[mapserver-dev] MS RFC 68: Support for combining features from multiple layers (Call for discussion)

[Jeff McKenna]

Hi Tamas,

Oh ok I see.  Can you add this example to the RFC so it documented?  Thanks.

-jeff



On 11-02-14 11:03 AM, Tamas Szekeres wrote:
> Hi Jeff,
>
> 'Cluster:FeatureCount' would be added to the CLUSTER layer as an
> additional attribute. In this regard it can be used as the text data
> when labelling the features, I imagine something like:
>
> LAYER
>    TYPE POINT
>    NAME cluster
>    CONNECTIONTYPE CLUSTER
>    PROCESSING "SOURCELAYER=layername"
>    CLASS
>      EXPRESSION ([Cluster:FeatureCount] > 1)
>      TEXT ([Cluster:FeatureCount])
>      LABEL
>      ...
>      END
>    END
>    ...
> END
>
> I would prefer this option to work even with the STYLEITEM "AUTO"
> setting, where the styles for the source layer would be retrieved only
> for the individual shapes not for the clustered shapes. I hope I can
> provide a working example soon.
>
> Best regards,
>
> Tamas
>
>
>
> 2011/2/14 Jeff McKenna <jmckenna at gatewaygeomatics.com
> <mailto:jmckenna at gatewaygeomatics.com>>
>
>     Hello Tamas,
>
>     This is a very interesting proposal.  I often have a hard time to
>     picture these changes beforehand, but I am trying to now and I have
>     a question: when it comes to clustering, I think a common need is
>     also to label the number of clustered points...and since you'll know
>     how many features are combined according to your proposal (I see
>     'Cluster:FeatureCount' mentioned in the RFC) I believe we must also
>     expose this clustercount in the label object.  How I am not sure,
>     but to me this will be a common need.  Or is this mentioned in the
>     RFC already?
>
>     -jeff
>
>
>
>
>     On 11-02-10 6:02 PM, Tamas Szekeres wrote:
>
>         Hi Devs,
>
>         In reference to the feature request from WhereGroup
>         <http://www.wheregroup.com/en/welcome> we did some refinement in the
>
>         specification and I've created MS-RFC-68
>         (http://trac.osgeo.org/mapserver/browser/trunk/docs/en/development/rfc/ms-rfc-68.txt)
>         which contains a reasonable implementation to this option without
>         requiring significant modifications in the MapServer codebase
>         and the
>         current processing chain. With regards to the timing, I already have
>         some working code in place which could be finished by the time
>         of the
>         feature freeze of MapServer 6.0 we are targeting now (feb. 28).
>
>