Re: [udig-users] Unable to apply a spatial Filter using CQL language in

[Jody Garnett <jody.garnett@xxxxxxxxx>]

Thanks - I have figured out what the problem is (not the CQL parser but the way it is used by the table view; the table view only downloaded the attributes for speed - not the geometry) as such CQL is unable to test against the_geom).

I have a workaround committed on trunk right now:

- I set the produced filter as the filter for the layer (I should really have a style page for this)

I think the best fix would be to evaulate the filter against the real data; and then take the set of generated FeatureIds and use that to update the selection in the table view. What do you think (or you do not care as long as it works?)

Please review the bug report with screen snap; it also shows that the units are ignored for this dwithin test.

Jody

On Mon, Feb 9, 2009 at 8:33 PM, Jacques LESCOT <jacques.lescot@xxxxxxxxxxxxxxxx> wrote:

Ok, thanks.
As suggested by Jody, I open a bug : see http://jira.codehaus.org/browse/UDIG-1475

Regards,
Jacques

Mauricio Pazos a écrit :

On Saturday 07 February 2009 11:45:31 Jacques LESCOT wrote:
  

Not from you specifically. I understood that Mauricio will have a look
into the problem, and then I am just hoping some news were posted on
this mailing list when the problem is detected/solved. As usual, sooner
would be the best, but it seems that there are many problems on all the
spatial statements in CQL ... and I guess it would take time ...

Regards,
Jacques
    

Hi Jacques: we were analyzing the problem the last week. We were thinking 
about the better way to solve it. CQL Is OK in version higher than 2.2.x.  Now 
we were waiting the Jesse analysis. He was talking about backport CQL 2.6. I 
don't know what is the exact status, when it is clear we offer our help to 
solve it.


Cheers.