[postgis-users] Measures (LRS) (renamed)

Christoph Spoerri spoerri at duke.edu
Tue Mar 9 14:55:12 PST 2004


Dave, 

I downloaded the (hourly) cvs tar.gz file from your website and checked the 
lwgeom.h file. I have couple questions:

- should we have a 2d point with a M value (x,y,m) ? Or would this complicate 
things only?

- how would one determine if a point has a M value? Could we use one of the x 
in the LWGEOM types (xxSDtttt)?

- the old geometry type are still valid, correct? If so, are they only a 
wrapper for the new type or does the code need to differentiat between the 
two types?

thanks,
Christoph

btw. does the lwgeom code already compile and work?

On Tuesday 09 March 2004 04:21, Dave Blasby wrote:
> Quoting Christoph Spoerri <spoerri at duke.edu>:
>
>
> Christoph,
>
> This is actually good timing.  I just finished implementing code required
> to implement the lwgeom.h I proposed a little while ago.
>
> Currently, I have support for (x,y) and (x,y,z).  I could add (x,y,z,m)
> support, I just have to figure out exactly how to do it.
>
> This would almost certainly make thing way easier for you, but more
> complicated for me.
>
> Take a quick look at lwgeom.h (in CVS and on the list) - I had to make a
> bunch of minor changes during the implementation, but nothing at all major.
>
> I guess we'll have to change "is3d" to "dimentionality type" (2d, 3d, 4d),
> and abstract the info out in POINTARRAY.
>
> I'll see if i can put some thought in tomorrow.
>
> dave
>
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users



More information about the postgis-users mailing list