[Cubicweb] annotating divs with rql and vid

Florent Cayré fcayre at logilab.fr
Thu May 24 19:21:23 CEST 2012


Le 24/05/2012 19:00, Nicolas Chauvat a écrit :
> On Thu, May 24, 2012 at 06:30:25PM +0200, Sylvain Thénault wrote:
>    
>> On 24 mai 18:04, Aurélien Campéas wrote:
>>      
>>> Le 24/05/2012 18:00, Adrien Di Mascio a écrit :
>>>        
>>>> On 24/05/2012 17:33, Sylvain Thénault wrote:
>>>>          
>>>>> On 24 mai 17:02, Adrien Di Mascio wrote:
>>>>> The latter ==<div id=foo rql=...>  ?
>>>>>            
>>>> Yes, that's what I meant.
>>>>          
>>> But do we all agree that what's really needed is along the lines of
>>>
>>> <div class="viewtop", callparams="..." ...>
>>>
>>> where callparams contains the complete set of properly serialized
>>> arguments to the orignal view ?
>>>        
>> yes, where call params include vid, rql (to get back the rset) and extra
>> arguments as discussed with pyves.
>>      
> I am not sure I understand Aurélien's proposal. Why class= ? Why
> hiding rql inside callparams ?
>
> My intention is to expose the rql/vid and make it clear that it is the
> foundation of the framework and how the HTML got generated.
>
> The rest of the page needs to be able to manipulate the rql/vid. If
> you encode them in callparams, you will need a couple of js function to
> encode/decode them and it will be more difficult to read/debug.
>
>    

My 2 cents :

I am not sure how comfortable it will be to manipulate the rql as a 
string anyway, particularly in javascript. If Sylvain has an idea of how 
to get facets working with client-side rql requests disabled (more for 
DoS protection reasons than security ones IMHO), I am curious to see how 
it could be used to manipulate the RQL query more easily.




More information about the Cubicweb mailing list