[GRASS5] Re: New operation to v.select: noverlap

[Wolf Bergenheim]

Allan, good news! I just figured out a way to do the contains operator.
is contained by can be doe by reversing a and b.

Here is what I plan to do, this is a bit technical, but it will serve as
a project plan (;
 * Check that every point of B is in A by calling Vect_point_in_area()
 * If this is true, then check that all points of A are not included in
	B with the same function. This is to ensure that A does not
	contain surprises like this:
+--+  +----+
|  | *|  A |
+ ++--+--+ |
| || *|  | |
| |+--+  | |
| |     B| |
| +------+ |
+----------+
Here A does not contain B (even if all ponts of B are inside A), because
of the "bay" marked by *.

Markus, Radim how does this sound?

--Wolf

On 13/12/05 16:24, Wolf Bergenheim wrote:
> On 13/12/05 16:02, Allan Metts wrote:
> 
>>At 04:56 PM 12/12/2005, Wolf Bergenheim wrote:
>>
>>
>>
>>>This patch adds another option to v.select, "noverlap". This is the
>>>opposite of overlap. So all fetaures not in b are selected from a.
>>
>>
>>Thanks, Wolf...
>>
>>I recently had the need for "a contains b" and "a is-contained-by b"
>>in v.select. Any chance you want to add these as well, since you're on a
>>roll ? :)
>>
> 
> 
> Ahmm.. I'm not sure I would know how to do that. The noverlap was easy,
> since there was no need to do any checking. All I had to do was reverse
> the overlap case. But it sounds like an interesing challange. I'll
> probably do it, but no promises! (;
> 
> --W
> 

-- 

<:3 )---- Wolf Bergenheim ----( 8:>