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

[strk]

On Wed, Feb 17, 2010 at 11:36:05AM -0800, Martin Davis wrote:

> I agree that users should be allowed to load topologically invalid data, 
> but structurally invalid data seems like it's more trouble to handle 
> than it's worth.  Is it not an option to simply always close rings on 
> input to the database? 

Close the ring below and you'll see how much of an option that is:

 +-----------+
 |           |
 | +-------+ |
 | | +---+ | |
 | | |   | | |
 |   |   |   |
 +---+   +---+

You'd go from structural invalidity to topological invalidity.

> If you allow structurally invalid rings into the database, I think you 
> wind up with a situation where users never know if their data is OGC SFS 
> compliant or not - which I think would do more harm than good to the 
> repuation of PostGIS.

Isn't this the whole reason for ST_isValid to exist ?

--strk;

  ()   Free GIS & Flash consultant/developer
  /\   http://foo.keybit.net/~strk/services.html