 10Aude: Use SimpleSiteLink (for now) in UpdateRepo [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 [13:19:19] JeroenDeDauw: MultiLangConstraintDetector and LabelDescriptionDuplicateDetector. See PreSaveChecks as an entry point. [13:19:51] JeroenDeDauw: why, you want to work on #62643? [13:20:06] there's already a draft, i'll add you [13:20:11] was going to hack at it myself just now [13:20:45] DanielK_WMDE: please link [13:20:50] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 (owner: 10Aude) [13:20:52] https://gerrit.wikimedia.org/r/#/c/62643 [13:20:55] That's not WB [13:21:09] DanielK_WMDE: anyway, no [13:21:39] !bug 62643 [13:22:25] DanielK_WMDE: did you have a better suggestion for the name of SomeProperty? https://github.com/wmde/Ask/tree/master/src/Language/Description [13:22:34] (03CR) 10Aude: [C: 032] Add missing wikibase.utilities dependency to snakview [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123631 (owner: 10Adrian Lang) [13:22:35] I'm not entirely happy with that name [13:22:48] JeroenDeDauw: "ValueExists"? [13:22:51] (03Merged) 10jenkins-bot: Add missing wikibase.utilities dependency to snakview [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123631 (owner: 10Adrian Lang) [13:22:57] that's basically the assertion [13:24:10] JeroenDeDauw: that was a bug id, not a patch id :) [13:24:47] DanielK_WMDE: that name is worse than the current one... [13:24:47] https://gerrit.wikimedia.org/r/#/c/121422/ [13:24:54] https://bugzilla.wikimedia.org/show_bug.cgi?id=62643 [13:25:11] JeroenDeDauw: i don't like it much either. can't think of anything else [13:25:24] anyway, coffee break [13:25:27] (03CR) 10Aude: [C: 032] Reorder snakview dependencies [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123632 (owner: 10Adrian Lang) [13:25:56] (03Merged) 10jenkins-bot: Reorder snakview dependencies [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123632 (owner: 10Adrian Lang) [13:25:56] PropertyDescription is better, though perhaps a bit deceptive [13:26:04] PropertyValueDescription is less deceptive [13:26:31] Though the "value" in there is not the same as in some other "PropertyValueStuff" we have [13:34:34] reimann: you had a lua question earlier? [13:37:03] hoo: hoo: yes, thanks for asking. is it somehow possible to use the LUA debug console for pages not connected to the module? I would need something like "Context title" from Special:ExpandTemplates to test the module in combination with the data retrieved from WD [13:37:21] Not possible at the moment, not [13:37:24] * no [13:37:31] ok. thx [13:37:47] (03CR) 10Aude: [C: 032] Altered property page layout [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123587 (owner: 10Henning Snater) [13:37:50] That would require a Scribunto patch... you might want to look in bugzilla for a bug asking for such a feature [13:38:06] (03Merged) 10jenkins-bot: Altered property page layout [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123587 (owner: 10Henning Snater) [13:38:11] ok. i'll create one. [13:38:20] (03CR) 10Aude: [C: 032] Use SimpleSiteLink (for now) in UpdateRepo [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 (owner: 10Aude) [13:38:44] (03Merged) 10jenkins-bot: Use SimpleSiteLink (for now) in UpdateRepo [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124309 (owner: 10Aude) [13:39:02] (03PS2) 10Aude: Set font of the wb-firstHeading toolbar back to sans-serif [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 [13:39:55] DanielK_WMDE: JeroenDeDauw can you please review https://github.com/DataValues/Common/pull/7 ? [13:41:09] [13WikibaseDatabase] 15thiemowmde comment on commit 14713b20e: "SQLite shall recognize a sequence of one or more WHITESPACE characters as a WHITESPACE token." https://www.sqlite.org/hlr40000.html 02http://git.io/pD3cIQ [13:41:42] (03PS2) 10Aude: Fix edit restriction handling [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 [13:45:00] [13Common] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/9jeUHg [13:45:00] 13Common/06master 1471b756c 15Jeroen De Dauw: Merge pull request #7 from DataValues/mismatch... [13:45:07] thanks :) [13:45:10] [13Common] 15JeroenDeDauw 04deleted 06mismatch at 14315bf6e: 02http://git.io/iV6vkQ [13:45:46] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 (owner: 10Aude) [13:46:07] (03CR) 10Aude: [C: 032] Set font of the wb-firstHeading toolbar back to sans-serif [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 (owner: 10Aude) [13:46:31] (03Merged) 10jenkins-bot: Set font of the wb-firstHeading toolbar back to sans-serif [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124311 (owner: 10Aude) [13:47:37] (03CR) 10Thiemo Mättig (WMDE): "> why do we need css class names on repo?" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123441 (owner: 10Bene) [13:48:59] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 (owner: 10Aude) [13:49:16] (03CR) 10Aude: [C: 032] Fix edit restriction handling [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 (owner: 10Aude) [13:49:33] (03Merged) 10jenkins-bot: Fix edit restriction handling [extensions/Wikibase] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124310 (owner: 10Aude) [13:52:06] oh, why do i get a fatal error in central auth :( [13:54:02] oO [13:54:56] why does CentralAuthHooks have over 1500 lines of code! [13:56:17] because it's more fun that way, right? [13:56:32] (03CR) 10Thiemo Mättig (WMDE): "> the prefix is necessary in this situation to prevent unwanted collisions" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [13:56:51] Saddest part is that much of the current mess was only added recently when we switched over towards having a login wiki :( [13:57:09] hoo: did you get the email from Kai? (and probably Abraham) regarding your tool on labs? [13:57:25] if kai should work on it he needs an answer soon [13:57:28] Tobi_WMDE: (More or lesS) just replied [13:57:39] hoo: awesome [13:57:44] Not sure what questions are open regarding to it :P [13:57:50] He can also ping me on IRC if that's better [13:58:02] or something like that [13:58:07] * hoo is flexible [13:58:37] hoo: IIRC he asked where to find the sourcecode and the documentation [13:59:08] I don't have any documentation and not really something you could consider source code :P [13:59:21] maybe DanielK_WMDE did that, back when he did the DBQs? [13:59:46] hoo: ok, then this is probably what you can write him as an answer [14:01:08] ok, will do that [14:01:20] thx [14:02:33] hoo: https://bugzilla.wikimedia.org/show_bug.cgi?id=63623 [14:03:19] reimann: Great... thanks :) [14:03:31] A feature I would like to have had while testing as well [14:03:49] [13WikibaseDatabase] 15addshore comment on commit 14713b20e: Lovely, then it just needs a few tests to make sure nothing else breaks with the changes to what can be matched 02http://git.io/4GZpRA [14:03:51] hm? whut? [14:03:54] hoo: also https://gerrit.wikimedia.org/r/#/c/124321/ [14:04:04] DanielK_WMDE: Database queries project on Toolserver [14:04:08] unless you think a reason not to do that [14:04:19] oh that [14:04:22] there is no source code [14:04:23] I took that over from you(?) in late 2010(?) :P [14:04:32] it was just a way to request queries via jira [14:04:42] database queries? [14:04:47] aude: Yeah, sounds sane... but should probably be added to the dblist? [14:04:53] basically, a place to open tickets of the form "could you run this query real quick..." [14:04:57] aude: yes [14:05:00] on-off stuff [14:05:03] hoo: no, list it separate [14:05:07] maybe [14:05:15] we thought about writing software support for it, but in the end, it was all manual via jira [14:05:17] DanielK_WMDE: https://github.com/mzmcbride/database-reports [14:05:19] and never really toook off [14:05:33] don't know if that fits the use case [14:05:42] Yeah... I was doing almost all of it until I didn't have the time anymore [14:05:59] some of the questions resulted in sheer insane queries :P [14:06:16] glor-ia might like help with the database reports [14:06:41] We also need help in that area [14:07:02] I don't think there are many people around who like to fiddle with such big, messy and hard to understand queries [14:07:19] except of weird people like /me [14:07:29] heh [14:11:59] [13WikibaseQuery] 15addshore pushed 1 new commit to 06master: 02http://git.io/XF4abQ [14:11:59] 13WikibaseQuery/06master 1456e001a 15addshore: Merge pull request #14 from wmde/EntitiesByPropertyValue... [14:12:11] [13WikibaseQuery] 15addshore 04deleted 06EntitiesByPropertyValue at 140230562: 02http://git.io/GLO-fA [14:12:52] aude: Anyway, why should it not be on the dblist if it takes data from the real wikidata? Because it uses a "mocked" siteid? [14:13:27] hi [14:15:27] hoo: it gets change dispatching if on the list, though shouldn't be a problem i suppose [14:15:45] then we can have a place to test it [14:16:08] as long as it doesn't propagate to wikidata that should be fine [14:16:21] Thiemo_WMDE, aude, JeroenDeDauw: I can has review? https://gerrit.wikimedia.org/r/#/c/121689/15 and https://gerrit.wikimedia.org/r/#/c/123860/ and the rest of the chain... [14:16:23] my patch configures it not to do that [14:16:35] Lydia_WMDE: https://bugzilla.wikimedia.org/show_bug.cgi?id=63255 close and open a bug to fix AbuseFilter, maybe? [14:16:48] or do you want to keep it as a tracking bug? [14:17:07] aude: Then that's fine... so I don't really see the problem [14:17:25] DanielK_WMDE: how finished is this? should we +2 this even if we find nitpicks? or do you want us to -1 because of nitpicks? [14:17:47] (03CR) 10Aude: bug 62644) move snak validation into ChangeOps (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [14:18:56] * aude is nitpicking [14:20:28] real issue is that daniel's patch makes it challenging for me to insert rubbish snaks so i can work on display of formatting errors [14:20:45] :p [14:20:50] heh :) [14:21:30] DanielK_WMDE: https://gerrit.wikimedia.org/r/#/c/120857/2/repo/includes/WikibaseRepo.php Ok, so you answered the why, but not the how [14:23:13] use a Setting? [14:23:15] aude: ^ [14:23:58] Feels kind of like our $GLOBALS['evilDataValueMap'] [14:24:09] (03CR) 10Aude: "nitpicks" (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [14:24:17] aude: yes and no - validation in the api is getting stricter, but i'm removing validation from the "save the entity" code path. [14:24:35] so injecting stuff directly should become easier. maybe make a little maintenance script for that? [14:25:05] it'S now also fesible to introduce a flag for the API to skip validation. if enabled in the config. [14:25:09] (03CR) 10Aude: bug 62644) move snak validation into ChangeOps (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [14:25:15] DanielK_WMDE: nice [14:25:43] Line 9: NOTE: This uncovers teh use of invalid data in test case. [14:25:44] teh? [14:25:52] :D [14:26:05] all teh dataz! [14:26:25] hoo: from Settings? or make a hookpoint here, so stuff buiding on top of wikibase can add types that way. [14:26:59] hoo: actually, this should not just be a list, but a mapping of item-type <-> content-model [14:27:00] DanielK_WMDE: Like wfRunhook( 'WikibaseContentModels', $foo ); [14:27:07] err, entity type [14:27:13] right [14:27:18] yea, like that [14:27:44] ok, I think I prefer the hook over the setting as I don't think this is a "setting" [14:27:48] i don't reaqlly like hooks for this, but it makes slightly more sense than a config var. it's not configurable. it's extensible. [14:28:09] hoo: exactly :P [14:28:45] what we really want is a registry. but we want stuff to get registered early on during initialization. so we could use a global array. ugh. [14:28:54] leaves... a hook. [14:29:25] registry gives me windows-shiver :P [14:29:27] hoo: if you do this, hunt for places where with use lists of entity ids and/or content models in the code, there is a handful of such places [14:29:39] Yeah, there are a few [14:29:40] make them use WikibaseRepo::bla [14:30:22] oh, and don't forget to make that array( &$foo ) [14:30:51] :) [14:30:57] Needs a rebase first... [14:32:19] (03PS3) 10Hoo man: Check the content model of pages in EntityPerPageBuilder [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 [14:32:27] so many places in central auth, it does "WikiMap::getWiki( wfWikiID() ); " [14:32:28] [13jenkins-job-builder-config] 15tobijat created 06jbbtest (+1 new commit): 02http://git.io/hPPa0A [14:32:28] 13jenkins-job-builder-config/06jbbtest 14ad5999a 15Tobias Gritschacher: (bug 63624) Add JUnit test report to browsertest job [14:32:32] and assumes wikiid is in WikiMap [14:32:35] (03CR) 10Hoo man: [C: 04-1] "Rebased, still needs love" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [14:32:47] if it's not, wiki explodes with fatal error [14:33:13] i don't want to spend time to fix, right now, though [14:33:41] I wanted to fix that once... big, big mess... I gave up :P [14:33:49] it's not just one place [14:33:53] [13jenkins-job-builder-config] 15tobijat opened pull request #6: (bug 63624) Add JUnit test report to browsertest job (06master...06jbbtest) 02http://git.io/JnYU9w [14:34:04] and then i don't understand what the hook is doing [14:34:17] and what a good way to handle the error is (other than throw exception) [14:34:23] At some point it should use Sites, I guess... [14:34:39] which hook? [14:34:41] and if the sites table is empty [14:35:03] it could show warning or do something else which is sane [14:35:05] * aude files a bug... this one not for lydia :) [14:35:19] probably but the code is confusing [14:35:53] +1 [14:37:01] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [14:37:43] [13jenkins-job-builder-config] 15tobijat pushed 1 new commit to 06jbbtest: 02http://git.io/whPHIw [14:37:43] 13jenkins-job-builder-config/06jbbtest 149d8934a 15Tobias Gritschacher: Just testing something for bug 63624 [14:38:30] do we have a Wikibase hooks.txt file? [14:38:36] (03CR) 10Bene: "So we'd need another name in config for each badge. Isn't that a bit redundant? Additionally, the config should not contain that much info" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123441 (owner: 10Bene) [14:38:48] can't find one, although we have quite a few hooks [14:39:11] oh, why does bugzilla think the memory numbers in https://bugzilla.wikimedia.org/show_bug.cgi?id=63625 are gerrit commits ? [14:39:28] hoo: don't think so [14:39:57] mh.. AbuseFilter has one, Translate has one (AFAIR) and I even added on to RenameUser :P [14:40:27] aude: Because they have 8+ digits... :P [14:40:34] lol [14:41:34] [13jenkins-job-builder-config] 15tobijat pushed 1 new commit to 06jbbtest: 02http://git.io/xDlBqg [14:41:34] 13jenkins-job-builder-config/06jbbtest 1444378ff 15Tobias Gritschacher: Removed test line from README [14:42:42] Hey [14:47:45] (03PS3) 10Bene: Insert css styles for badges on client [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 [14:49:34] aude: Jenkins still broken? [14:50:34] benestar: shouldn't be [14:50:41] you might need to rebase changes [14:50:48] ok, we'll see ;) [14:50:50] if they are based on old, broken stuff [14:52:26] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [14:53:27] 9_) [14:53:30] 9_9 [14:53:50] probably needs rebase [14:54:00] e_e [14:54:00] xd [14:54:03] yep [15:01:02] What is this? [15:03:14] JohnLewis: hiiiii [15:03:31] Hi [15:03:53] (03PS4) 10Bene: Insert css styles for badges on client [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 [15:06:38] (03PS16) 10Daniel Kinzler: (bug 62644) move snak validation into ChangeOps [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 [15:06:51] (03CR) 10Daniel Kinzler: (bug 62644) move snak validation into ChangeOps (034 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [15:10:08] (03PS1) 10Aude: Fix site links group filtering in ItemView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124332 [15:10:39] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [15:10:56] [13jenkins-job-builder-config] 15tobijat pushed 1 new commit to 06jbbtest: 02http://git.io/Y077IQ [15:10:56] 13jenkins-job-builder-config/06jbbtest 14adaeb7b 15Tobias Gritschacher: Again some test for bug 63624 [15:11:16] aude: thx :D [15:11:26] -_- [15:11:31] (03PS4) 10Hoo man: Check the content model of pages in EntityPerPageBuilder [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 [15:12:13] Thiemo_WMDE: ok, you're right :P [15:12:19] see the patch set ;) [15:13:26] benestar: I feel really sorry. must be frustrating to you. sorry. [15:14:50] i want this to be almost perfect right from the start because it touches so many subsystems. basically all of them. repo and client extensions and both our wikidata.org repo and all wikipedia clients. [15:16:53] it took me some time to understand your basic idea with the per-client option. it's really nice. :-) [15:17:36] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [15:19:31] (03CR) 10Hoo man: "Addressed Daniel's comments, introduced WikibaseRepo::getContentMappings()" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [15:19:41] Hey what is this thing? [15:21:15] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124332 (owner: 10Aude) [15:24:17] Thiemo_WMDE: thanks :) [15:24:32] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [15:25:05] Thiemo_WMDE: why do you think we need css class names in repo, too? [15:25:38] it works in a totally different way, there will be around 100-200 badges, which would all need another name [15:27:40] (03PS5) 10Hoo man: Check the content model of pages in EntityPerPageBuilder [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 [15:27:44] [13WikibaseDataModel] 15JeroenDeDauw created 06terms (+1 new commit): 02http://git.io/9aES-Q [15:27:44] 13WikibaseDataModel/06terms 144b0b439 15jeroendedauw: Add new Term classes [15:28:43] addshore: Thiemo_WMDE: congraths, you totally confused me now with these two PRs on WB DB! [15:28:50] xD [15:29:10] (03CR) 10Hoo man: "Workaround NoBadDependencyUsageTest (as it also scans comments...)" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [15:29:47] hoo: can you open one for abuse filter please? that one can then be closed [15:30:26] Will do... not sure whether it's a Wikibase or AbuseFilter thing [15:30:32] should probably test with wikitext [15:32:20] :) [15:33:09] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [15:33:36] Lydia_WMDE: Looks like a Wikibase thing :/ [15:33:49] ok [15:33:52] thx for testing [15:34:35] whats the year before +00000000001-01-01T00:00:00Z [15:34:42] -00000000001-01-01T00:00:00Z ? xD [15:34:46] right? :P [15:34:48] (03CR) 10Thiemo Mättig (WMDE): [C: 04-1] "Almost there. :-) Very nice, thanks for the new patch set. The only problem is that the file WikibaseClient.hooks.php got lost for some re" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [15:35:18] [travis-ci] wmde/WikibaseDataModel/terms/4b0b439 : jeroendedauw The build failed. http://travis-ci.org/wmde/WikibaseDataModel/builds/22453450 [15:36:14] addshore: as long as the first year is ~=[,,_,,]:3 i'll be happy [15:36:19] [13jenkins-job-builder-config] 15tobijat pushed 1 new commit to 06master: 02http://git.io/OTmVsw [15:36:19] 13jenkins-job-builder-config/06master 14890d171 15Tobi Gritschacher: Merge pull request #6 from wmde/jbbtest... [15:37:10] well it would be great if someone could work out what they think the epoch timestamp of -00000000001-01-01T00:00:00Z would be :P [15:39:55] [13Time] 15addshore created 06sinceEpoch (+1 new commit): 02http://git.io/D-xt3Q [15:39:55] 13Time/06sinceEpoch 14f739985 15addshore: Introduce SinceEpochCalculator [15:40:06] DanielK_WMDE: JeroenDeDauw ^^ still needs more tests and work [15:40:16] would be good to get come inial comments [15:40:23] Lydia_WMDE: https://bugzilla.wikimedia.org/show_bug.cgi?id=63632 now do your bug black-magic :D [15:41:56] Thiemo_WMDE: will do this in another patch ... [15:42:08] Thiemo_WMDE: im sure youll want to look at http://git.io/D-xt3Q also [15:46:30] [13Time] 15thiemowmde comment on commit 14f739985: I really think this is not acceptable. If you are searching for stuff that happened on `2000-03-01` this will also return stuff that happened on `2000-02-29`. 02http://git.io/_3sUwg [15:47:49] (03PS1) 10Aude: Remove unused $wgExtensionMessagesFiles global [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124340 [15:48:30] someone want chocolate? :) [15:49:26] (03CR) 10Addshore: [C: 032] Remove unused $wgExtensionMessagesFiles global [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124340 (owner: 10Aude) [15:49:37] airmail the chocolate? :) [15:49:42] oh noes... to slow [15:49:50] addshore: help yourself with the chocolate [15:49:51] (03Merged) 10jenkins-bot: Remove unused $wgExtensionMessagesFiles global [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124340 (owner: 10Aude) [15:49:53] * aude brings chocolate to zürich [15:49:56] :D [15:50:01] :DD [15:50:07] even found the german key :) [15:50:21] aude: there IS chocolate in zürich [15:50:26] there is! [15:50:33] lol, lots of chocolate [15:51:15] ok, back to help figure out why localisation update fails [15:51:27] (possibly wikidata issue) [15:51:28] * hoo needs a break :P [15:51:35] oh well :/ [15:52:44] (03CR) 10WikidataJenkins: "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124340 (owner: 10Aude) [15:55:41] addshore: i can't find my initial comments about epoch/epoche/timestamps on github. help! [15:56:00] are they on the pull req in WikibaseQueryEngine? [15:56:14] https://github.com/wmde/WikibaseQueryEngine/pull/21 [15:57:03] addshore: tanks! <-- unintentional typo intentionally not fixed [15:57:59] Thiemo_WMDE: I think I know how to account for leap years actually, just chop of the last 4 digits from the year like we do in other places and use that instead of always using 1970 [16:01:59] [13WikibaseDataModel] 15JeroenDeDauw 04force-pushed 06terms from 144b0b439 to 143fccec1: 02http://git.io/ap9HWA [16:01:59] 13WikibaseDataModel/06terms 143fccec1 15jeroendedauw: Add new Term classes [16:02:09] [13WikibaseDataModel] 15JeroenDeDauw opened pull request #34: Add new Term classes (06master...06terms) 02http://git.io/D8sm8w [16:03:35] [13WikibaseDataModel] 15JeroenDeDauw 04force-pushed 06terms from 143fccec1 to 14a682140: 02http://git.io/ap9HWA [16:03:35] 13WikibaseDataModel/06terms 14a682140 15jeroendedauw: Add new Term classes... [16:06:17] addshore: sure. i think we did exactly this in other places. :-) but I'm afraid the simple $years*$secondsPerYear will still be a problem. [16:06:37] :P [16:07:13] indeed, I would probably have to calculate home many leap years are within the number of years :/ [16:07:40] You can haz 20 new files! https://github.com/wmde/WikibaseDataModel/pull/34 [16:07:48] merge merge merge :D [16:08:43] addshore: probably something like this. i would love to help now that i understand what this is for. :-) [16:09:06] Lydia_WMDE: can we put https://bugzilla.wikimedia.org/show_bug.cgi?id=63623 in the sprint, if we can't do https://gerrit.wikimedia.org/r/#/c/124321/ for now? [16:09:19] not sure that's something i can actually work on though [16:09:36] Thiemo_WMDE: is it ok if we implement the hook in another patch so that this can get merged right now? [16:10:09] benestar|afk: i think so, yes. not sure if it's ok for the rest of the team. [16:10:35] [travis-ci] wmde/WikibaseDataModel/terms/3fccec1 : jeroendedauw The build has errored. http://travis-ci.org/wmde/WikibaseDataModel/builds/22456296 [16:10:49] whatttt [16:13:26] :D [16:13:32] travis says NOM [16:20:24] http://askhelixfossil.com/#ew4h51 [16:20:30] JeroenDeDauw: ^ [16:20:33] here you go [16:22:16] so Thiemo_WMDE what about leap years in BC [16:27:26] [travis-ci] wmde/WikibaseDataModel/terms/a682140 : jeroendedauw The build was fixed. http://travis-ci.org/wmde/WikibaseDataModel/builds/22456392 [16:28:14] (03PS1) 10Aude: New build, fixes for phpunit tests, typography and page protection bug [extensions/Wikidata] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124348 [16:35:40] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikidata] (mw1.23-wmf21) - 10https://gerrit.wikimedia.org/r/124348 (owner: 10Aude) [16:35:51] * aude cries.... central auth is setting a header() all because IE requires it [16:53:28] (03PS2) 10Aude: Fix site links group filtering in ItemView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124332 [16:53:30] (03PS1) 10Aude: Use TestSites site in ItemContentTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124353 [17:00:11] (03CR) 10Bene: "This will be done in another patch. Can we merge this one now?" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123447 (owner: 10Bene) [17:01:01] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124332 (owner: 10Aude) [17:04:38] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124353 (owner: 10Aude) [17:27:31] (03PS4) 10Bene: Support getting single sitelinks with ClientSiteLinkLookup [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123449 [17:29:56] (03PS5) 10Bene: Support getting single sitelinks with ClientSiteLinkLookup [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123449 [17:30:59] Thiemo_WMDE: ok, what says the team? [17:31:58] (03PS1) 10Aude: Add clearfix css for edit link in Vector skin [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124358 [17:34:56] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123449 (owner: 10Bene) [17:39:54] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123449 (owner: 10Bene) [17:43:04] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124358 (owner: 10Aude) [18:28:05] lazowik: lolwat [18:28:17] lazowik: http://askhelixfossil.com/#uq0o4i [18:28:59] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: Shouldn't there be an array_key_exists check? 02http://git.io/F7813w [18:29:12] lazowik: http://askhelixfossil.com/#uefza1 [18:31:27] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: Is this going to have methods like equals()? 02http://git.io/_1_4sA [18:32:14] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: Is this a mistake that needs to be fixed before this is merged? 02http://git.io/h0a_Sg [18:33:19] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: Again, shouldn't this do an array_key_exists check? 02http://git.io/9gnLEQ [18:34:44] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: This is exactly how $this->terms is constructed in the constructor. Why not simply return it as it is? 02http://git.io/MWFBzg [18:37:00] has any dev some time atm to help me? [18:37:29] JeroenDeDauw: :p [18:37:30] https://en.wikipedia.org/wiki/Twitch_Plays_Pok%C3%A9mon [18:37:39] haven't heard about that? [18:37:49] [13WikibaseDataModel] 15JeroenDeDauw comment on pull request #34 14a682140: I can yes. Now we have dedicated objects, there are many more things we can potentially put in here, in a clean way. 02http://git.io/_WYuqg [18:38:44] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: I'm fine with that. But shouldn't this be in the inline documentation then? 02http://git.io/ulY2gg [18:39:54] [13WikibaseDataModel] 15JeroenDeDauw comment on pull request #34 14a682140: $array[$term->getLanguageCode()] = $term->getText();... 02http://git.io/ehL-LA [18:40:39] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 14a682140: Sure, just asking. isEmpty, equals, compareTo and such are things I would love to see here some day. :-) 02http://git.io/Aro5yA [18:41:09] [13WikibaseDataModel] 15JeroenDeDauw 04force-pushed 06terms from 14a682140 to 1430fac0a: 02http://git.io/ap9HWA [18:41:09] 13WikibaseDataModel/06terms 1430fac0a 15jeroendedauw: Add new Term classes... [18:41:44] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 1430fac0a: That was the other comment. I already deleted it. ;-) 02http://git.io/g-9d8w [18:46:45] hoo: aude: after https://gerrit.wikimedia.org/r/#/c/122124/ did you check if https://bugzilla.wikimedia.org/show_bug.cgi?id=48086 still works. Because this was the reason why the lines that you now removed were in there.. [18:47:36] Tobi_WMDE: Not sure it worked with the old version, but should work with the new one [18:47:45] would be bad if that bug gets introduced again.. [18:48:37] it did not work before the additional checks got added. now they got removed.. so please check [18:48:41] don't think so, the change can only (theoretically) make it that people see stuff as editable which actually isn't [18:48:46] compared to the old version [18:49:05] but it can't have regressions where it disallows editing of content which you could edit beforehand [18:49:36] I'll have a look later on, but I doubt it's something to worry about [18:49:39] see https://gerrit.wikimedia.org/r/#/c/98796/ [18:50:50] Tobi_WMDE: Ah, that makes sense... so nothing to worry about [18:50:50] yeah, please check. I don't expect anything critical there.. but since we already backported that change we should be sure [18:51:02] I'll test it later on [18:53:18] [13Time] 15JeroenDeDauw comment on commit 14f739985: DecimalMath, which is somewhat similar, is in the DataValues namespace: https://github.com/DataValues/Number/tree/master/src/DataValues... 02http://git.io/i5OvQA [18:53:37] Thiemo_WMDE: merge? :D [18:54:53] [13WikibaseDataModel] 15thiemowmde comment on pull request #34 1430fac0a: Please expand this description a bit. To me this sounds like it's throwing an Exception if I break the "only one" rule. But it silently drops extra values and uses the last one. I think this should be described.... 02http://git.io/quX5zw [18:56:14] JeroenDeDauw: should i merge and you expand the doc in a follow-up? [18:56:53] Thiemo_WMDE: what doc? [18:57:10] JeroenDeDauw: https://github.com/wmde/WikibaseDataModel/pull/34#discussion_r11359473 [18:58:10] Thiemo_WMDE: bah [18:58:19] Thiemo_WMDE: the true specification is in the tests, not the doc ;p [18:58:45] is this the new "the spec is in the code"? [18:58:56] ;))) [19:00:20] [13WikibaseDataModel] 15JeroenDeDauw 04force-pushed 06terms from 1430fac0a to 1414f0ca9: 02http://git.io/ap9HWA [19:00:20] 13WikibaseDataModel/06terms 1414f0ca9 15jeroendedauw: Add new Term classes... [19:01:00] Thiemo_WMDE: no, having it in the tests actually makes sense and takes effort to do [19:01:31] hey, i was kidding. ;) [19:02:24] [13WikibaseDataModel] 15thiemowmde closed pull request #34: Add new Term classes (06master...06terms) 02http://git.io/D8sm8w [19:13:12] [travis-ci] wmde/WikibaseDataModel/terms/14f0ca9 : jeroendedauw The build has errored. http://travis-ci.org/wmde/WikibaseDataModel/builds/22470224 [19:15:04] [13WikibaseDataModel] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/SC7_Lg [19:15:04] 13WikibaseDataModel/06master 1495a76a0 15jeroendedauw: Optimize imports for tests [19:19:41] [13WikibaseDataModel] 15JeroenDeDauw created 06testLoader (+1 new commit): 02http://git.io/feZXtA [19:19:41] 13WikibaseDataModel/06testLoader 141b8498c 15jeroendedauw: Replace custom classmap loading for tests with PSR-4 loader [19:19:56] [13WikibaseDataModel] 15JeroenDeDauw opened pull request #35: Replace custom classmap loading for tests with PSR-4 loader (06master...06testLoader) 02http://git.io/f1Tv5A [19:25:36] fuuuuuuuuuu php [19:26:31] Thiemo_WMDE: lowevely how you cannot type hint the contents of an iterator :( [19:29:55] [13WikibaseDataModel] 15JeroenDeDauw pushed 1 new commit to 06master: 02http://git.io/In2sWg [19:29:55] 13WikibaseDataModel/06master 14ee4578d 15Jeroen De Dauw: Update AliasGroupList.php [19:33:15] [13WikibaseDataModel] 15thiemowmde created 06code-smell (+1 new commit): 02http://git.io/AmcIFA [19:33:15] 13WikibaseDataModel/06code-smell 141712a03 15Thiemo Mättig: Avoid code smell warnings by adding LogicExceptions... [19:33:50] [13WikibaseDataModel] 15thiemowmde opened pull request #36: Avoid code smell warnings by adding LogicExceptions (06master...06code-smell) 02http://git.io/kmebag [19:35:58] [13WikibaseInternalSerialization] 15JeroenDeDauw created 06move (+1 new commit): 02http://git.io/6rVMag [19:35:58] 13WikibaseInternalSerialization/06move 148fda9be 15jeroendedauw: Move over new Term classes to DataModel [19:37:19] [13WikibaseDataModel] 15JeroenDeDauw comment on pull request #36 141712a03: The type is called callable http://www.php.net/manual/en/language.types.callable.php... 02http://git.io/KN9xbg [19:38:49] [13WikibaseDataModel] 15JeroenDeDauw comment on pull request #36 141712a03: This is silly. It's clear this cannot happen. If the static code analysis tools did not suck they would find that as well. The called method above ALWAYS throws an exception. 02http://git.io/wNUfhA [19:44:33] Tobi_WMDE: hoo|away confirmed / autoconfirmed works way it should (no problem) [19:46:03] it' s bug prone and usually mistake to duplicate core functionlity like we did there in the js [19:46:08] aude: Great :) Everything else would have surprised me, though [19:46:27] Yep... just checking the global is the right way to go... [19:46:38] as we saw, buggy [19:47:20] (03CR) 10Thiemo Mättig (WMDE): (bug 62644) move snak validation into ChangeOps (0310 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [19:49:08] Thiemo_WMDE: re your comment of languageCode vs Language objects: somewhat moot point in DM - it does not know about Language objects. [19:49:19] Also, we should try not to use Language objects directly in WB [19:49:24] Evil class [19:50:28] agree. however, avoiding $language as variable name helps a lot. it's so unclear. [19:50:42] agree [19:53:38] (03CR) 10Thiemo Mättig (WMDE): [C: 031] "I'm curious why the removed method wasn't marked as deprecated? I assume it's not used any more. If that's true (as the tests should have " [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [19:59:04] Thiemo_WMDE: two changes seem to be ready to get merged --> today? [20:00:04] benestar: was a long day. not sure if i'm around long enough. [20:00:57] [13WikibaseDataModel] 15JeroenDeDauw created 06getFingerprint (+1 new commit): 02http://git.io/JdUT_A [20:00:57] 13WikibaseDataModel/06getFingerprint 14c20741a 15jeroendedauw: Added Entity::getFingerprint [20:01:07] [13WikibaseDataModel] 15JeroenDeDauw opened pull request #37: Added Entity::getFingerprint (06master...06getFingerprint) 02http://git.io/nzrgMA [20:02:03] Thiemo_WMDE: hehe, ok but I hope you are satisfied with the patches now [20:03:32] (03CR) 10Aude: (bug 62644) move snak validation into ChangeOps (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [20:05:10] (03CR) 10Thiemo Mättig (WMDE): (bug 62644) move snak validation into ChangeOps (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [20:05:43] (03PS1) 10Bene: DNM Insert HTML for badges in ItemView [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124391 [20:06:16] w00t [20:09:11] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124391 (owner: 10Bene) [20:10:26] [13WikibaseDataModel] 15JeroenDeDauw pushed 1 new commit to 06getFingerprint: 02http://git.io/cVewKg [20:10:26] 13WikibaseDataModel/06getFingerprint 144a391b0 15jeroendedauw: Added Entity::setFingerprint [20:10:59] (03CR) 10Thiemo Mättig (WMDE): [C: 04-1] (bug 62644) snak validation in more change ops (034 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [20:11:54] i really love doing code reviews. but it takes sooo muuuch tiiimeee. ;-) [20:20:03] (03CR) 10Thiemo Mättig (WMDE): [C: 04-1] "Almost a +2, just a typo and a question." (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [20:27:08] [travis-ci] wmde/WikibaseDataModel/code-smell/1712a03 : Thiemo Mättig The build passed. http://travis-ci.org/wmde/WikibaseDataModel/builds/22472471 [20:43:32] (03CR) 10Daniel Kinzler: (bug 62644) move snak validation into ChangeOps (0311 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121689 (owner: 10Daniel Kinzler) [20:44:56] (03CR) 10Daniel Kinzler: "Which removed method, exactly?" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123860 (owner: 10Daniel Kinzler) [20:45:07] (03CR) 10WikidataJenkins: "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124456 (owner: 10L10n-bot) [20:46:22] [travis-ci] wmde/WikibaseInternalSerialization/move/8fda9be : jeroendedauw The build passed. http://travis-ci.org/wmde/WikibaseInternalSerialization/builds/22472642 [20:48:38] (03CR) 10Daniel Kinzler: (bug 62644) snak validation in more change ops (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123872 (owner: 10Daniel Kinzler) [20:50:43] (03CR) 10Daniel Kinzler: (bug 62644) No snak val in ClaimModificationHelper (032 comments) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124317 (owner: 10Daniel Kinzler) [21:13:55] [travis-ci] wmde/WikibaseDataModel/getFingerprint/c20741a : jeroendedauw The build passed. http://travis-ci.org/wmde/WikibaseDataModel/builds/22474352 [21:40:04] [travis-ci] wmde/WikibaseDataModel/getFingerprint/4a391b0 : jeroendedauw The build passed. http://travis-ci.org/wmde/WikibaseDataModel/builds/22474929 [21:55:57] mmh [21:55:59] aude: Around? [22:17:33] * addshore waves [22:21:44] addshore the dinosaur! [22:21:51] :> [22:21:56] hope you enjoyed :P [22:22:02] very :D [22:22:17] * Lydia_WMDE wonders what addshore will do tomorrow then ;-) [22:23:35] hehe :P [22:23:46] [13WikibaseDataModel] 15addshore 04deleted 06testLoader at 141b8498c: 02http://git.io/F9qAeQ [22:24:30] (03PS1) 10Hoo man: Check for array_key_exists in SettingsArray before throwing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 [22:24:48] This is needed as PHP says the offset exists if the value [22:24:49] is null, while HHVM claims it doesn't (which is in accordance [22:24:49] to the PHP manual, in contrast to PHP's own behaviour). [22:24:55] addshore: ^ [22:25:00] This is why I love PHP so much :) [22:26:00] lol :P [22:26:24] (03CR) 10Ori.livneh: "LGTM. Ok to merge?" (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:26:55] array_key_exists( $this, $settingName [22:26:55] ) [22:27:08] mh? [22:27:09] should this not be the other way around? Or an I missing something and tired? [22:27:26] Second :P [22:27:27] $key, $array [22:27:37] $this is the key? O_o [22:27:37] oh that, right [22:27:40] * hoo slaps [22:27:45] himself and php [22:27:45] :P [22:27:59] (03CR) 10Addshore: [C: 04-1] Check for array_key_exists in SettingsArray before throwing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:28:07] good night folks [22:28:08] <3 [22:28:12] night Lydia_WMDE :) [22:28:16] (03PS2) 10Hoo man: Check for array_key_exists in SettingsArray before throwing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 [22:28:20] Lydia_WMDE: Good night :) [22:28:36] that looks better hoo ;p [22:29:13] will wait for jenkins [22:29:17] :) [22:29:32] Probably leave the bug open after... to wait for the upstream fix?! [22:29:32] god knows what the tests for this actually test ;p [22:29:46] (03CR) 10Addshore: [C: 031] "will +2 after jenkins +2Vs" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:31:11] (03CR) 10WikidataJenkins: [V: 04-1] "Build Failed" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:31:18] lets ignore that one ;p [22:31:48] wait what [22:31:57] thats ps1 right? *looks* [22:32:09] yes :P which should indeed fail ;p [22:32:16] oh right [22:32:17] :P [22:32:20] addshore: If we're ignoring WikidataJenkins, can we ignore you too? :p [22:32:30] JohnLewis: never :> [22:32:55] addshore: If your code fails, do you become minushore? :p [22:33:44] no :O [22:33:49] im always posotive ;p [22:34:29] unsigned longlong happines = ~0; [22:34:34] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:34:45] (03CR) 10Addshore: [C: 032] Check for array_key_exists in SettingsArray before throwing [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124480 (owner: 10Hoo man) [22:34:57] and bang goes the dynamite [22:34:58] * long long [22:35:22] pity that PHP doesn't have unsigned ints... all these jokes don't work with it :P [22:39:12] (03CR) 10Addshore: [C: 031] Slightly enlarge Snak type (custom/unknown/no value) icons [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120201 (owner: 10Thiemo Mättig (WMDE)) [22:39:59] so the wikiquote launch is still on for tomorrow? [22:40:02] (03CR) 10Addshore: [C: 032] Use TestSites site in ItemContentTest [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124353 (owner: 10Aude) [22:41:47] rschen7754: Yes [22:41:51] ok [22:41:53] (03CR) 10Addshore: [C: 032] Fix issue of blocked user seeing page protection instead of block message [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/124308 (owner: 10Aude) [22:45:08] (03CR) 10Addshore: [C: 031] "looks nice" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123562 (owner: 10Henning Snater) [22:47:49] (03CR) 10Addshore: [C: 032] Add option in travis setup to run client or repo tests only, or both [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123863 (owner: 10Aude) [22:47:55] (03PS5) 10Addshore: Have travis test variety of setups (client-only, repo-only and both) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [22:52:45] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123864 (owner: 10Aude) [22:53:04] (03CR) 10Addshore: [C: 04-1] Add / allow injecting experimental mode param in LanguageFallbackChainFactory (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/119502 (owner: 10Aude) [22:55:02] (03CR) 10Addshore: [C: 031] (bug 57693) supply id for claim group [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/121361 (owner: 10Daniel Kinzler) [22:59:13] (03CR) 10Addshore: [C: 031] Support getting single sitelinks with ClientSiteLinkLookup [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/123449 (owner: 10Bene) [23:02:47] (03CR) 10Addshore: [C: 04-1] Check the content model of pages in EntityPerPageBuilder (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [23:07:39] [travis-ci] wikimedia/mediawiki-extensions-Wikibase/master/f793122 : Addshore The build has errored. http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/22485501 [23:09:41] night all [23:12:34] 'night, addshore [23:15:56] (03PS6) 10Hoo man: Check the content model of pages in EntityPerPageBuilder [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 [23:16:23] (03CR) 10Hoo man: Check the content model of pages in EntityPerPageBuilder (031 comment) [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [23:16:43] (03CR) 10Hoo man: "Addressed Addshore's comment" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [23:16:49] [= [23:17:05] addshore: That's not how you go to bed :P [23:17:06] :D [23:17:24] Ok, I wanted to go to bed like more than an hour ago :P [23:21:09] (03CR) 10WikidataJenkins: [V: 032] "Build Successful" [extensions/Wikibase] - 10https://gerrit.wikimedia.org/r/120857 (owner: 10Hoo man) [23:24:35] [travis-ci] wikimedia/mediawiki-extensions-Wikibase/master/a18d00d : Addshore The build was broken. http://travis-ci.org/wikimedia/mediawiki-extensions-Wikibase/builds/22485918