[postgis-users] NumInteriorRing(s) ?

[Paul Ramsey]

Yeah...

"NumInteriorRing" is superficially an error in the specification, and  
I brought it up in the SFSQL committee a couple months ago.  It turns  
out, it was deliberate.  SFSQL is an older specification (late 90s)  
and they were limiting the length of their function names.  The only  
way they could fit in NumInteriorRings was to drop the 's'. (Me, I  
would have turned all the "Num*" functions into "N*" functions, but  
hindsight is 20/20.)

And then the author of the conformance guidd I guess did what I did  
and assumed NumInteriorRing was a typo, and fixed it for the  
conformance document.

In any event, we should support the "officially mis-spelled" version  
as well in order to be *completely* true to the spec.

Paul

On 9-Jan-06, at 5:24 AM, Arnaud Lesauvage wrote:

> Hi all !
>
> I used two documents as references to write my spatial queries :
> - OpenGIS® Implementation Specification for Geographic
> information. Simple feature access. Part 1: Common architecture
> - OpenGIS® Implementation Specification for Geographic
> information. Simple feature access. Part 2: SQL option
> Both are the latest specifications that I got from the OpenGis web
> site.
>
> These two documents reference the function NumInteriorRing as the
> function to determine the number of interior rings for a given
> polygon.
> The latest document also talks about NumInteriorRings (notice the
> 's' at the end) in the "Conformance Test" paragraph, and only in
> this one (and only once, in the recapitulative table). At the end
> of the document, the test query for conformance is :
> SELECT AsText(ExteriorRing(shore))
> FROM lakes
> WHERE name = 'Blue Lake';
>
> It appears that this function is implemented as NumInteriorRings
> in PostGIS, but I am quite puzzled that the official specification
> describes this function with two different spellingsand that
> PostGIS apparently implements the "less-official" one (with an 's').
>
> Thanks for your opinion on this !
>
> Arnaud
>
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users