W3C home > Mailing lists > Public > www-webdav-dasl@w3.org > October to December 2006

Re: Using property expansion and sets in criteria

From: Julian Reschke <julian.reschke@gmx.de>
Date: Mon, 20 Nov 2006 10:43:59 +0100
Message-ID: <456178DF.6070700@gmx.de>
To: werner.donne@re.be
CC: www-webdav-dasl@w3.org

Werner Donné schrieb:
> 
> Hi,
> 
> Consider a question to a server like: "Which VCRs have the
> activity x in the activity-set of their checked-in version?"
> Two things would be necessary for this. First, where the
> "prop" element is expected, a nested prop element should be
> allowed as well; something reminiscent of the expand property
> report. Second, there should an "in" operator for all those
> properties which are in fact sets.
> 
> Here is a possible example of the above question:
> 
> <d:searchrequest xmlns:d="DAV:">
>  <d:basicsearch>
>    <d:select>
>      <d:prop><d:displayname/></d:prop>
>    </d:select>
>    <d:from>
>      <d:scope>
>        <d:href>http://hostname/container1/</d:href>
>        <d:depth>infinity</d:depth>
>      </d:scope>
>    </d:from>
>    <d:where>
>      <d:in>        <d:property name="checked-in">
>          <d:property name="activity-set"/>
>        </d:property>
>        <d:literal>http://hostname/act/my_activity</d:literal>
>      </d:in>
>    </d:where>
>  </d:basicsearch>
> </d:searchrequest>
> 
> Regards,
> 
> Werner.

It is a well-known restriction of the basicsearch grammar that it is 
simple. This is by design. When it was created, the assumption was that 
at a later point it would be replaced by something that is more XML 
aware, such as something XPath-based or even XQuery.

That being said, the functionality you're looking for can of course be 
implemented as an optional extension to basicsearch (that's why there is 
grammar discovery, see 
<http://greenbytes.de/tech/webdav/draft-reschke-webdav-search-09.html#rfc.section.4>).

The right next step would be to write a self-contained proposal, and try 
to get at least one additional implementor behind it (otherwise it would 
make little sense to standardize it).

Best regards, Julian
Received on Monday, 20 November 2006 09:44:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 22 March 2009 03:38:10 GMT