Last modified: 2006-06-04 01:37:47 UTC
Halló! For http://jadesukka.homelinux.org:8180/betawiki/Betawiki:Templates/Template:Task some documentation is available at http://jadesukka.homelinux.org:8180/betawiki/Project:Templates/Template:Task/description#documentation The template includes several enhancements for [[en:Template:To do]] but due to the lack of requested extensions is using around 170 objects. It is vanity to create a template which reverts all language localisations. This attempt shows that at the moment it is possible to make some programming with the existing features in MediaWiki. It does not make sense to be excessively restrictive regarding programming because of vandalism and should focus on providing simpler solutions to what is available anyhow at this moment. Developers should not tray to solve all requests alone. An intermediate platform (based on settings, templates and roles) should be available too. I did run into a problem regarding priority managment for "template:task" which supports priority handling. It is stored in a extra page ".../to do/priority" and supports 5 values. Unfortunately this "atribute" MUST be specified before being able to continue. It would make sense to use a construct "IFEXISTS page" in templates. Only if a page exists actions could be performed display / hide the "output". A simple example would be to add a link to an archive of tasks performed already. Bug 233 relates to this because it is a kind of "IFEXISTS pages LIKE '{{NAMESPACE}}:{{PAGENAME}}/%'" (% in MySQL). IFEXISTS <object> should relate also to parameters of a template. This way parameters could be optional to that respect that the number of trailing | would NOT be mandatory. In order not to interfeare with existing templates IFEXISTS <object> would return TRUE for EMPTY parameters. It is not my intention to provide a specification but to encourage taughts about this. Another question would be how long it takes to implement after procedures about status asigned etc. passed. Best regards Reinhardt [[user:gangleri]]
This seams to relate to bug 364 and [[meta:Extended template syntax]]. Can anybody provide status information?
Would anyone care to provide a concise summary of what this is about?
changes summary from "extensions to template syntax IFEXISTS <object>; IFNOTEXISTS <object>; IFEQUALS <object>; ELSE" to "new constructs in template syntax"
Then it's defenetly a duplicate of bug 364, marking it as such. *** This bug has been marked as a duplicate of 364 ***
Dear friends! maybe the summary does not explain comment 0 it is definitevely *not* a duplicate of bug 364 IFEXISTS <object> could be used in templates as IFEXISTS [[{{NAMESPACE}}:{{PAGENAME}}/archive]] THEN <code> normaly [[{{NAMESPACE}}:{{PAGENAME}}/archive|archive]] IFEXISTS [[{{NAMESPACE}}:{{PAGENAME}}/to do]] THEN <code> normaly [[{{NAMESPACE}}:{{PAGENAME}}/to do|to do]] IFEXISTS [[Help:{{PAGENAME}}/to do]] THEN <code> normaly [[Help:{{PAGENAME}}|Help]] IFEXISTS some sort of query as LIKE '{{NAMESPACE}}:{{PAGENAME}}/%'" (% in MySQL) could be used with the apropriate variant of [[Spacial:Allpages]] to make a link to subpages as required in bug 233 The difference to bug 364 is that a *parameter* might be in the template but involved in some code *but* the generated / referenced *object* does not exist. IFNOTEXISTS [[{{NAMESPACE}}:{{PAGENAME}}/to do/priority]] THEN <code> would generate default code for this situation. IFEQUALS [[{{NAMESPACE}}:{{PAGENAME}}/to do/priority]] = done THEN <code> would be helpfull from the begining. Other parser keywords as ELSE, ELSEIF and END or whatever would be there names are required to make somthing meaningfull. This is not the place to make a full specification. If there are doubts about vandalism a way should be found. Possibly restricting the usage (runtime effects) to protected template pages. In order to design such templates everybody could do this inside <pre> ... </pre>. But as already told restrictions sould not be the first issue to thing about here. Regards Reinhardt [[user:gangleri]]
I'm hazarding a guess that the core issue here was fixed with the introduction of Tim's ParserFunctions.
