[00:35:43] is there any rough estimate on when queries will be ready? e.g. a few weeks, a few months? [00:57:19] Emw: idk, i do know that the QueryEngine was recently merged into master though :) [08:35:53] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63854 [08:37:56] New patchset: Henning Snater; "(bug 48145) Introducing listrotator widget to set time precision" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62585 [08:39:53] Change merged: jenkins-bot; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/58879 [08:53:25] New patchset: Tobias Gritschacher; "Remove getParsers from DataType interface" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/61599 [08:55:13] New review: Tobias Gritschacher; "$parsers parameter could be removed from the constructor." [mediawiki/extensions/DataValues] (master) C: 2; - https://gerrit.wikimedia.org/r/61599 [08:55:28] Change merged: jenkins-bot; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/61599 [08:56:34] New review: Tobias Gritschacher; "can this be abandoned, since this file does not exist anymore..?" [mediawiki/extensions/Wikibase] (master) C: -2; - https://gerrit.wikimedia.org/r/61591 [09:03:23] Abraham_WMDE: hangout? [09:03:34] not that i had much to say... [09:04:04] I'm in [09:15:36] New patchset: Henning Snater; "(bug 48145) Introducing listrotator widget to set time precision" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62585 [09:15:56] New patchset: Henning Snater; "(bug 48145) Simplified input extender widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62586 [09:16:18] New patchset: Henning Snater; "(bug 48145) Applying input expander widget to TimeInput expert" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62152 [09:16:27] New patchset: Henning Snater; "(bug 48145) Introducing listrotator widget to set time precision" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62585 [09:16:32] New patchset: Henning Snater; "(bug 48145) Simplified input extender widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62586 [09:18:15] New patchset: Henning Snater; "(bug 48145) Close icon for input extender" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62587 [09:23:48] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Added listrotator to select calendar" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63640 [09:23:57] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Applied toggler widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63641 [09:25:40] New patchset: Henning Snater; "(bug 48145) Input extender: Triggering "animationstep" event" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63835 [09:41:43] DanielK_WMDE: I have a string with the DB content, i.e. the JSON object from the DB. How do I turn it into RDF? [09:41:51] i guess, first turn it into an entity [09:41:57] and then turn the entity into rdf, right? [09:42:03] what classes are relevant? [09:43:42] Denny_WMDE: usually, you'd already have an Entity object. Where do you have the string from? [09:43:46] from the dump? [09:46:00] Denny_WMDE: anyway. rough flow: $entity = EntityFactory::newFromBlob(); RdfSerializer::serializeEntity. [09:47:16] New patchset: Henning Snater; "(bug 48145) Simplified input extender widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62586 [09:47:17] New patchset: Henning Snater; "(bug 48145) Close icon for input extender" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62587 [09:47:23] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Added listrotator to select calendar" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63640 [09:47:25] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Applied toggler widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63641 [09:47:28] New patchset: Henning Snater; "(bug 48145) Input extender: Triggering "animationstep" event" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63835 [09:48:39] Denny_WMDE: it would be a bit nicer to construct a Revision object from the ionfo in the dump, and get the EntityContent (and, in turn, the Entity) from that. But I don't quite remember how to do that. [09:48:43] DanielFriesen: yes, from the dump [09:48:59] sorry wrong daniel [09:49:04] DanielK_WMDE: yes, from the dump [09:49:10] New patchset: Henning Snater; "(bug 48145) Adding calendar hint message" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63859 [09:49:26] the whole revision? that sucks, that's much more overhead [09:49:50] not much, but constructing a Revision object sucks. [09:49:59] well, then let's not do it :) [09:50:06] The advantage is that we can have alkl the metaq info and can include the revision ID etc in the output [09:50:08] so, how do I create an entity? [09:50:21] do we need the revision ID for the output? [09:50:35] no, but it's useful to have it. anyway. [09:50:46] like i said above: $entity = EntityFactory::newFromBlob() [09:51:02] sweet thanks, i missed that [09:51:08] New patchset: Daniel Werner; "enhancement of time.Time.validate, making sure required fields for given precision are set" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63991 [09:53:58] Danwe_WMDE: can you has a look at this code? https://gerrit.wikimedia.org/r/#/c/63967/3 _without_ checking my comments on PS1 first [09:59:10] Denny_WMDE: the $format parameter should come from the dump, btw [09:59:36] i can't just take null? [09:59:40] nope [09:59:41] where in the dump? [09:59:59] iirc, the revision has a content-format attribute (or member) [10:00:31] you can also just fake it and use the CONTENT_FORMAT_JSON constant - and hope the dump nerver contains anything different [10:00:59] i'll do that for now :) [10:01:12] put a //TODO there [10:01:39] text/x-wiki ? [10:01:47] yea, that one [10:01:51] ok [10:01:54] oh, i just looked again: null will work [10:02:04] pity it comes *after* the content [10:02:09] ...but it's not recommended. you should always tell it what's actually in the dump [10:02:36] not wikitext but the format one [10:02:40] Denny_WMDE: yea... my mistake, i'm afraid. [10:02:51] JeroenDeDauw: sure [10:03:15] sadly, the xml schema for the dump specifies the order, so changing the order now would (technically) break B/C [10:03:32] i cursed over that before :) [10:03:49] New patchset: Henning Snater; "(bug 48145) Handling "animationstep" event in value variation" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63836 [10:04:22] but, uh... how are you processing the dump? [10:05:34] with a php script? [10:05:42] using bzfile [10:05:50] streaming over it [10:05:56] how else? :) [10:06:27] Denny_WMDE: by using the code we already have for importing xml dumps. it's a bit gnarly, let me look again how exactly it's used [10:06:44] true, would have worked :) [10:08:44] $source = new ImportStreamSource( $handle ); [10:08:46] $importer = new WikiImporter( $source ); [10:09:27] $importer->setRevisionCallback( $yourCallback ); [10:09:42] ...set all the other callbacks to null (or a dummy, if needed) [10:10:10] $importer->doImport() [10:10:14] Denny_WMDE: ---^ [10:10:40] adding a todo [10:10:54] Denny_WMDE: your callback will then get a WikiRevision object as its first param (that's not a Revision object!) [10:11:17] this will take care of getting you the format, and dealing with any other stuff regarding the xml format [10:12:46] Denny_WMDE: you can then use WikiRevision::getContent(), you don't need the factory. [10:12:51] much cleaner and more robust [10:13:27] ah, that is actually sweet [10:13:30] ok, replacing [10:13:55] Denny_WMDE: note that the dump may contain multiple revisions of the same entity. not sure if there's a simple way to find the current one [10:14:15] New patchset: Daniel Werner; "[time.js] Added tests for time.Time.parse" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64035 [10:14:17] no, i am taking the current dump [10:14:21] (ideally, the revision tag would simply have a "latest" attribute, but i don't think it does) [10:14:39] the current dump has only one revision per page [10:15:03] the history dump contains the multiples, but also newest first every time, which is neat [10:15:06] yes, but ideally, the code should be robust against someone using somethign else as input. [10:15:34] anyway [10:15:51] make sure you set *all* other callbacks to null; otherwise, things get written to the database [10:16:05] (WikiImporter is a horrible tangle) [10:16:43] New review: Daniel Kinzler; "can't reproduce the issue locally off hand. will keep poking" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63849 [10:20:00] New patchset: Daniel Werner; "[time.js] Added tests for time.Time.parse" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64035 [10:22:30] New patchset: Jeroen De Dauw; "Make the RDF tests not fail when the RDF stuff is not installed" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63660 [10:24:03] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63660 [10:33:55] aude, JeroenDeDauw, hashar: where can I see the config used for jenkin's mwext-Wikibase-client-tests job? [10:34:10] the failure of https://gerrit.wikimedia.org/r/#/c/63849/ is rather odd [10:34:32] this fails: /srv/ssd/jenkins/workspace/mwext-Wikibase-client-tests/extensions/Wikibase/client/tests/phpunit/includes/api/ApiClientInfoTest.php:64 [10:34:51] huh [10:34:54] indicating that Settings::get( 'repoArticlePath' ) returns null [10:35:20] i can't reproduce it locally [10:35:44] DanielK_WMDE: at first I will add an assertNotEmpty() [10:35:57] well the test is wrong for one [10:36:05] * aude fix [10:36:09] New review: Jeroen De Dauw; "We already have the SIteLink object to represent site+page." [mediawiki/extensions/Wikibase] (master) C: -2; - https://gerrit.wikimedia.org/r/63967 [10:36:22] hm... i'm setting the value explicitly in my local install. will play with this [10:36:39] DanielK_WMDE: the conf is generated by jenkins job builder, a python script that uses specifications written in YAML. [10:36:53] uurrr... [10:36:54] DanielK_WMDE: with a labs account, you should be able to browse https://integration.wikimedia.org/ci/job/mwext-Wikibase-client-tests/configure which list the shell commands executed [10:37:08] which is: cd tests/phpunit && php phpunit.php --log-junit "$WORKSPACE/junit-phpunit-allexts.xml" --testsuite extensions [10:37:09] DanielK_WMDE: if you want me to update the test to inject the settings, i can do [10:37:24] hashar: i need to see the LocalSettings [10:37:48] aude: do we even have a good way to inject these settings right now? [10:37:56] i think we can [10:38:06] * aude looks [10:38:16] DanielK_WMDE: look at the workspace : https://integration.wikimedia.org/ci/job/mwext-Wikibase-client-tests/ws/ . The LocalSettings for the last build is https://integration.wikimedia.org/ci/job/mwext-Wikibase-client-tests/ws/LocalSettings.php/*view*/ [10:38:18] you can change the global value, and then change it back after the test [10:38:27] that's an option [10:39:00] hashar: thanks! [10:39:15] aude: a sucky option, but i don't think there's a better one right now [10:39:46] hashar: there are no wikibase settings there at all [10:39:56] this api module would be a good one to come up with a better design to integrate with mw api classes [10:40:13] (maybe that's intentional?) [10:40:20] at the end of LS.php there is require_once( '/var/lib/jenkins/tools/extensions-loader.php' ); which lookup for files under /extensions/ It will include /extensions/Wikibase/Wikibase.php which has some weird configuration :-] [10:40:44] hashar: DanielK_WMDE i think all the configuration magic is done in Wikibase.php [10:40:47] based on the switches [10:40:48] hashar: what's the url to view that? [10:40:50] (evil) [10:41:02] DanielK_WMDE: https://integration.wikimedia.org/ci/job/mwext-Wikibase-client-tests/ws/LocalSettings.php/*view*/ [10:41:18] no, the Wikibase one [10:41:25] over the last 2 weeks or so, Jeroen and I worked on providing a default entry point for Wikibase [10:41:36] i can't see the link [10:41:37] depending on the Jenkins Job name, it will load a different set of files [10:41:44] but think there's nothing important there [10:41:58] na the LocalSettings.php is whatever install.php created [10:42:07] Wikibase.php in Wikibase exxtension root is the key part [10:42:11] extension [10:42:29] we could potentially make Jenkins look up for a file named TestSettings.php at the root of the extension [10:42:30] anyway the test is doing it wrong [10:43:07] hashar: basically, i want to know if anything beyond the defaults is applied. the example settings? anythign special? [10:43:19] * aude does not think so [10:43:25] for the repo, i think example settings are used [10:43:34] New review: Jeroen De Dauw; "(1 comment)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63849 [10:44:04] yeah Wikibase.php does require_once __DIR__ . '/Query/ExampleSettings.php'; [10:44:22] aude, hashar: ok, if i use defaults only, without any local config, the test fails [10:44:26] i'll investigate [10:44:29] but for the client yes, there does not seem to be any specific settings [10:44:43] Change abandoned: Jeroen De Dauw; "yup" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/61591 [10:46:31] ok, thanks hashar [10:48:06] ooops?! [10:48:24] hashar: $wgArticlePath appears to be "false"... [10:48:27] is that intentional? [10:48:43] while LocalSettings.php is evaluated, i mean [10:49:32] oh, grrr. DefaultSettings has $wgArticlePath = false; [10:49:56] apparently, it's initialialized only after all settings have been loaded [10:50:04] whaaa! [10:50:14] great, need to hook into setup to fix this [10:50:15] but [10:50:19] this wasn't changed by my test [10:50:24] err, by my patch [10:50:31] it should already be broken?... [10:50:45] Easy to find out :) [10:54:08] New patchset: Hashar; "Tweak ApiClientInfoTest::testGetUrlInfo" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64036 [10:54:11] ah, no. [10:54:18] now the defaults get evaluated much earlier [10:54:24] they kind of have to. [10:54:26] DanielK_WMDE: https://gerrit.wikimedia.org/r/64036 should make the ApiClientInfoTest::testGetUrlInfo a bit more useful [10:54:27] hmpfs [10:54:32] DanielK_WMDE: assertTrue() is evil :-] [10:54:48] hashar: at least without a menaingful message it is, yes [10:54:50] if you rebase your changes on top of it, that will give you a nicer message hopefully [11:02:44] hashar: well, i wonder how to fix the actual issue... [11:03:56] actually, screw it. [11:04:08] New patchset: Aude; "Inject settings into api client info function" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64037 [11:04:18] defaulting the *repo's* article path to *client's* $wgArticlePath is broken anyway [11:04:20] DanielK_WMDE: my attempt https://gerrit.wikimedia.org/r/#/c/64037/ [11:06:00] aude: hm... where'S the Settings object from the test injected into the API module? [11:06:17] injected into the function [11:06:51] i don't have a wrapper for the entire api module yet [11:07:22] the part that does $this->getResult()->addValue( 'query', 'wikibase', $data ); could be split out [11:07:28] then also test $data [11:08:15] * aude working on another patch [11:09:03] aude: well, the thing is, if the API module uses one settings object to generate the output, and the test uses *another* settings object to check the output, what guarantees us that they will be the same? [11:10:37] New patchset: Daniel Kinzler; "(bug 47610) Use $wgWBXXXSettings instead of $wgWBSettings" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63849 [11:15:02] Change merged: Jeroen De Dauw; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64036 [11:57:07] aude: did you look at the changeHandler tests yet? [11:58:54] maybe DanielK_WMDE had a look at the changeHandler tests ? [11:59:20] New patchset: Henning Snater; "(bug 48145) TimeInput expert styling improvements" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64038 [12:01:06] Tobi_WMDE: busy, no [12:04:30] aude: n.P. [12:08:51] New patchset: Henning Snater; "(bug 48145) Handling "animationstep" event in value variation" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63836 [12:09:15] New patchset: Henning Snater; "(bug 48145) TimeInput expert styling improvements" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64038 [12:09:36] Tobi_WMDE: got a pointer? my inbox is flooded [12:09:56] * aude does not think it's simple to fix [12:10:10] what's the issue anyway? [12:10:10] DanielK_WMDE: https://bugzilla.wikimedia.org/show_bug.cgi?id=48453 [12:10:30] these tests are currently excluded on jenkins [12:10:51] hrrrmmm.... [12:10:57] now the question is how to reproduce this... [12:11:31] just run the test, should fail for you too [12:11:37] (on a client) [12:11:41] with sqlite [12:11:43] Tobi_WMDE: nope+ [12:11:45] empty sites table [12:11:51] ah [12:12:02] also on mysql [12:12:07] it's evil to assume the sites table is poupulated or touch it [12:12:12] e.g. insert sites [12:12:24] we did that wrong in a number of places [12:12:33] aude: it's just as evil as any other database fixture [12:12:38] not sure that's the entire problem here [12:12:40] heh :) [12:12:45] aude: and we have a simple method to populate it with test data [12:12:58] aude: it's also failing when the sites table is not empty [12:13:02] well for sites, it could be possible to populate a sites list but not the db [12:13:07] Tobi_WMDE: ok [12:14:25] aude: what do you mean by "but not the db"? [12:14:38] DanielK_WMDE: use the SitesList object? [12:15:04] aude: i guess you could force the global instance [12:15:15] ok, can reproduce it now, with an empty sites table [12:16:29] wait, whut? [12:16:32] how to marge two item? [12:16:46] asiaworldcity: there is currently no way to do that but manually [12:16:57] it's on the todo list, but pretty tricky to implement [12:17:43] thanks [12:18:55] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Applied toggler widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63641 [12:19:23] New patchset: Henning Snater; "(bug 48145) Input extender: Triggering "animationstep" event" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63835 [12:23:23] aude, Tobi_WMDE: well, the Real Fix (tm) would be to just ditch the client side commont handling, and use the Summary class in the client as well. [12:23:56] * aude nods and not simple [12:24:06] I#m not sure I understand [12:24:07] important on our todo list [12:24:34] i can't look right now but maybe when i'm done with this api module thing and some open items [12:24:50] * aude of course wants to make sure comments are not broken [12:26:23] New patchset: Henning Snater; "(bug 48145) Implemented input extender jQuery widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62151 [12:26:23] New patchset: Henning Snater; "(bug 48145) Input extender: Triggering "animationstep" event" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63835 [12:26:23] New patchset: Henning Snater; "(bug 48145) Adding calendar hint message" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63859 [12:26:24] New patchset: Henning Snater; "(bug 48145) Introducing listrotator widget to set time precision" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62585 [12:26:24] New patchset: Henning Snater; "(bug 48145) Simplified input extender widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62586 [12:26:24] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Applied toggler widget" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63641 [12:26:24] New patchset: Henning Snater; "(bug 48145) Applying input expander widget to TimeInput expert" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62152 [12:26:25] New patchset: Henning Snater; "(bug 48145) TimeInput expert styling improvements" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64038 [12:26:25] New patchset: Henning Snater; "(bug 48145) TimeInput expert: Added listrotator to select calendar" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/63640 [12:26:31] flooding :) [12:26:57] New patchset: Henning Snater; "(bug 48145) Close icon for input extender" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/62587 [12:27:31] rebasing... [12:32:17] New patchset: Daniel Werner; "Constructor tests for time.Time and introduction of time.validTimeDefinitions" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64042 [12:32:17] New patchset: Daniel Werner; "added equals function for Time, and using it in DataValue now" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64043 [12:35:59] aude: ok, i kind of fixed it, in a very simple way. but the test will break if there's a non-standard sites table in the DB. [12:36:10] because for *some* things, it does read from the db [12:36:33] New patchset: Tobias Gritschacher; "Selenium tests for entityselector as search-box" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64045 [12:36:36] because Itewm::getSiteLinks() uses global state to construct site objects [12:36:39] can take a look [12:36:43] bad [12:36:49] i'll have a brief look at that [12:37:23] aude: perhaps don't try to have a look at *everything*... [12:38:12] heh [12:39:58] damn. baby alert. i'll push a (partial?) solution in about 40 minutes [12:41:31] New patchset: Tobias Gritschacher; "Selenium tests for entityselector as search-box" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64045 [12:49:01] New patchset: Aude; "Inject settings into api client info module" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64037 [12:53:18] DanielK_WMDE: when you are back https://gerrit.wikimedia.org/r/64037 [12:53:55] sure the approach could be generalised and polished for all our api modules [12:54:25] but think right direction and i'm only considering this one part of code now [13:11:20] New patchset: Tobias Gritschacher; "Selenium tests for entityselector paging" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64047 [13:24:52] New patchset: Henning Snater; "(bug 48145) TimeInput expert refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64049 [13:29:17] New patchset: Jeroen De Dauw; "Work on DescriptionMatchFinder [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64050 [13:30:39] New patchset: Jeroen De Dauw; "Have PropertyValue return the value is contains in getValue rather then itself" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64051 [13:35:14] Denny_WMDE1: https://gerrit.wikimedia.org/r/#/c/64053/ [13:35:26] Denny_WMDE1: https://gerrit.wikimedia.org/r/#/c/64051/ [13:37:34] New review: Aude; "needs a rebase" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64037 [13:39:20] New patchset: Aude; "Inject settings into api client info module" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64037 [13:39:20] Change merged: Denny Vrandecic; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64051 [13:44:31] New patchset: Jeroen De Dauw; "Put PropertyNotFoundException into its own file and added PropertyId field" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64054 [13:44:52] New patchset: Jeroen De Dauw; "Put PropertyNotFoundException into its own file and added PropertyId field" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64054 [13:54:29] New patchset: Daniel Kinzler; "(bug 48453) Set site's local ID in tests." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64058 [13:59:52] New review: Daniel Kinzler; "(1 comment)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63849 [14:07:18] New review: Daniel Kinzler; "Ok, here's the thing: I14422cd85 got a serious flaw. I already had the fix committed locally, but di..." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63845 [14:07:54] JeroenDeDauw: please have another look at https://gerrit.wikimedia.org/r/#/c/63845/ and https://gerrit.wikimedia.org/r/#/c/63849/ [14:08:11] i didn't think this would cause so much confusion. [14:08:44] it's just: A got merged, while I already had A' locally. SO I propose to revert A so A' can be applied cleanly. [14:10:16] New patchset: Henning Snater; "(bug 48145) TimeInput expert refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64049 [14:15:13] New patchset: Aude; "(bug 48473) validate claim guid in api and return error" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63787 [14:17:01] New review: Daniel Kinzler; "Note that the follow-up also fixes the failing test case." [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/63845 [14:18:31] New patchset: Henning Snater; "(bug 48145) List rotator widget refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64062 [14:22:18] hm. who killed gerrit [14:22:19] ? [14:22:33] mark did [14:22:44] guh [14:23:00] ah, it's back [14:23:02] * aude trying to setup the easyrdf submodule to point to gerrit now [14:23:29] do we have easyrdf-lite in gerrit? [14:23:49] we do [14:24:01] i'll send a mail around in a moment [14:24:18] we need to use that, since github is not accessible in the deployment environment [14:25:36] DanielK_WMDE: meh, the importer uses its own WikiRevision class, which does not know about ContentHandler yet [14:27:40] DanielK_WMDE: ignore. it is just my PhpStorm giving me wrong methids [14:35:28] New review: Jens Ohlig; "Too much code duplication from the client. Better to find a place where repo and client can both use..." [mediawiki/extensions/Wikibase] (master) C: -1; - https://gerrit.wikimedia.org/r/62321 [14:36:25] New review: Jens Ohlig; "(1 comment)" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/62321 [14:38:35] New patchset: Jeroen De Dauw; "Work on DescriptionMatchFinder [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64050 [14:42:55] Denny_WMDE1: more chocolate for you https://gerrit.wikimedia.org/r/#/c/64064/ [14:44:26] New patchset: Aude; "Point easyrdf submodule to gerrit repo" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64066 [14:44:48] New patchset: Jeroen De Dauw; "Removed PropertyValue class" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64067 [14:46:44] New patchset: Tobias Gritschacher; "Selenium tests for handling of entities w/o labels in statements UI" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64068 [14:49:12] New patchset: Aude; "Point easyrdf submodule to gerrit repo" [mediawiki/extensions/Wikibase] (mw1.22-wmf4) - https://gerrit.wikimedia.org/r/64069 [14:49:53] Denny_WMDE1: Tobi_WMDE https://gerrit.wikimedia.org/r/#/c/64066/ (easy chocolate) [14:50:04] Denny_WMDE1: https://gerrit.wikimedia.org/r/#/c/64070/ [14:50:20] Denny_WMDE1: and after that one, this one https://gerrit.wikimedia.org/r/#/c/64067/ [14:50:46] not to overwhelm, but i recommend we put this https://gerrit.wikimedia.org/r/#/c/63710/ as a backport [14:50:50] New patchset: Jeroen De Dauw; "Work on DescriptionMatchFinder [DO NOT MERGE]" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64050 [14:51:05] since i suspect the issue may appear in production [14:51:19] might want to run tests again with the new submodule and this [14:51:20] New patchset: Henning Snater; "(bug 48145) TimeInput expert refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64049 [14:51:21] or not [14:51:27] New patchset: Henning Snater; "(bug 48145) List rotator widget refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64062 [14:52:44] aude: not sure about the backport [14:52:55] we can see if the issue appears [14:53:01] Change merged: jenkins-bot; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64066 [14:53:07] aude: what's new with the submodule? [14:53:12] Tobi_WMDE: easyrdf [14:53:14] it is the same code, isnt it? [14:53:16] yes [14:53:21] new home for it [14:53:35] so there is no need to run all of the tests agin [14:53:45] not for the submodule [14:56:36] New patchset: Henning Snater; "(bug 48145) List rotator widget refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64062 [14:58:16] New patchset: Jeroen De Dauw; "Add subject_id field to all property snak tables" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64071 [14:59:33] Denny_WMDE1: also https://gerrit.wikimedia.org/r/#/c/64071/ [15:01:59] Change merged: Denny Vrandecic; [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64067 [15:06:30] DanielK_WMDE: are properties also wikibase-item models? [15:07:16] New patchset: Jeroen De Dauw; "Add subject_id field to all property snak tables" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64071 [15:07:44] Denny_WMDE1: Amended - if you think this should already be all specified correctly now, then also have a look at the other definitions [15:07:51] New patchset: Henning Snater; "(bug 48145) List rotator widget refactoring" [mediawiki/extensions/DataValues] (master) - https://gerrit.wikimedia.org/r/64062 [15:08:24] Denny_WMDE1: no, they would be wikibase-property [15:08:24] ok, will [15:08:38] Denny_WMDE1: use the constants [15:08:40] DanielK_WMDE: ok, thanks [15:08:43] Denny_WMDE1: I guess in particular the claims table [15:09:16] Change merged: Denny Vrandecic; [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64071 [15:09:49] ah, jenkins is failing on the deployment branch :( [15:10:18] PHP Fatal error: Call to undefined method DataTypes\DataType::getParsers() [15:14:50] Tobi_WMDE: any idea ? [15:14:54] * aude running tests [15:15:23] New patchset: Tobias Gritschacher; "Selenium tests for deleted-item handling in statements UI" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64073 [15:15:41] aude: can u send me link [15:15:47] https://integration.wikimedia.org/ci/job/mwext-Wikibase-repo-tests/212/console [15:16:03] https://gerrit.wikimedia.org/r/#/c/64069/ [15:16:52] aude: this got merged today: https://gerrit.wikimedia.org/r/#/c/61599/ [15:16:59] JeroenDeDauw: yeah, they would be some things i do not entirely agree with in schema [15:17:00] but it was not backported (and should not) [15:17:08] how to comment? [15:17:16] hmmmm, ok [15:17:24] so, is jenkins running the wikibase tests with master of datavalues? [15:17:31] i am running all the tests myself [15:17:33] Denny_WMDE1: mail, AFK, make a commit - it all works :) [15:17:36] although not selenium [15:17:52] it should run the tests with the respective mw1.22-wmf4 branch of datavalues [15:17:57] yes [15:18:08] but it seems not to do this [15:23:17] Hello! [15:25:59] hi pragunbhutani [15:26:18] aude: today was officially the last day of lectures at my uni :) [15:26:24] yay! [15:27:09] Hi. I want propose a little change on http://bits.wikimedia.org/static-1.22wmf4/extensions/Wikibase/lib/resources/wikibase.ui.Tooltip.js is anybody can help me? :) [15:27:41] aude: if you want to merge it, just remove jenkins as reviewer [15:28:01] Tobi_WMDE: ok, once done running tests myself [15:28:07] so i can check verified :) [15:29:12] aude: I've been thinking, for my project I've proposed to submit my work through a series of small commits rather than working on a branch [15:29:38] so can I maybe start working on a few things right away? [15:29:40] New review: Daniel Kinzler; "probably good to merge, got some questions though" [mediawiki/extensions/Wikibase] (master) - https://gerrit.wikimedia.org/r/64037 [15:29:57] pragunbhutani: i think that's good [15:30:20] aude: http://192.168.2.251:8080/job/wikibase-phpunit/272/console [15:30:28] ebraminio: what do you want to change in the tooltip? [15:30:42] Tobi_WMDE: thanks [15:30:43] aude: I haven't had time to look over that last bug yet, https://bugzilla.wikimedia.org/show_bug.cgi?id=47726 [15:30:48] I'd like to get that done first [15:30:52] ok [15:31:34] aude: when we last talked you gave me some pointers on how to get started [15:31:49] can you tell them to me again please? [15:32:32] aude: i want to implement this https://www.wikidata.org/wiki/Wikidata:Contact_the_development_team#add_link_to_errors_and_alarms_messages [15:33:16] pragunbhutani: what we need it the entity id of the connected item for a wikipedia page [15:33:45] aude: I don't want do it of-course, I want find someone to implement it :P [15:33:46] yep, [15:33:49] so, look into how that can be done [15:33:57] okay [15:34:14] relevant classes could be SiteLink and EntityId [15:34:18] and the SiteLinkTable [15:34:34] then we'll need to use a hook to add it to the page info [15:35:09] ebraminio: i think we have a bug ticket for that and it's a long known issue [15:35:21] not sure the status, though [15:36:25] aude: oh, ok. Can you find the bug? [15:36:39] one sec [15:36:43] aude: can we replace wb-tooltip-error-details content like [15:37:05] https://www.wikidata.org/w/index.php?title=MediaWiki%3AGadget-slurpInterwiki.js&diff=11304427&oldid=11241754 [15:37:37] https://bugzilla.wikimedia.org/show_bug.cgi?id=41602 [15:38:02] https://bugzilla.wikimedia.org/show_bug.cgi?id=46717 [15:38:11] might be duplicate bugs [15:39:00] thanks [15:39:31] reza1615: not sure if it could work [15:39:42] maybe [15:39:59] I think in commo.js we can hack it [15:40:10] common.js [15:40:59] should be fixed in wikibase itself [15:41:21]