I think a maintenance script is a good idea.  Perhaps even a button on a
SMW "special page" to trigger the script for easy use when you need it
right away.

I'm wondering if the effort to generally improve the way templates work
with SMW might go hand in hand with the effort to formally support SMW in
templates with Mediawiki (Without having to patch parser.php)

While I'm rambling on to this other topic it would be wonderful to not have
to apply a patch to Mediawiki down the road.  I'm trying to formally move
our wiki  into more of a "production" state, and telling our Unix/system
admins that they need to apply a patch to parser.php every time they do a
quarterly software update on Mediawiki probably won't sit to well with
them.  As it is now I play that role and do the patching, but it's not
really my job. I just happen to be the techy wiki evangelist at the
moment...

One more thing.  You guys probably don't hear this enough: Thanks! to all
the SMW developers,  you guys are doing an awesome job.

Mike

                                                                           
      Mike Axelrod                                                         
      Solution Architect                                                   
      Enterprise Architecture & Integration Team                           
      Excellus, Inc.                                                       
      [EMAIL PROTECTED]                                         
      office: 585-301-8345                                                 
                                                                           





                                                                           
             "Yaron Koren"                                                 
             <[EMAIL PROTECTED]                                             
             m>                                                         To 
             Sent by:                  "[EMAIL PROTECTED]"      
             semediawiki-user-         <[EMAIL PROTECTED]>      
             [EMAIL PROTECTED]                                          cc 
             rceforge.net              [EMAIL PROTECTED] 
                                       net                                 
                                                                   Subject 
             04/19/2007 10:50          Re: [Semediawiki-user] Regenerating 
             AM                        child content for "ask" queries     
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Hi,

Inline queries work by doing database queries on the tables that store
the attributes and relations that have been defined. Whenever you save
(or re-save) a page that defines attributes and relations, the old
attributes and relations (if there were any) get removed from the
database, and the new ones are added. After that, the new data should
show up in inline queries immediately; though you might need to do a
refresh on the page *that holds the query*, to prevent caching.

I'm guessing that the problems you're all experiencing are brought on
by the use of templates. If pages define attributes and relations
indirectly through templates, and you change some of the semantic
structure of a template, that change won't be reflected in the
database: you'll need to re-save each of the pages that use that
template. If you change around the categories in a template, that
change will show up in the database, without any work on your part;
though it takes a few minutes. But that need to re-save is certainly
annoying, especially since there can often be a lot of trial-and-error
in deciding how templates should be structured.

I think what's needed (and it's been discussed before a few times) is
a maintenance script that wipes out all the semantic data in the
database and recreates it. It seems like a lot of people would benefit
from such a tool.

-Yaron


On 4/18/07, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
wrote:
> Yes, I'd like to understand the real mechanism here also.  I too have
> witnessed this similar behavior. The problem was worse for me when I had
> dozens of pages that I had to "re-edit".   How could I know that if i
just
> waited they would eventually show up...  What is the "wait time"?
> Minutes?, hours?
>
>       Mike Axelrod
>       Solution Architect
>       Enterprise Architecture & Integration Team
>       Excellus, Inc.
>       [EMAIL PROTECTED]
>       office: 585-301-8345
>
>
>
>
>
>
>
>              "Lionel Brits"
>              <[EMAIL PROTECTED]
>              l.com>
To
>              Sent by:                  "Rich Mintz" <[EMAIL PROTECTED]>
>              semediawiki-user-
cc
>              [EMAIL PROTECTED]
[EMAIL PROTECTED]
>              rceforge.net              net
>
Subject
>                                        Re: [Semediawiki-user]
Regenerating
>              04/18/2007 01:12          child content for "ask" queries
>              PM
>
>
>
>
>
>
>
>
>
> I don't know if this is superstition, but it seems like the pages enter
> some sort of queue after they're refreshed, because I've noticed that the
> ask queries *eventually* give the correct results without intervention.
But
> this is only anecdotal and I could be wrong. If you figure this out,
please
> share your results.
>
>
> On 4/18/07, Rich Mintz <[EMAIL PROTECTED]> wrote:
>   Hello,
>
>   I'm trying to get more information about how to regenerate the
>   relations that are needed for pages to come up in an "ask" query.
>   Specifically:
>
>   * I have pages called Manhattan, Brooklyn, Queens, etc., that have
>   [[category:borough]] and [[is in city::New York City]] set for them.
>   * On the page "New York City", I create:
>
>   <ask>
>   [[Category:Borough]]
>   [[Is in city::New York City]]
>   </ask>
>
>   * But my borough pages don't show up in the "ask" results until I
>   *edit* those borough pages individually.  Touching them (i.e.,
>   decaching them) apparently is not sufficient.
>
>   Can someone point me to documentation that explains a little more
>   about the conditions under which a page's categories and relations
>   become available for querying?  We would like to find or write a
>   utility or bot that touches (or edits, etc.) pages periodically for
>   this purpose, but we need to understand exactly what it needs to do.
>   I've seen documentation of this type, but now I can't find it.
>
>   Rich Mintz
>
>
-------------------------------------------------------------------------
>   This SF.net email is sponsored by DB2 Express
>   Download DB2 Express C - the FREE version of DB2 express and take
>   control of your XML. No limits. Just data. Click to get it now.
>   http://sourceforge.net/powerbar/db2/
>   _______________________________________________
>   Semediawiki-user mailing list
>   Semediawiki-user@lists.sourceforge.net
>   https://lists.sourceforge.net/lists/listinfo/semediawiki-user
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Semediawiki-user mailing list
> Semediawiki-user@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/semediawiki-user
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Semediawiki-user mailing list
> Semediawiki-user@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/semediawiki-user
>

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Semediawiki-user mailing list
Semediawiki-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/semediawiki-user



-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Semediawiki-user mailing list
Semediawiki-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/semediawiki-user

Reply via email to