[postgis-devel] Call for 1.4.2 and 1.5.1 (Handling of Invalid Geometries)

[Kevin Neufeld]

That makes the most sense to me ... but that breaks the 1st bullet in your proposed functional contract for 
ST_CleanGeometry.
  * Returned geometry will have same dimensionality of input

-- Kevin

On 2/17/2010 1:19 PM, strk wrote:
> On Wed, Feb 17, 2010 at 10:18:08PM +0100, strk wrote:
>> On Wed, Feb 17, 2010 at 01:15:54PM -0800, Kevin Neufeld wrote:
>>> Interesting.  So does this mean that a 2-ordinate polygon would get
>>> converted to a LINESTRING and a 3-ordinate polygon would automatically have
>>> a fourth point added, thus closing the ring?
>>>
>>> What would these clean to?
>>> 'POLYGON((0 0),(1 1 2 2))'
>>> 'POLYGON((0 0),(1 1 2 2),(3 3, 4 4, 5 5))'
>>
>> Can you provide them in HEXWKB form for me to check ?
>> PostGIS won't let me write that in the db....
>
> Beside the sarcasm, the plan is for the above to become
> 'GEOMETRYCOLLECTION(POINT(0 0),LINESTRING(1 1, 2 2))'
> 'GEOMETRYCOLLECTION(POINT(0 0),LINESTRING(1 1, 2 2),LINESTRING(3 3, 4 4, 5 5))'
>
> --strk;
>
>    ()   Free GIS&  Flash consultant/developer
>    /\   http://foo.keybit.net/~strk/services.html
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel