[postgis-devel] Re: Rounding off coordinates (from Postgis-users)

[strk]

Gino, about removing points:
	- You'll never introduce self-intersections
	- You might make LineString collaps to Point
	- You might make Polygon collaps to LineString or Point

Collapses can be handled by either add up to the required amount
of points for the geometry (I'd suggest this - although I'm not
sure about OGC compliance) or modify geometry type (could drive
clients crazy) or set them to NULL (could be needed to localize a
feature).

--strk;

On Tue, Jun 29, 2004 at 07:02:01AM -0700, Paul Ramsey wrote:
> Gino,
> Good help to be had on postgis-devel.
> The only "internal" debugging scheme in pgsql is classic "print your 
> state until you die" debugging, which you can do using the pgsql 
> logging facility (there should be lots of commented out examples in the 
> postgis code). For more advanced debugging, you should use the standard 
> GNU debugger, gdb.  The trick there is to not attach gdb to the main 
> postmaster, but to the postgres backend you are actually running the 
> command within.
> P
> 
> On Tuesday, June 29, 2004, at 06:14 AM, Gino Lucrezi wrote:
> 
> >>Write up the function... base it on the parameters of the standard 
> >>pgsql
> >>round function that Charlton has demonstrated below, it would be a
> >>useful addition.
> >
> >I started writing both a ROUND and a TRUNC function.
> >
> >I got yesterday's CVS build, and started browsing the code to see 
> >where I could add it.
> >
> >My model is the translate function, which works in a similar way; it 
> >has a geometry argument and some numeric arguments, it scans all 
> >points in the geometry and performs some manipulation on them based on 
> >the numeric arguments.
> >
> >It compiles perfectly, but it crashes PostgreSQL with no explaination.
> >
> >I need of some debugging tips... is there any debug facility in 
> >PostgreSQL/Postgis?
> >
> >By the way, I was also thinking about Markus and strk's comments. It's 
> >not just a matter of rounding off point coordinates, you'd have to 
> >remove duplicates and I fear it might still cause some incorrect 
> >geometries. It would still be OK for human reading.
> >Any suggestions would be welcome, too.
> >
> >Gino Lucrezi
> >Penta Consulting Services Srl
> >
> >
> >
>      Paul Ramsey
>      Refractions Research
>      Email: pramsey at refractions.net
>      Phone: (250) 885-0632
> 
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel