[01:32:49] hey - in parserhooks/SMW_SMWDoc.php, is: class SMWSMWDoc extends ParserHook ... I can't find ParserHook in MW and don't know why - thanks [01:35:16] I mean to say I don't why I can't find ParserHook in svn.wikimedia.org [06:39:06] New review: Nikerabbit; "(1 comment)" [mediawiki/extensions/SemanticForms] (master) C: -1; - https://gerrit.wikimedia.org/r/53270 [09:01:08] pfff. I change two templates and runJobs.php has already been running 6h [09:01:23] I feel the desire to complain about bad performance being bad! [09:02:15] 2013-03-12 08:06:55 refreshLinks2 Template:Project start=33681 end=35244 t=837064 good [12:13:53] yaron: thanks for the ED patch [12:15:08] Oh, yeah - well, thanks for finding the issue. [12:23:23] yaron: could I trouble you to push it onto referata? [12:23:39] I want to demo this wiki to a colleague, so it can't be on my local box... [12:25:11] Oh, yeah, okay. [12:25:49] Done! [12:25:56] Many thanks Yaron!! [12:28:52] Looks beautifull! [12:29:33] Wow - cool! [12:30:38] http://das.referata.com/wiki/DAS_registry_import [12:46:17] do subobjects work like SIO? i.e. should this work: {{#ask: [[my sub obj::+]]}} [12:47:18] do subobjects work like SIO? i.e. should this work: {{#ask: [[my sub obj::+]]}} [12:47:57] http://semantic-mediawiki.org/wiki/Help:Adding_subobjects # Doesn't mention querying [12:48:14] Yes - you can query them like regular pages. [12:48:29] What was that Referata URL, again? [12:48:43] http://das.referata.com/wiki/DAS_registry_import [12:48:47] My IRC client crashed, for some reason. [12:49:15] Calls a template for each 'row' returned, which calls #subobject in turn [12:49:21] but I don't see any results in my #ask [12:50:12] Oh, yeah, that doesn't look too good right now. [12:50:40] yaron: added some 'noise' in the template for debugging, just to show it gets called [12:50:49] Ah. [12:51:22] Just made the template very explicit [12:51:40] but still nothing here: http://das.referata.com/w/index.php?title=Talk:DAS_registry_import&action=edit [12:52:47] * dbolser populates the page for 'Has subobject' just in case ... [12:53:54] I don't understand how the 'identifier' works I guess [12:54:09] it doesn't work like SIO [12:54:19] Yes, the syntax is different. [12:54:33] I'll update teh man page later, g2g just now [12:54:39] The first parameter of #subobject should be blank - it's not a property name. [12:54:44] right [12:54:52] * dbolser sets a dummy property [12:56:43] The identifier idea is quiet nice (just didn't understand at first) [12:57:01] http://das.referata.com/wiki/DAS_registry_import :-) [13:02:08] Looks better. [13:05:19] I don't think the identifier thing is nice, by the way... I actually find it pointless. [14:31:33] Change merged: jenkins-bot; [mediawiki/extensions/SemanticMaps] (master) - https://gerrit.wikimedia.org/r/53149 [14:56:36] Change merged: jenkins-bot; [mediawiki/extensions/SemanticWatchlist] (master) - https://gerrit.wikimedia.org/r/53021 [16:56:48] New patchset: Mwjames; "ParserHook re-factoring, remove static binding, and add unit tests" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/52598 [17:05:18] New patchset: Mwjames; "ParserHook re-factoring, remove static binding, and add unit tests" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/52598 [17:35:47] New review: Mwjames; "Since git seems to have a problem with renaming files and doing a rather an extended change on a ren..." [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/52598 [18:21:33] New patchset: Foxtrott; "new parameters 'root' and 'start level'; bugfix; some clean-up" [mediawiki/extensions/SemanticResultFormats] (master) - https://gerrit.wikimedia.org/r/52266 [18:35:49] hi everyone; anyone know if it's possible to conditionally pass arguments to #subobject, e.g., something like {{#subobject:id |prop1 = {{{val1}}} {{#if: {{{val2|}}} | {{!}} prop2 = {{{val2}}} }} }} ? [18:42:01] tayloj: I would think that's possible, yes. [18:42:40] Though in that specific case, it shouldn't be necessary - passing a null value to #subobject is the same as not setting it at all. [18:43:13] I say "something like" because that doesn't actually work. e.g., the code at http://pastebin.com/rhqYPyxZ gives me a caution sign with "v1 | p2 = v2" cannot be used as a pagename, [18:44:30] Oh, I didn't realize that passing a null would have the desired effect. [18:52:44] Any idea about that conditional text getting handled strangely? [18:55:36] That's not so big a deal now, since the passing null works (didn't even think to try that; d'oh). [19:00:30] Okay, cool. I assume the issue was with trying to pass the pipe over. [19:04:00] I don't have a great grasp yet on which order things get evaluated, but I'd hoped that the same technique that works for, e.g., conditional rows in tables would work in the subobjects, but I guess not. [19:17:54] tayloj: no, pipes in parser functions are different from pipes in tables. [19:47:24] I guess what it comes down to is that the outermost parser function splits things by | first, so calls to parser functions inside don't have the chance to add more arguments? [20:31:11] tayloj: no, the issue is just that, for a parser function, {{!}} isn't the same as | . [20:31:33] (Whereas, for a table, they're the same.) [20:54:05] I still feel like I'm missing something, then. Since {{!}} expands to |, if {{#if: text | {{!}} foo = bar }} were completely expanded in {{#fun {{#if: text | {{!}} foo = bar }} }} *before* {{#foo: ... }} were processed, then the call to fun would be {{#fun: | foo = bar}} . [20:54:39] That'd be pretty backwards, though since then the *parser function* fun wouldn't actually get to do its own parsing of the text inside {{#fun: ... }} [20:57:15] #fun needs to split up ... by literal | first, it would seem, and that's why {{#if: ...}} inside {{#fun: ... }} can't go and "splice in" more pipes. [20:57:39] Anyhow, none of this is particular to SMW, and I'm reading up a bit more on parser functions in general [21:01:01] Just like CL macroexpands the outer before the inner (https://groups.google.com/d/msg/comp.lang.lisp/8_kwPkKjGwI/tGtcLcCP3mAJ) since the outer can manipulate the inner. [22:48:30] New patchset: Mwjames; "ParserHook re-factoring, remove static binding, and add unit tests" [mediawiki/extensions/SemanticMediaWiki] (master) - https://gerrit.wikimedia.org/r/52598 [22:52:08] New review: Mwjames; "As requested, all #set, #subobject, #ask, #show, #concept are now using constructor dependency injec..." [mediawiki/extensions/SemanticMediaWiki] (master) C: 1; - https://gerrit.wikimedia.org/r/52598