[GRASS-dev] Re: v.buffer creates artefact - bug?

[Maciej Sieczka]

Maris Nartiss wrote:

> I downloaded provided location and can confirm bug. I get same buffer
> as You. Running v.clean or v.build on vector does not change anything.
> v.digit fixes problem, as You wrote.

"OK". So there is something fishy here - 3 people confirmed till now.

> One strange thing - v.info says, that there is one island. I can not
> see any such feature in v.digit. Am I missing something?

This is as expected. A separate area is an island at the same time,
AFAIK. When there are two such areas, like on the attached screendump,
there are two islands - and so forth:

$ v.info -t checkit
nodes=4
points=0
lines=0
boundaries=2
centroids=2
areas=2
islands=2
faces=0
kernels=0
primitives=4
map3d=0

> GRASS 6.2.1 and 6.3-cvs (~ week old)
> Gentoo GNU/Linux on i686 (Intel P4 Celeron M)
> GCC 4.1.1

I can reproduce this on a 32bit Pentium M Dothan (2.6.15-27-686) and on
a 32bit AMD K7 Barton (2.6.15-27-k7), both Ubuntu 6.06, GCC 4.0.3.

Hamish said it was not reproducible on his "Debian/Sarge, gcc 3.3.5,
linux 2.4.27-3-686 on a P4".

Maybe GCC version makes the difference? I'll try GCC 3 when possible.

Hamish,

Are you still not able to reproduce this problem on your setup?

Maciek
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 2islands.png
Type: image/png
Size: 2244 bytes
Desc: not available
Url : http://lists.osgeo.org/pipermail/grass-dev/attachments/20070131/6fdd7244/2islands.png