From: [EMAIL PROTECTED] on behalf of Mick Semb Wever
> On Wed, 2008-06-25 at 13:37 +0200, [EMAIL PROTECTED] wrote:
> > Am I missing something here? Is there any other way of defining this?
> > I haven't seen this implemented anywhere else, but I think it's been
> > mentioned earlier as something that should be possible to do.
> 
> Either you subclass the ResultPagingNavigator__ classes providing
> slightly different "pagers" working off separate but still hardcoded ids
> & parameters.

Since AbstractSearchCommand still only uses the hardcoded "offset", this would 
prevent the actual search result from being paged (the offset will always be 
zero).

> And/or you build/improve a ResultPagingNavigator that works off flexible
> id and parameters. The ideal case i can think of is that each command
> can have a pager, so you could scroll through 'pages' of navigation, or
> have multiple results lists on the page each with their own pager. For
> example something like id="offset<navigator-name>" &
> offset<-navigator-name>=X 

I'm not quite sure I understand what you mean. What I've been playing with is 
adding a paging-navigation-id attribute to CommandConfig, which will be used if 
set. Are we on the same page? ;)

So views.xml looks like:

            <navigation id="offset">
                <result-paging id="addressOffset" command-name="address" 
page-size="15" number-of-pages="10"/>
                <result-paging id="whiteOffset" command-name="white" 
page-size="15" number-of-pages="10"/>
                <result-paging id="yellowOffset" command-name="yellow" 
page-size="15" number-of-pages="10"/>
            </navigation>

and modes.xml looks like:

            <catalogue-command id="yellow" [...] 
paging-navigation-id="yellowOffset"/>

Anyways, since this forces changes in AbstractSearchCommand (and other classes) 
in SESAT I'm assuming this won't be in until 2.18. Am I right assuming this? 
I've created SKER4785 and set fix version to 2.18, please move to 2.17 if you 
think that's ok. ;)

-- 
Magne
_______________________________________________
Kernel-development mailing list
[email protected]
http://sesat.no/mailman/listinfo/kernel-development

Reply via email to