[mapserver-dev] RFC 62: Support Additional WFS GetFeature Output Formats

Lime, Steve D (DNR) Steve.Lime at state.mn.us
Fri Oct 8 23:52:38 EDT 2010

Hi Frank: For multi polygons there is already code availible to detect inner and outer rings. It's been awhile since I wrote it but it was necessary for both the GML writer and the shapeObj to GEOS translation. It's mapprimitive.c and you can see example usage in mapgml.c or mapgeos.c. Seems to work just fine. Folks like Bart really put it through the wringer with WFS testing...


From: mapserver-dev-bounces at lists.osgeo.org [mapserver-dev-bounces at lists.osgeo.org] On Behalf Of Frank Warmerdam [warmerdam at pobox.com]
Sent: Friday, October 08, 2010 4:25 PM
To: Stephen Woodbridge
Cc: mapserver-dev at lists.osgeo.org
Subject: Re: [mapserver-dev] RFC 62: Support Additional WFS GetFeature Output   Formats

Stephen Woodbridge wrote:
> Regarding MULTI* support, I assume that you would have to support export
> features from all data sources, like postgis. It is common for the
> shp2pgsql loader to create MULTI objects by default. And while
> shapefiles are seem to be mostly NOT MULTI* by default I almost always
> find a random MULTI* in various files. I'm also wondering if the output
> of geometry processing function using GEOS that are built into mapserver
> will be able to be exported via this WFS writer? If so, then it would
> need to support all geometry types possible from that, which would
> include collections in addition to MULTI*.


I agree - I'll need to support multipoint, multiline and multipolygon
geometries.  Multipolygon will be a bit involved since in the
shapeObj there is no real way to distinguish between inner rings
of a polygon and outer rings of additional polygons implying I
need to produce a multipolygon.

> I guess this raises these questions in my mind:
> Do users need to be able to filter the output based on geometry type?

I don't see a compelling need for this, though there are already
mechanisms to do this for OGR inputs with a FILTER statement.

 > Do  we need to provide MULTI* dumpers to base type?

Do you mean a mechanism to split one multi feature into
several simple features?  I don't see this as critical.

I'll incorporate support for multigeometries in the RFC.

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent

mapserver-dev mailing list
mapserver-dev at lists.osgeo.org

More information about the mapserver-dev mailing list