[Cubicweb] annotating divs with rql and vid

Adrien Di Mascio adimasci at gmail.com
Thu May 24 23:27:09 CEST 2012


On Thu, May 24, 2012 at 9:39 PM, Sylvain Thénault
<sylvain.thenault at logilab.fr> wrote:
>
> On 24 mai 18:55, Nicolas Chauvat wrote:
> > On Thu, May 24, 2012 at 05:24:36PM +0200, Sylvain Thénault wrote:

> The work currently done in the js code could (should?) be done on
> the server side : simply post the facets form, that's enough to build
> the rql, get a rset, apply the view and returning the result (along with
> values to update the facets form display). And this results in a single
> HTTP request instead of 3 as of today (one to build the rql, one to
> execute
> it and apply the view, one for the facets form values).

Other queries are also performed to refresh other boxes / components
(e.g. the action box).

> Corrolary points:
>
> * what can't we have by disabling rql input, beside allowing user to type
>  arbitrary rql?

There are quite a few places in CW or cubes where urls with explicit
rql and vid parameters are generated. If you disable the rql parameter
(which is not that hard), you'll get bitten there.



More information about the Cubicweb mailing list