CGAL users discussion list

[Andreas Fabri <>]

Hello,

I just have a look at your data.
You have two slanted rectangles that have an almost overlapping edge.
Even if this works with an exact number type, it would make
no sense to run the Delaunay conforming algorithm with these
close segments, as the algorithm would produce many many triangles
between the close segments.

I am looking if the nap rounding package is the solution
for this problem. The idea is to snap segment endpoints
which are almost on another segment to the segment.
This also avoids the case that you have a segment that
intersects another segment, and gets split in two segments
with one of them having length almost zero.

Best,

Andreas

On 27/08/2014 15:09, HuyLe wrote:
> I have isolated further the issue, my guess is that two constraints (#56 and
> 69 in the attachment conforming.cpp
> <http://cgal-discuss.949826.n4.nabble.com/file/n4659749/conforming.cpp>  )
> overlap each other, or are very close. In the example there are many
> constraints that overlap which doesn't cause issue. Is there a way to
> configure the algorithm to take care of such situation?
>
>
>
> --
> View this message in context: 
> http://cgal-discuss.949826.n4.nabble.com/assert-in-refine-edges-h-tp4659748p4659749.html
> Sent from the cgal-discuss mailing list archive at Nabble.com.

--
Andreas Fabri, PhD
Chief Officer, GeometryFactory
Editor, The CGAL Project

phone: +33.492.954.912    skype: andreas.fabri