[07:49:14] Change merged: Tobias Gritschacher; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24553 [08:03:23] Change merged: Tobias Gritschacher; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24556 [08:07:20] New patchset: Tobias Gritschacher; "removed hardcoded sleep" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24775 [09:32:46] DanielW_WMDE: if you want to work on 40355 (rc) that's fine [09:33:37] i am working now on the client updating when items get deleted (and undeleted) and should have patch for that shortly [09:34:06] oops [09:34:14] DanielK_WMDE: ^ :) [09:34:37] DanielW_WMDE: please change your nick back to Danwe :) [09:35:28] DanielW_WMDE: do not listen to aude on this one! [09:36:38] no! [09:37:08] * jeblad_WMDE goes #atsji [09:39:46] me pokes DanielK_WMDE [09:44:28] DanielK_WMDE, jeblad_WMDE: i'm gonna upload a fix for UI breakage in https://gerrit.wikimedia.org/r/#/c/24541/ now. [09:45:40] Anything special that failed? [09:45:53] New patchset: Tobias Gritschacher; "(Bug 40407, Bug 40408) Change from item and items into entity and entities" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24541 [09:46:21] Ah, that one! cool! [09:46:33] jeblad_WMDE: the whole changeset needs a rebase though. [09:46:56] Yeah it colides with a lot of stuff [09:48:03] going to look at 24541 unless someone else is doing the rebase? [09:48:10] Anyone? [09:48:28] * jeblad_WMDE pokes all with an electric cattleprod [09:49:27] jeblad_WMDE: I'm not gonna do it. [09:54:49] I get sick and nobody else, thats the usual.. [09:55:46] jeblad_WMDE: let's hope it doesn't spread. get well soon! [09:55:58] Not sure why, but me getting so sick of flue and cold started after working with some chemicals at the electric power supply company [09:56:43] Its 12 hour from you get it until outbreak [09:57:01] And it wasn't Henning, I got it over at Spandau [09:57:59] Others gets superhuman strength by exposure to weird chemicals, I only got supersensitivity to cold and flue.. #darn [09:59:20] aude: that's my new crappy irc client always changing my nick [10:00:06] heh [10:00:22] Be careful when using glyfosat, that shit is not as safe as they say [10:03:27] This way of addressing stuff in datastructures are somewhat fragile.. [10:03:31] https://gerrit.wikimedia.org/r/#/c/24541/1..2/repo/resources/wikibase.ui.PropertyEditTool.EditableValue.js [10:03:41] I would not recomend to do it [10:04:49] Code doe like that has a nasty habit of breaking and then you get fatal errors [10:04:58] -doe [10:04:58] 04Error: Command “doe” not recognized. Please review and correct what you’ve written. [10:05:14] :D [10:05:19] -doe [10:05:19] 04Error: Command “doe” not recognized. Please review and correct what you’ve written. [10:07:08] jeblad_WMDE: what you mean? do you mean the dot notation? [10:07:25] Yes [10:08:44] Try to avoid the form if possible, remember this data, not code [10:09:21] this is.. [10:09:37] http://www.dev-archive.net/articles/js-dot-notation/ [10:09:53] your suggestion simply is not correct for js [10:10:10] dot notation is faster to write and clearer to read as the article says [10:10:17] under the hood it is exactly equivalent [10:10:39] and the dot notation has the big advantage of being autocompleted by decent editors [10:10:50] Feel free, I've been using code built like that for some years and it breaks horribly [10:10:58] in javascript? [10:11:23] and *how* does it break? [10:11:23] In javascript among others [10:11:37] The basic assumption is that code and data is the same [10:11:43] That is very seldom true [10:12:01] But use it if you like it, but watch out for fatal breakage [10:12:16] *how* would it break?? [10:12:28] what kind of fatal brekage? [10:13:12] If you do foo['something'].bar you know you do something dangerous [10:13:26] If you do foo.something.bar you don't [10:14:04] When the assumption that "something" doesn't hold the code will break [10:14:44] with functions being assignable to variables, the difference between code and data is indeed much smaller and much less cut clear than you describe [10:14:46] And when "something" is data supplied from someone else the code is on thin ice [10:14:47] ask the lispers [10:15:01] jeblad_WMDE: it's JSON, as the name says "Object Notation", and the way to access data in an object is with the dot-notation. It breaks if the object changes but that's how it works. [10:15:17] it would break with array notation as well [10:15:21] sure [10:15:24] what advantage would the array notation bring? [10:15:55] I've written lisp since the 80', I know a good part of the problems with claiming data is code [10:16:42] Always access data after making sure the dat is in fact there, never assume the data is there [10:19:27] but why is it different for code? [10:19:31] code is as malleable as data [10:20:18] jeblad_WMDE: I see your point. we shoud add some further checks indeed.. [10:20:22] the original question is: why is the array notation better than the dot notation? why would that not fail horribly? doesn't both stop the script from running? [10:21:53] Change merged: Daniel Werner; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24775 [10:27:31] Change merged: Daniel Werner; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24558 [10:32:49] jeblad_WMDE: are you working on the gettoken stuff? i'm about to dig into your change set. [10:33:03] hi DanielK_WMDE [10:33:10] hi aude [10:33:36] Denny_WMDE: I see no difference. checks should be done in both cases. [10:33:58] i have enough to do at the moment for handling item deletions/undeletions, noexternallinks thing, tests etc. [10:34:15] if you want to work on recent changes, that's fine, although i'd be interested in poking at it some later maybe [10:34:24] jeblad_WMDE: https://gerrit.wikimedia.org/r/#/c/24541/ requires a rebase [10:34:27] TobiasG_WMDE: agree [10:35:19] aude: have you already written code for the recentchanges injection? [10:35:25] DanielK_WMDE: not yet [10:35:44] ok. i'll poke at it then, after looking into the gettoken stuff [10:36:04] * aude doesn't like to start on too many things while others are still in "doing" stage [10:37:57] if the gettoken stuff gets done in the next half hour, we can still put it in the to-review version. otherwise we will just have it later. [10:38:03] jeblad_WMDE: what's the status of the rebase? [10:38:46] DanielK_WMDE: when are you in berlin again to talk about 29c3? [10:38:58] Lydia_WMDE: thursday [10:39:01] k [10:39:08] yay, 29c3 ;) [10:39:13] :D [10:39:30] i'll schedule something for you, jens and me then [10:39:45] cool [10:43:46] New patchset: Daniel Kinzler; "(Bug 39366) Kill use of gettoken [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24559 [10:43:49] I just deleted all data from test system to fix database errors... [10:44:23] New review: Daniel Kinzler; "rebased, still failing tons of tests. looking into that." [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/24559 [10:49:08] New patchset: Daniel Werner; "Added missing JS dependency to jquery.uls which was causing failing QUnit tests" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24782 [10:54:32] jeremyb: poke when you're around please :) [10:59:00] New patchset: Daniel Werner; "Removed QUnit test for autocomplete.autocompleteString()" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24783 [11:02:31] DanielK_WMDE: I sent a mail with two changes that should REALLY go into the review branch [11:03:18] since they significantly lower the posibility of failing QUnit test stuff^^ [11:04:54] Danwe_WMDE: Anja ist in charge of the branch... just submit them to the branch and add anja and me as a reviewer. [11:05:19] anja can just merge it, or I'll have a look later [11:06:23] DanielK_WMDE: I already submitted them to the actual master, can I add you for review there and you pick the changes afterwards? [11:08:07] Danwe_WMDE: you could, but if you submit them to master and beta, it's more likely that a) i don't forget and b) someone else does it :) [11:08:46] jeblad_WMDE: that session problem in your gettoken change is indeed quite strange. [11:08:55] it even kilsl the debugger >_< [11:10:59] AnjaJ_WMDE: what is the name of the review branch? [11:12:18] Denny_WMDE: wikidata-wmfphase1beta [11:12:24] TobiasG_WMDE: ^^ [11:12:30] AnjaJ_WMDE: thx [11:18:27] OK, the repo client replication on our test system is working again. [11:18:41] Silke_WMDE: yay! [11:20:34] New patchset: Daniel Werner; "QUnit test for autocomplete.autocompleteString()" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24787 [11:23:04] New review: Daniel Kinzler; "seems fine, not tested" [mediawiki/extensions/Wikibase] (master); V: 0 C: 2; - https://gerrit.wikimedia.org/r/24782 [11:23:12] New review: Daniel Kinzler; "seems fine, not tested" [mediawiki/extensions/Wikibase] (master); V: 0 C: 2; - https://gerrit.wikimedia.org/r/24783 [11:23:23] New patchset: Daniel Werner; "Added missing JS dependency to jquery.uls which was causing failing QUnit tests" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24789 [11:25:31] DanielK_WMDE: AnjaJ_WMDE: done [11:26:03] great. anja is looking into it. let me know if you need my help [12:01:54] DanielK_WMDE: Keep the gettoken for now and flag it as deprecated? [12:02:27] jeblad_WMDE: no, i found the problem. running some more tests, but i think that's it. [12:02:39] DanielK_WMDE: There is an upcoming problem with prefixes, it clashes with how something not found is marked and handled [12:03:03] how exactly does it cause a problem? [12:03:14] The rest of the API and our code marks not found items with negative numbers [12:03:30] Negative strings would be a new concept.. [12:03:33] :D [12:03:50] is anything actually relying on that? [12:04:00] why don't we just use the full ID and "missing"? [12:04:18] ok, test passed, amending [12:04:20] We could stick with flags in the returned structure, I think that is okey, but havent checked anything [12:04:27] i'll also submit a follow up soon [12:04:35] What did cause the problem? [12:04:50] let me put this on gerrit, then you can see the diff [12:04:54] it's two things, really [12:05:16] I suspected a wacko session, or a reset session [12:05:36] there'S a static token cache. it needs to be reset. that was the first problem [12:06:03] the second problem was that getItemToken didn't return the token, but an entiry array of stuff. just the wrong thing from the result structure. [12:06:18] New patchset: Daniel Kinzler; "(Bug 39366) Kill use of gettoken [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24559 [12:06:50] New review: Daniel Kinzler; "fixed token handling in test cases" [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/24559 [12:07:11] getItemToken? Didn't I fix that? Perhaps I didn't uploaded before I left on friday [12:07:27] But nice work! [12:07:34] jeblad_WMDE: look at the diff between patch set 3 and 4 [12:07:49] well, took me nearly 3 hours :D [12:09:44] Strange, I thought I ddn't use the token cache [12:21:25] Change merged: Daniel Kinzler; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24782 [12:21:41] Change merged: Daniel Kinzler; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24783 [12:22:04] Change abandoned: Daniel Kinzler; "wrong branch" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24787 [12:22:16] Change abandoned: Daniel Kinzler; "wrong branch" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24789 [12:22:41] Danwe_WMDE: to submit a change to wikidata-wmfphase1beta, you have to use git review wikidata-wmfphase1beta [12:23:02] therwise, you are just creating a topic branch on master [12:23:12] great [12:23:20] learned that on friday - the hard way ;) [12:23:32] So I have to abandone them and do it again or what? [12:23:57] i abandoned them [12:23:58] yea [12:24:05] you can reuse your local branch [12:24:11] sure [12:24:24] just do commit --amend and remove the change id [12:24:27] so it will be a new change [12:24:37] then submit again [12:24:45] btw, i merged them to master already [12:24:52] * jeblad_WMDE makes a note to self that git can screw up in strange and mysterious ways [12:25:08] jeblad_WMDE: gerrit, in that caser [12:25:22] and it is kind of consistent, though it could be smarter about these things [12:25:39] I have a method that is a mixture of old and new code.. [12:25:46] like "if the user is on a remote branch, they probably also want to submit to that branch" [12:25:54] It looks vejjjy wejjjd.. [12:26:30] jeblad_WMDE: hehe, oh yea. automatic merges *usually* work. but sometimes they fail in very strange way. [12:26:41] diff/merge is looking at this stuff though a keyhole [12:26:53] a merge system that was aware of scopes and function names would be awesome [12:27:12] Its strange that it works so well so often.. [12:27:50] New patchset: Daniel Werner; "Added missing JS dependency to jquery.uls which was causing failing QUnit tests" [mediawiki/extensions/Wikibase] (wikidata-wmfphase1beta) - https://gerrit.wikimedia.org/r/24791 [12:28:27] deceptive :) [12:29:18] New patchset: Daniel Werner; "Removed QUnit test for autocomplete.autocompleteString()" [mediawiki/extensions/Wikibase] (wikidata-wmfphase1beta) - https://gerrit.wikimedia.org/r/24792 [12:29:44] DanielK_WMDE: there they are [12:34:50] New patchset: John Erling Blad; "(Bug 40407, Bug 40408) Change from item and items into entity and entities" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24541 [12:35:19] Change merged: Anja Jentzsch; [mediawiki/extensions/Wikibase] (wikidata-wmfphase1beta) - https://gerrit.wikimedia.org/r/24791 [12:35:43] Change merged: Anja Jentzsch; [mediawiki/extensions/Wikibase] (wikidata-wmfphase1beta) - https://gerrit.wikimedia.org/r/24792 [12:37:24] It seems like https://gerrit.wikimedia.org/r/#/c/24541/3 works here, but after kill-gettoken is merged it probably need another rebase [12:38:38] TobiasG_WMDE: Can you kick of another run with selenium tests? I checked Qunit tests and they work [12:39:04] I have done additional changes in the UI.. :D [12:39:33] New patchset: Denny Vrandecic; "Tests for SpecialItemDisambiguation" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24793 [12:44:04] jeblad_WMDE: sure, but u have to wait ~30 mins [12:44:20] it's testing the review branch at the moment [12:46:11] thats fine for me.. [12:46:42] I think this should go in there, but thats up to AnjaJ_WMDE [12:47:01] which change, jeblad_WMDE ? [12:47:23] change from use of item/items to entity/entities [12:47:24] yep also wondering which ;) [12:47:39] https://gerrit.wikimedia.org/r/#/c/24541/ ? [12:47:42] tests are running, so I doubt another merge and test run is doable before 3pm [12:48:08] and it is not even reviewed and merged yet to master… i am afraid it's too late, sorry [12:48:40] I don't mind, but then tere will be breaking changes later [12:48:51] yes, we will have these anyway i am afraid [12:48:58] https://gerrit.wikimedia.org/r/#/c/24541/ [12:49:11] we just need to communicate that [12:49:47] There are also a type-flag that needs to be added, and.. we have used a few deprecated functions.. [12:49:47] Lydia_WMDE: can you write that down as a todo item: in case we expect breaking changes to the API soon after deployment, let the bot writers know [12:50:01] k [12:50:13] the use of deprecated functions should not force us to change the API… ? [12:50:40] No, but they should be removed soon-ish [12:50:56] agree [12:55:07] New review: Aude; "tests look good to me and appear to test the right stuff." [mediawiki/extensions/Wikibase] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/24793 [13:00:38] New patchset: Daniel Kinzler; "Cleaned up token handling for test cases" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24794 [13:03:22] New patchset: Daniel Kinzler; "(Bug 39366) Kill use of gettoken" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24559 [13:04:30] New patchset: Daniel Kinzler; "Cleaned up token handling for test cases" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24794 [13:05:21] New patchset: Tobias Gritschacher; "Tests for SpecialItemDisambiguation" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24793 [13:05:33] aude: you can merge it if you want. [13:07:36] jeblad_WMDE: AnjaJ_WMDE will rock the tests! [13:07:41] :-)= [13:12:05] New review: Aude; "looks good but haven't actually tried this it out with selenium to verify this works" [mediawiki/extensions/Wikibase] (master); V: 0 C: 2; - https://gerrit.wikimedia.org/r/24793 [13:12:06] TobiasG_WMDE: how do i try it? [13:18:34] New patchset: Jeroen De Dauw; "Fixed incorrect check in test" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24800 [13:22:55] New review: Tobias Gritschacher; "works." [mediawiki/extensions/Wikibase] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/24793 [13:22:55] Change merged: Tobias Gritschacher; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24793 [13:32:49] It's a TobiasG_WMDE: http://wikidata-docs.wikimedia.de/testcoverage/ [13:35:55] New patchset: Jeroen De Dauw; "Added uniqueness restriction for label+description in the same language for items" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24567 [13:35:56] New patchset: Jeroen De Dauw; "Added getMatchingJoinedTerms method to TermCache" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24566 [13:35:56] New patchset: Jeroen De Dauw; "Add support for self joins to getMatchingTerms" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24563 [13:52:42] New patchset: Jeroen De Dauw; "Put datatypes, valueparsers and valuevalidators into their own namespaces" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24801 [13:53:21] New patchset: Jeroen De Dauw; "killed obsolete comment" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24802 [13:54:37] Change merged: Jeroen De Dauw; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24802 [13:55:45] New patchset: Jeroen De Dauw; "Put datatypes, valueparsers and valuevalidators into their own namespaces" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24801 [13:55:45] New patchset: Jeroen De Dauw; "Fixed incorrect check in test" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24800 [13:59:18] Danwe_WMDE: ~=[,,_,,]:3 [14:05:25] New patchset: Jeroen De Dauw; "Update names and links to plural" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24805 [14:07:24] New patchset: Jeroen De Dauw; "Update name of valueparsers extension" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24806 [14:07:46] New review: Jeroen De Dauw; "Depends on I3caeb76899fc911800a4e09f74ed176472a2d4fe" [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/24806 [14:08:58] New patchset: Jeroen De Dauw; "Update names and links to plural" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24805 [14:19:02] jeblad_WMDE: "Rebase and additional fixes"? now, how do i find those fixes? do you want me to review everything again? [14:19:11] unlikely to happen today - or tomorrow. [14:19:55] It was broken after rebase [14:20:24] New patchset: Jeroen De Dauw; "added datavalues extension group" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24807 [14:20:29] ah - if the fixes just fix the rebase, ok. [14:20:55] would be nice to have an easy way to find them, but i don't know a good way to do that :/ [14:30:55] The diff solution in gerrit is crap [14:32:21] seems like something has gone wrong in the last patchset in 24541, it fails as a parent [14:32:22] Change merged: Daniel Kinzler; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24541 [14:32:42] oops.. [14:32:53] O_O [14:32:56] jeblad_WMDE: just merged it. [14:33:01] w3orks for me [14:33:07] what fails? [14:33:21] I pull and squash and sees what happen [14:33:23] * DanielK_WMDE is afk for 10, family calls [14:36:46] New patchset: John Erling Blad; "(Bug 40476) Add a type marker to the entity output by the API" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24808 [14:48:27] New patchset: Jeroen De Dauw; "Increment to 0.2 alpha" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24810 [14:57:07] Lydia_WMDE: why did you put the stickers on your laptop upside down? noob! :p [15:27:38] what do you do if for interwiki links a page that is linked to is moved? [16:05:03] shouldn't the client be reset as well as the repo? [18:18:15] New review: Daniel Werner; "That's ok, just waiting for the "official" ok to increment the number to 0.2 here" [mediawiki/extensions/Wikibase] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/24810 [18:18:55] hi Lydia_WMDE [18:19:34] hey jeremyb :) [18:19:55] jeremyb: is there a git repo for wikidata.org content that i can clone, change and push changes to? [18:20:04] yes [18:20:04] jeremyb: also I have a favicon now \o/ [18:20:12] yay! i was going to ask you that anyway ;) [18:20:17] :D [18:20:32] can you point me to the repo? then i will have a go at it tomorrow [18:20:49] sure. i'm happy to make changes for you too [18:21:07] i need to experiment a bit tbh [18:21:13] want to add the logo for example [18:21:16] and change the text [18:22:14] (and secretly I want the "i used gerrit" badge :P [18:22:25] but shhhhh [18:23:09] Lydia_WMDE: do you want to be shot or what? o.O [18:23:20] * Lydia_WMDE pets jeremyb [18:23:21] e [18:23:28] * Lydia_WMDE pets JeroenDeDauw [18:24:08] Lydia_WMDE: can you answer my question about what happens if a client page moves (interwiki/stage 1)? [18:24:47] giftpflanze: hi! i think at the moment nothing happens - needs to be adjusted manually [18:24:56] later more magic will happen afaik [18:25:15] * jeremyb pets Lydia_WMDE! [18:25:34] * Lydia_WMDE purrs [18:25:37] hope so [18:26:35] lol :D [18:27:54] Lydia_WMDE: here you go: operations/mediawiki-config.git:docroot/www.wikidata.org [18:28:13] jeremyb: thx! will poke it hard tomorrow [18:28:22] Lydia_WMDE: i'll be around for the next ~2 hrs [18:28:32] nah - i'll not touch it now [18:28:38] k [18:28:39] too late and i deserve some tv time :P [18:28:55] you can do TV at the same time! [18:29:27] hah [19:12:49] New patchset: Jeroen De Dauw; "Added DataType to Property" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24829 [19:18:49] New patchset: Jeroen De Dauw; "fix version numbers and some naming issues" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24868 [20:05:05] New patchset: Jeroen De Dauw; "Match naming changes in DataValues [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24882 [20:05:07] New patchset: Jeroen De Dauw; "Changed DataValue namespace to DataValues and implemented NumberValue [DO NOT MERGE]" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/24883 [20:10:12] New patchset: Daniel Kinzler; "fix pollForChanges skipping changes" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24884 [20:12:42] New patchset: Daniel Kinzler; "fix item id in debug output" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24885 [20:18:26] New patchset: Daniel Kinzler; "Make pollForChanges terminate optionally." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24887 [20:22:38] New patchset: Daniel Kinzler; "Inject wikibase updates into recentchanges table." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24889 [20:25:46] New review: Jeroen De Dauw; "Hmm. It's less broken then before like this. Still needs a bug to have it fixed properly." [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/24884 [20:28:14] New review: Daniel Kinzler; "@jeroen: I agree in general, but your objection on principle should not block a bug fix. File a bug ..." [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/24884 [20:30:14] New patchset: Daniel Kinzler; "Inject wikibase updates into recentchanges table." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24889 [20:33:39] New patchset: Daniel Kinzler; "Clients should not share tables per default." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24891 [20:37:19] JeroenDeDauw: Are you active? [20:37:27] Probably a stupid question [20:37:57] Let me rephrase; is you in the office or are you playing gp [20:37:59] O? [20:38:11] playing go? [20:38:21] :P [20:40:43] DanielK_WMDE: I find very few deprecated functions in our code but it seems like mediawiki itself uses them due to our calls [20:41:50] jeblad_WMDE: really? where? deprecated functions should make tests fail. [20:42:05] Change merged: Reedy; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24891 [20:42:48] I updated GlobalFunctions with wfDeprecated to trace down the calls [20:42:58] And yeah, tests fail [20:45:08] New patchset: Jeroen De Dauw; "Revert "Clients should not share tables per default."" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24909 [20:45:15] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24909 [20:45:39] New review: Jeroen De Dauw; "Reedy: wtf?" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24891 [20:45:58] Feck [20:46:09] * aude hides ;) [20:46:11] It slipped in the middle of my .gitreview commits [20:46:38] Reedy: oh, you did not even want to merge it? :p [20:46:42] nooope [20:46:43] Even the best among us can make errors,.. :D [20:46:53] >_> [20:46:55] Numbers are hard [20:47:20] hehe [20:48:13] Reedy: are you code reviewing the extenstions? [20:48:27] JeroenDeDauw: Is it possible for ChangeRow::getAge to ever get an invalid timestamp from getField? [20:48:50] New patchset: Jeroen De Dauw; "Re-apply "Clients should not share tables per default." fd0a81a12f6d4ac2953e2b7fa0c9d492ea5b5ee6" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/24910 [20:49:43] New review: Jeroen De Dauw; "See comments on original patchset" [mediawiki/extensions/Wikibase] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/24910 [20:51:33] jeblad_WMDE: sure, if the set field is invalid