> Hi,
> 
> On 09/01/2008, Ard Schrijvers <[EMAIL PROTECTED]> wrote:
> >
> > You cannot query for locks
> > through a DASL, because (AFAIU but do not know exactly) a lock is a 
> > derived property, which might not be something store on the 
> document 
> > itself. Just like you cannot DASL on <d:supported-privilege-set/>. 
> > They are derived properties, which might be a result of 
> some of its parents.
> 
> 
> Ah, that makes sense. I'll have to go with retrieving all the 
> documents, and just cache it until it screams ...

Well....the best way to do this, is through recursion (with depth=1),
and cache each result seperately. You then have 2 things that work
better:

1) you have a lot of small propfinds instead of one gigantic, where the
latter will in the end kill the repository when it gets large enough
2) changing a document won't invalidate the huge single propfind, but
will only invalidate a trail of cached propfinds. So, if at depth 3 a
document is changed, only 3 entries will be deleted, and need
refetching...

then again, we still have one problem: a change on the lock of a
document won't trigger a jms, so you won't be able to cache it
actually....

You can't do without it?

> 
> Thanks for the comprehensive answer!

your welcome, although I probably do not have the answer you were hoping
for :-) 

Ard

> 
> 
> Andrew.
> --
> Sourcesense: Making sense of Open Source
> Tel: +44 (0)870 741 6658  Fax: +44 (0)700 598 1135
> Web: http://www.sourcesense.com/
> ********************************************
> Hippocms-dev: Hippo CMS development public mailinglist
> 
********************************************
Hippocms-dev: Hippo CMS development public mailinglist

Reply via email to