CGAL users discussion list

[thann <>]

Laurent Rineau (CGAL/GeometryFactory) wrote
> I would like to see the full build log (with -v) of one example where the 
> exceptions are caught correctly, and using the same compiler and same
> version 
> of the CGAL library. There must be a difference.

I agree. I revoke my previous statement though that the examples work with
the used build setup. It was merely never the case that the exception was
actually thrown with the input to the triangulation. Forcing the same input
also gives the phenomenon. 

However, I tried Sebastiens (sorry for writing your name wrong earlier)
suggestion to compile the examples via CGALs CMake scripts and the above
mentioned modified example with changed input actually *does* work that way.
I used the arguments of the resulting gcc calls on my example program and
it's working there, too. I think it's some difference in the invoking of the
linker. I might report back if I find the critical difference to be a
non-trivial issue. 

Thanks for your help!



--
View this message in context: 
http://cgal-discuss.949826.n4.nabble.com/Uncaught-CGAL-Uncertain-conversion-exception-tp4656528p4656559.html
Sent from the cgal-discuss mailing list archive at Nabble.com.