[00:00:00] and I hate mailing lists [00:02:08] Merlissimo: https://en.wikipedia.org/w/api.php?action=paraminfo&modules=wbgetitems [00:02:32] Maybe [00:02:53] Now onto more important things: ihazadmin? [00:02:55] Oh, sod that [00:02:56] http://wikidata.org/w/api.php?action=query&meta=siteinfo&siprop=extensions [00:03:25] Merlissimo: Ask them to add the NS info to siteinfo http://wikidata.org/w/api.php?action=query&meta=siteinfo&siprop=namespaces [00:03:29] As an extra property maybe [00:07:07] that's a good idea, Reedy [00:07:22] Merlissimo: I do get them from time to time ;) [00:57:20] Is there not a bot going through backfilling labels on already existing pages? [01:13:44] Who do we think will be first to add selective importing of IWs from a list? [01:14:05] Mainly to exclude ones that redirect to other pages that have articles... but are distinct in other projects [01:53:28] https://www.wikidata.org/w/index.php?title=Special:CreateItem&site=enwiki&page=Nexus+4 [01:53:37] What's the point of those last 2 parameters when they don't get used? :/ [02:13:31] Reedy: It seems you're going to be right about our page-count exceeding enwiki's, especially when we enable the functionality to create properties [03:16:08] I accidentally created a duplicated item - how do I flag it for deletion? [03:16:29] GabrielF: https://www.wikidata.org/wiki/Wikidata:Requests_for_deletions [03:16:37] thx [03:16:40] np [03:18:06] I'm not sure why the original didn't come up in search though [03:18:23] Search is still buggy [03:18:31] What do you use to search, GabrielF? [03:18:33] Recently it hasn't been working as well... [03:18:40] Riley: just the search bar [03:18:46] GabrielF: Use http://www.wikidata.org/wiki/Special:ItemByTitle [03:18:47] Riley: https://www.wikidata.org/w/index.php?search=Jules+Verne&title=Special%3ASearch [03:19:30] Riley: that doesn't work if the page doesn't have any interwikis/sitelinks [03:19:38] suboptimal solution, IMO [03:19:53] How many pages don't have sitelinks though? [03:20:18] We can't tell; There's no list of pages without sitelinks... [03:20:48] http://www.wikidata.org/wiki/Special:ShortPages [03:20:56] Riley: https://www.wikidata.org/wiki/Q48957 <- here's one [03:21:05] that you created [03:21:09] oops [03:21:19] :P [03:21:20] I am handling Q48957 leave alone :) [03:24:10] Thanks for letting me know, fixed. [03:28:44] jeblad_WMDE: are you the one w/ translation admin? [03:29:12] Yeah, as staff [03:30:03] Only staff member who has it, well aren't you special :) [03:30:39] jeblad_WMDE: since the other (sole) translation admin is not online atm, care to mark http://www.wikidata.org/wiki/Template:User_admin for translation? [03:42:09] * Hazard-SJ wonders when people will stop complaining about pages needing archival and actually do something like support the archivebot request [03:46:49] Hazard-SJ: maybe when they know it exists [03:46:52] :) [03:47:29] PiRQuadratus: Maybe they should try looking then :P [04:13:41] phew [04:13:53] I just finished fixing editnotices [05:17:02] i'm getting an error with the import interwiki gadget [05:17:20] "Save error: The external client site did not provide page information." [05:17:26] http://www.wikidata.org/wiki/Q49149# [05:18:54] hm [05:19:39] the es interwiki link is a redirect, don't know if that matters [05:19:41] GabrielF: works for me [05:19:43] https://www.wikidata.org/w/index.php?title=Q49149&diff=450680&oldid=450675 [05:20:08] hm maybe there was a session error or something [05:20:43] GabrielF: I suggest [[User talk:Tpt]] [05:20:43] 10[42] 10https://www.wikidata.org/wiki/User_talk:Tpt [05:23:15] GabrielF: even better: https://www.wikidata.org/wiki/User_talk:Tpt/slurpInterwiki.js [05:28:58] bai [06:19:31] Reedy: why does http://www.wikidata.org/wiki/Special:Statistics show -1 for active users? [06:19:39] is it because that stat is disabled? [06:58:48] can someone take a look at [[Anaheim, California]] on enwiki? there's an interlanguage link to a polish template that's showing up but I can't find it in the interlanguage links section (maybe a template is generating it?) [06:58:48] 10[43] 10https://www.wikidata.org/wiki/Anaheim%2C_California [07:04:57] never mind I figured it out [07:49:20] hi [09:19:34] New review: Henning Snater; "Just two or three minor problems (which you probably would have noticed if you had tests already). R..." [mediawiki/extensions/DataValues] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/33183 [09:23:44] * jeblad_WMDE must get his as moving [09:24:04] *ass [09:26:08] DanielK_WMDE: poke [10:34:53] New patchset: Aude; "refactor changes metadata stuff [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33319 [10:39:04] DanielK_WMDE: kann i modify the json page content directly? [11:04:24] Merlissimo: only via the import interface. not recommended. [11:04:33] Merlissimo: wbedititem is pretty close though [11:04:50] err, wbeditentity, sorry. [11:05:08] the data parameter is pretty much what the plain json would be [11:05:43] but how i know that this page content is a wikidata entity? [11:05:52] json=wikidata? [11:06:03] no [11:06:29] Merlissimo: ask the api. revisions and pageinfo should report the contentmodel. [11:06:45] no, only contentformat [11:06:51] http://de.wikidata.org/w/api.php?action=query&titles=Q25173&prop=info|revisions|langlinks&rvprop=content [11:07:26] Merlissimo: https://de.wikidata.org/w/api.php?action=query&titles=Q25173&prop=info|revisions|langlinks&rvprop=content|contentmodel [11:07:45] i guess content should imply contentmodel, because it's impossible to interpret the content without knowing the model [11:07:48] ah [11:08:33] the contentformat=json could be used for other things as well. think gadget definitions, image metadata, etc. [11:09:13] contentmodel tells you what structure to expect. contenformat tells you how to deserialize. [11:10:13] i did not know the new rvprop value [11:10:37] i'll make it implicite [11:12:41] New patchset: Tobias Gritschacher; "fixed failing test in TermCacheTest" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33323 [11:20:23] Merlissimo: https://gerrit.wikimedia.org/r/#/c/33325/ [11:22:15] i think that will help others a lot [11:25:42] Abraham_WMDE: before i forget: https://bugzilla.wikimedia.org/show_bug.cgi?id=42065 should go into the status mail to wmf. it's not urgent, but important to get some feedback on this. [11:25:58] one part of the change propagation discussion hinges on if and how this is implemented [11:27:53] New review: Jeroen De Dauw; "You removed the 42 at the end!!! :<0" [mediawiki/extensions/Wikibase] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/33323 [11:45:06] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33323 [11:46:06] New patchset: Daniel Werner; "Created $.valueview and sub-system for creating DataValue edit view widgets" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/33183 [12:17:27] New patchset: Henning Snater; "Using queue to synchronize wikibase.Api QUnit tests" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33331 [12:26:04] New patchset: Aude; "cleaning up some code in wikibase client hooks" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33086 [12:26:04] New patchset: Aude; "refactor changes metadata stuff [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33319 [12:27:44] New patchset: Aude; "refactor changes metadata stuff [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33319 [12:47:34] New review: Daniel Kinzler; "Works and looks good, just would like some additional input on whether we should keep the old name C..." [mediawiki/extensions/Wikibase] (master); V: 1 C: 1; - https://gerrit.wikimedia.org/r/32921 [12:48:29] New review: Daniel Kinzler; "as per commit message" [mediawiki/extensions/Wikibase] (wikidata-wmfphase1beta); V: 0 C: -2; - https://gerrit.wikimedia.org/r/31849 [12:48:39] New review: Daniel Kinzler; "as per commit message" [mediawiki/extensions/Wikibase] (master); V: 0 C: -2; - https://gerrit.wikimedia.org/r/33319 [12:53:01] heya! who can I bribe to merge I344d7681? [12:53:05] I'd love to demo that today... [12:53:35] johl: where's john and jeroen? I was planning to pester them about this [12:55:05] ..humpeti..humpeti.. [12:55:11] you called me master? [12:55:45] ^-- DanielK_WMDE [12:55:48] * jeblad_WMDE is deep inside a scary fantasy [12:55:48] :P [12:55:59] hey jeblad_WMDE [12:56:14] care to look at I344d7681 for me? I'd love to see it merged for the demo [12:56:28] anything i can bribe you with? [12:56:29] (from now on i will call jeblad_WMDE "igor") [12:56:48] johl: hahaha! my thought exactly! i just didn't want to say it ;) [12:57:26] hm, I would like to merge some stuff about permissions.. [12:57:52] the queue seems pretty empty... add me on it, i'll have a look [12:58:02] jeblad_WMDE: does that fix the failures? :) [12:58:05] i just want to have the merge conflict resolution finally in :) [12:58:15] oops - afk for a minute [12:58:27] But it can wait if its more important,.. ie. permissions are'nt that important [12:59:03] I would really like to fix some stuff about the permissions, but it will take more time than an hour I think [12:59:16] AnjaJ_WMDE: is there any permission bugs? [12:59:58] New patchset: Aude; "refactor changes metadata stuff [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33319 [13:00:02] there were several bugs due to permissions (according to jeroen) [13:00:39] filed bugs, not fluffy bugs [13:00:56] Change abandoned: Aude; "squashed" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33086 [13:01:39] DanielK_WMDE: is the general direction i'm going with https://gerrit.wikimedia.org/r/#/c/33319/ okay to you? [13:02:01] trying to split item change stuff into subclass of entitychange and then [13:02:07] handlie it in particular way on the client [13:04:40] aude: i'm wondering if we really need separate subclasses. do they actually implement different functionality or contain different data? [13:04:52] aude: wouldn't somethign like getChangeType be enough?= [13:05:15] (could actually be the type id also use in the logging table) [13:05:51] aude: to answer your question: these should be handled separately on the client, sure - the question is just how to best achieve this. [13:08:05] aude: i'm not sure what ChangeRevision is for... isn't the Change object itself already representing the change, i.e. all relevant metadata, including the content diff, rc info, etc? Why, conceptually, should this be split into separate classes? [13:08:17] DanielK, I got errors [13:08:37] jeblad_WMDE: you always do :) post them somewhere. [13:08:41] and tell me how to reproduce [13:08:58] New review: John Erling Blad; "PHP Warning: require(/home/john/Workspace/Wikibase/lib/tests/phpunit/site/SiteObjectTest.php): fail..." [mediawiki/extensions/Wikibase] (master); V: -1 C: 0; - https://gerrit.wikimedia.org/r/32586 [13:09:25] Possibly a rename.. [13:10:02] DanielK_WMDE: i can see about doing the change with out change revision stuff [13:10:04] Or, .. I just pulled, shall check master [13:10:19] i'm getting this: [13:10:21] Fatal error: require(): Failed opening required '/DATA/var/www/daniel/wikidata/extensions/Wikibase/lib/tests/phpunit/site/SiteObjectTest.php' (include_path='/DATA/var/www/daniel/wikidata/tests/phpunit:.:/usr/share/php:/usr/share/pear') in /DATA/var/www/daniel/wikidata/includes/AutoLoader.php on line 1155 [13:10:29] on any test I run [13:10:46] master passes it seems [13:11:01] concerned about the wikibase client hooks getting huge and would like to separate some of the logic for different types of entities and changes [13:11:26] aude: yes, separating sounds good. but i want to understand the design, and the reasoning behind it. [13:11:52] i'm not arguing against your proposal - just saying I don't currently see why you did it that way. [13:12:28] it would make it easier, for example, provide the recent change comment [13:12:32] AnjaJ_WMDE: do you know anything about the SiteObjectTest thing? I guess that test was moving into core, but apparently, it's still references from wikibase [13:12:42] (probably will go in the revision table on the client) [13:12:45] isn't everyone seeing this error? [13:12:46] as a message [13:13:21] aude: any reason not to implement this in the EntityChange or even Change directly? [13:13:37] SiteObjectTest is in core.. [13:13:47] Why is it listed in lib? [13:13:47] aude: i'm trying to understand what the ChangeRevision is supposed to represent conceptually, and how that is different from what Change itself represents. [13:14:01] jeblad_WMDE: no idea, but it fails all tests for me [13:14:05] am i the only one? [13:14:15] JeroenDeDauw: [13:14:21] Fatal error: require(): Failed opening required '/DATA/var/www/daniel/wikidata/extensions/Wikibase/lib/tests/phpunit/site/SiteObjectTest.php' (include_path='/DATA/var/www/daniel/wikidata/tests/phpunit:.:/usr/share/php:/usr/share/pear') in /DATA/var/www/daniel/wikidata/includes/AutoLoader.php on line 1155 [13:14:23] ??? [13:15:01] is it just me? [13:15:22] jeblad_WMDE: want to merge it? [13:15:26] DanielK_WMDE: sorry, don't know [13:15:39] It insists on a test in the extension, but this is moved to core [13:15:39] hm... [13:15:55] i'm seeing it on the feature branch, should probably try master [13:15:59] maybe it just needs a rebase [13:16:10] for the change revision, it'd be nice to have a super lightweight representation of an external revision [13:16:36] and stuff like remote users [13:16:50] rather than dump everything into an array [13:16:52] It does need a rebase, and it have a merge conflict [13:17:03] ah, fixed in master, rebase works [13:17:08] JeroenDeDauw: never mind, sorry [13:17:22] jeblad_WMDE: already fixed that, will push an amend in a minute [13:17:31] * DanielK_WMDE is waiting for tests to compelte [13:18:05] * jeblad_WMDE is better than tests, a look and the code fails [13:18:07] still figuring out the details and how the code would work [13:18:12] New patchset: Daniel Werner; "Created $.valueview and sub-system for creating DataValue edit view widgets" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/33183 [13:19:42] aude: hm... lightweight representation of a remote revision sounds good. but to be useful, it should have all the nice getters etc. If it just wraps an array, it seems pretty useless. [13:20:12] DanielK_WMDE: agree [13:20:19] one step at at time [13:20:26] aude: also... where does the recentchanges info go? do we need *another* class for that? [13:20:46] I'd put that also in the remote revision class. Optionally. [13:21:00] no, i think how we're doing recentchanges in the client has to change slightly [13:21:05] actually... is there anything we need in RC that wouldn't be in the revision? [13:21:08] we need to insert a null revision in the client revision table [13:21:20] no, i'm eliminating a bunch of the rc stuff [13:21:26] that's not an alternative, that would be in addition to the rc injection [13:21:27] the metadata and [13:21:33] right [13:21:36] but yes, that has been on the todo list for a while. [13:21:49] but connecting the client revision and client rc needs to be done [13:22:01] upload? [13:22:15] e.g. rc_curid = page_id [13:22:16] aude: but to be useful in any way, we then need to store the item revision id along with the page revision, so we can create combined diffs for data and text [13:22:23] page_id is easier to reference [13:22:23] i think that can and should wait [13:22:31] let's for now just get the recentchanges stuff right [13:22:36] yes, we need item revision id [13:22:45] sure [13:23:00] until then, we can just use the page's current revision in the rc table [13:23:09] ...? [13:23:29] And gerrit died [13:23:31] what's in the rc table doesn't matter that much yet [13:23:36] New patchset: Daniel Kinzler; "(Bug 39836) Auto-resolve edit conflicts." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/32586 [13:23:38] the formatter re-writes it [13:23:49] New patchset: Henning Snater; "Implementing getEntities in wikibase.Api" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33335 [13:23:50] New patchset: Henning Snater; "wikibase.Api setLabel QUnit test" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33336 [13:24:03] New review: Daniel Kinzler; "rebased" [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/32586 [13:24:12] jeblad_WMDE: new version is up [13:24:26] tests running [13:25:12] basically, all the rc_info can be derived from revision info in the wb_changes metadata [13:25:42] Change merged: Henning Snater; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/33183 [13:25:53] I wrote a fuzzy similarity measure last night [13:26:23] oh [13:26:25] I'm toying with the idea of using it to set tags for abuse filter [13:26:35] * aude hungry ... back in a bit [13:27:07] And adding the similarity to apigetentities [13:27:58] New patchset: Henning Snater; "Minor API documentation update" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33337 [13:28:39] I think we should have chocholate when someone adds more tests.. [13:28:44] This takes time... [13:28:59] Change merged: Daniel Werner; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33337 [13:30:45] If I merge I must change the only patchset I have for demotime, and I cant fix it in 30 min! [13:30:49] O_O [13:31:58] jeblad_WMDE: oh? what does my patch conflict with? [13:32:14] some permission stuff [13:32:17] not important [13:35:12] I tried to manually trigger an edit conflict.. [13:35:26] And it failed to patch the edit [13:35:49] A loads and edit page C [13:35:59] B loads and edit page C [13:36:16] A makes an additional edit on page C and fails [13:36:25] That should work [13:38:15] REVIEW MY STUFF https://gerrit.wikimedia.org/r/#/q/%28+project:mediawiki/extensions/Wikibase+OR+project:mediawiki/extensions/Diff+OR+project:mediawiki/extensions/DataValues+%29+status:open,n,z [13:38:20] [13:38:48] New patchset: Jeroen De Dauw; "Added value formatter factory" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/33181 [13:38:58] New patchset: Jeroen De Dauw; "Added value validator factory" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/33176 [13:39:41] New review: John Erling Blad; "Tested in browser with one logged in user and one anonymous. The following case" [mediawiki/extensions/Wikibase] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/32586 [13:39:45] oh. the queue for wikibase seemed to only contain ui/js stuff. i guess i should have looked at DV too [13:39:54] but now i want to know what jeblad found [13:40:17] It passes tests, but the manual edit fails [13:40:19] jeblad_WMDE: whether that should work depends on whether A and B are editing the same thing [13:40:37] Point [13:41:43] jeblad: can't reproduce [13:43:13] If I do not edit the same field I still get a failure on the third edit [13:44:06] jeblad_WMDE: i don't. can you give me an exact stript of who is editing what and when? [13:44:12] New review: John Erling Blad; "Also" [mediawiki/extensions/Wikibase] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/32586 [13:44:44] one thing i noticed though: if both remove the same thing, there is a conflict, even though there isn't "really" a conflict [13:44:56] New review: Tobias Gritschacher; "Also the following is not working as expected:" [mediawiki/extensions/Wikibase] (master); V: 0 C: -1; - https://gerrit.wikimedia.org/r/32586 [13:46:36] DanielK_WMDE: I wonder if it something weird with aliases.. I think I saw something when I messed around with patching weeks ago [13:48:11] I wonder if the UI sends an add with one element and an empty remove [13:50:06] New review: Daniel Kinzler; "Seems fine at a first look, but I'd need to first dig into the DV architecture before I can approve ..." [mediawiki/extensions/DataValues] (master); V: 0 C: 1; - https://gerrit.wikimedia.org/r/33181 [13:50:42] jeblad_WMDE: are all three edits you made to aliases? what exactly did you edit in which order? [13:50:50] i tried with aliases, it seemd to work fine [13:50:58] are both your users using the same languageß [13:50:59] ? [13:51:11] label - description - alias and it fail [13:53:27] what - A edits label, B edits description, A edits alias? [13:53:34] what's the exact error message? [13:54:09] works perfectly for me... [13:54:32] A reload page C [13:54:41] B reload page C [13:54:59] A edits label on pageC [13:55:13] (english) [13:55:37] B edits label on page C (Norwegian) [13:55:55] hm, switch those languages [13:56:17] New patchset: Henning Snater; "Basic wikibase.Api setDescription test" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33340 [13:56:40] And A edits label on page C (Norwegian) and have an edit conflict [13:57:29] That should not give an edit conflict I think [13:58:12] Shall I merge and we can discuss later? I think something is wrong in the diff/patching [13:58:16] jeblad_WMDE: hm? [13:58:36] if A edits the label in norwegian after B did the same, sure, there's a conflict [13:58:42] Code freeze in 3... 2... [13:58:55] They edit in different languages [13:59:01] jeblad_WMDE: merge if it doesn't break anything, so i can demo :) [13:59:02] There are still edit conflicts [13:59:22] jeblad_WMDE: according to your script, they were both editing in norwegian [13:59:30] Nope [13:59:39] [14:55] B edits label on page C (Norwegian) [13:59:43] 14:56] And A edits label on page C (Norwegian) and have an edit conflict [13:59:53] no? [13:59:57] New review: John Erling Blad; "This code have issues, and it seems like it does not work properly" [mediawiki/extensions/Wikibase] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/32586 [14:00:04] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/32586 [14:01:28] ok, seeing a conflict now that perhaps shouldn't be there [14:01:33] will investigate [14:01:46] i think we can merge, but perhaps shouldn't be accepted until this issue is resolved [14:02:01] jeblad_WMDE: your script was not describing that case though :) [14:02:02] code freeze as cold as ice [14:02:19] * DanielK_WMDE shivers [14:02:20] AnjaJ_WMDE: You are free to tag. :) [14:02:44] Silke_WMDE: did you shout out code freeze already? ;) [14:03:48] * Silke_WMDE_ is shouting [14:03:53] GONG [14:04:20] :) [14:04:20] * jeblad_WMDE merged changesets with clear bugs and a -1.. [14:04:26] Muahahahahaaa... [14:04:49] jeblad_WMDE: two -1's [14:04:55] hm, didn't see tobi's comment before. wouldn't have recommended to merge with a -1 [14:05:09] TobiasG_WMDE: can't reproduce the problem as described [14:05:09] Two minus one, anybody with a -3? [14:05:12] * DanielK_WMDE is confused now. [14:05:21] One time, do we have a -3? [14:05:21] revert if you feel like it :) [14:05:37] it appears to work as expected in many cases. but apparently, not all [14:05:45] Second time, tthe gangster with the video beamer, will he give a -3? [14:05:54] hehehe [14:06:01] we can demo it [14:06:02] I'm fine with it as long as it does not break anything [14:06:34] DanielK_WMDE: you cannot reproduce the problem I described? hmhm.. [14:07:26] There was changes in Diff and I pulled, perhaps that created problems [14:07:27] TobiasG_WMDE: nope. need more info though. logged in or anon? which languages? etc [14:07:34] oh... [14:07:40] i guess i should pull diff too! [14:08:16] Perhaps we should complain about Jeroen instead.. [14:08:26] Nah, lets take Daniel this time! =D [14:09:55] DanielK_WMDE: User A (logged in) changes description in EN; User B (anon user) is not able to add an alias in EN [14:11:14] TobiasG_WMDE: did the description exist before, or was it created? [14:11:24] were there any aliases before, or is this the first one? [14:12:15] TobiasG_WMDE: huh. ok, i see it now. wtf? [14:12:28] DanielK_WMDE: description existed and was just modified; and no aliases existed before, so User B would try to add the first alias [14:12:54] New patchset: Jeroen De Dauw; "Construct PropertyValueSnak objects in create claim API" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33341 [14:13:30] TobiasG_WMDE: ah! seems like that only happens when creatign the first alias [14:13:33] DanielK_WMDE: if User B is a logged in user (of course different from User A) then it works [14:14:12] New patchset: Jeroen De Dauw; "Construct PropertyValueSnak objects in create claim API" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33341 [14:14:30] New patchset: Henning Snater; "Restructuring wikibase.Api's editEntity" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33342 [14:15:05] DanielK_WMDE: oh, yes you're right [14:15:57] http://bots.wmflabs.org/~wm-bot/dump/%23wikimedia-wikidata.htm [14:16:16] !g is https://gerrit.wikimedia.org/r/#q,$1,n,z [14:16:16] Key was added [14:16:27] !g 33341 | jeblad_WMDE [14:16:27] jeblad_WMDE: https://gerrit.wikimedia.org/r/#q,33341,n,z [14:16:49] !nyan | Henning_WMDE [14:16:49] Henning_WMDE: ~=[,,_,,]:3 [14:17:03] TobiasG_WMDE, jeblad_WMDE: ok, there seems to be an issue when stuff gets first added. will investigate [14:23:25] New patchset: Jeroen De Dauw; "Removed Hashable interface as we can now use the one in DVs" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33343 [14:25:34] New patchset: Jeroen De Dauw; "Removed Immutable interface as we can now use the one in DVs" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33344 [14:30:34] New review: Aude; "This makes the client dependent on DataValues, which should be okay." [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/33344 [14:32:22] grr... can't reproduce in a test case. what gives? [14:33:29] computers suck [14:33:31] Wikibase, Diff, DataValues are tagged 2012-11-14 - anything else? [14:33:36] You see, I'm marvelous when it comes to bugs [14:34:26] If there are none, they will magically pop up from nowhere when I start poking arouind [14:34:34] Reedy: that depends on how you mount the fans [14:34:51] jeblad_WMDE: you must have stopped looking, because it just disappeared o_O [14:35:28] JeroenDeDauw: mounting fans is for rock stars [14:36:44] O_o [14:43:11] New patchset: Tobias Gritschacher; "selenium test to catch bug 42101" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33346 [14:43:53] hm, looks like a bug in Diff::addReversibleOperations [14:46:51] New patchset: Jeroen De Dauw; "Removed Comparable interface as we can now use the one from DVs and added equals method to Snak objects" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33347 [14:46:53] I have some additional weird ideas for diffs, so its nice that you track down the bugs! =) [14:48:28] New review: John Erling Blad; "Will remove the whitspace later" [mediawiki/extensions/Wikibase] (master); V: 1 C: 2; - https://gerrit.wikimedia.org/r/33341 [14:48:28] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33341 [14:48:35] New review: Jeroen De Dauw; "Think the client already was dependent on DataValues" [mediawiki/extensions/Wikibase] (master); V: 0 C: 0; - https://gerrit.wikimedia.org/r/33344 [14:50:30] New review: John Erling Blad; "Gah.. ;)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33341 [14:53:30] New patchset: Jeroen De Dauw; "Drop support for integer argument in snak constructor" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33349 [14:54:42] DanielK_WMDE: oh really? [14:55:24] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33343 [14:55:59] JeroenDeDauw: i think i'm starting to understand what happens. it's related to ItemDiff::newFromItems generating empty MapDiffs. [14:56:18] empty diffs seem to cause problems in some way. [14:56:28] still investigation, the code is a bit confusing [14:57:54] JeroenDeDauw: can you explain how exactly EntityDiffObject::newFromEntities creates an instance? [14:57:55] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33349 [14:58:10] New patchset: Jeroen De Dauw; "Removed Immutable interface as we can now use the one in DVs" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33344 [14:58:46] it seems to use the constructor from ArrayObject or something? [14:58:52] New patchset: Aude; "refactor changes metadata stuff [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33319 [14:59:16] grr. this is frustrating [15:01:00] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33347 [15:01:30] DanielK_WMDE: as you can see, newFromEntities creates an array of DiffOp objects [15:01:42] And it feeds this to the constructor of MapDiff [15:01:51] Which takes an array of DiffOp [15:02:01] Not sure what's unclear there [15:03:05] DanielK_WMDE: hangout? [15:03:13] Change merged: John Erling Blad; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33344 [15:03:16] Abraham_WMDE1: i'm already there [15:03:24] the one linked from the calendar [15:03:31] äh [15:03:51] DanielK_WMDE: do'nt think so, that's only for our weekly call. Silke invited you to hangout [15:04:27] DanielK_WMDE: but you're right there is a hangout link [15:04:39] ...which we should use in the future ;) [15:07:29] New patchset: Jeroen De Dauw; "Move interface to correct place (orginially put it here to avoid merge conflict)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33353 [15:07:38] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33353 [15:16:07] New patchset: Jeroen De Dauw; "test" [mediawiki/extensions/Diff] (master) - https://gerrit.wikimedia.org/r/33354 [15:20:03] Change merged: Jeroen De Dauw; [mediawiki/extensions/Diff] (master) - https://gerrit.wikimedia.org/r/33354 [15:20:23] New patchset: Jeroen De Dauw; "Improved setup and entry point code" [mediawiki/extensions/Diff] (master) - https://gerrit.wikimedia.org/r/33190 [15:20:27] New patchset: Jeroen De Dauw; "Include test commented out for unknown reason" [mediawiki/extensions/Diff] (master) - https://gerrit.wikimedia.org/r/33191 [15:32:55] New patchset: Jeroen De Dauw; "fixed type hint" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33356 [15:33:05] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33356 [15:41:57] New patchset: Jeroen De Dauw; "remove use of unused use statements" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33357 [15:48:47] DanielK_WMDE: I will has a look at the diff thing, would be stupid if you spend a day on it if I can find it in a few mins [15:50:38] New patchset: Henning Snater; "wikibase.Api QUnit test for setting and removing site links" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33359 [16:07:46] New patchset: Jeroen De Dauw; "Added extra test to verify MapDiff correctly handles structures such as the aliases of Wikibase Entity objects" [mediawiki/extensions/Diff] (master) - https://gerrit.wikimedia.org/r/33374 [16:08:08] DanielK_WMDE: looks like the issue is not in Diff: https://gerrit.wikimedia.org/r/#/c/33374/ [16:09:37] JeroenDeDauw: still constructing a test case. looks like ItemDiffTest has been broken forever, and wasn't executed for some reason [16:09:43] now fixing it and adding a test case for this [16:09:57] :D [16:10:04] great those broken test cases no? :p [16:14:43] New patchset: Aude; "re-add wbc-after-page-move message in client" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33375 [16:15:08] DanielK_WMDE: you got some nice variable names in those tests btw [16:15:54] hm? in which tests? [16:16:21] nice of translate bot to revert my messages :( [16:18:43] New patchset: Henning Snater; "wikibase.Api setAliases QUnit tests" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/33376 [16:33:05] DanielK_WMDE: EntityDiffTest and deriving stuff [16:33:15] ie the broken ones [16:33:41]