[Dune] ISI final suggestion:

Andreas Dedner dedner at mathematik.uni-freiburg.de
Mon Jul 3 11:59:56 CEST 2006


your right we should fix some semantics. Some
suggestions:
1) if !en.leaf() then en.leafbegin()==en.leafend()
2) I would keep everything at the boundary as is, i.e.,
   both iterators return the boundary intersections and
   one can use the isBoundary method. At least for the
   leaf Iterator this is the desired behavior; for multigrid
   I don't know...
3) we can still include a method
   bool hasBoundaryInterstion()
   on the entity class - with a default implementation

Andreas
Peter Bastian wrote:
> I am still not clear about what happens with boundary intersections. Are
> they delivered by both intersection iterators?
> 
> -- Peter
> 
> Andreas Dedner wrote:
> 
>> I want to suggest, to change the interface
>> as given below. It seems to me that we can manage
>> this for 1.0, so I will change the interface
>> at the beginning of next week if I don't get any
>> negative feedback.
>> New interface:
>>
>> Dune Entity Class:
>>   IntersectionIterator ibegin () const DUNE_DEPRECATED
>>     {
>>       return realEntity.ibegin();
>>     }
>>   IntersectionIterator iend () const DUNE_DEPRECATED
>>     {
>>       return realEntity.iend();
>>     }
>>   LevelIntersectionIterator libegin () const
>>     {
>>       return realEntity.libegin();
>>     }
>>   LevelIntersectionIterator liend () const
>>     {
>>       return realEntity.liend();
>>     }
>>   LeafIntersectionIterator leafibegin () const
>>     {
>>       return realEntity.leafibegin();
>>     }
>>   LeafIntersectionIterator leafiend () const
>>     {
>>       return realEntity.leafiend();
>>     }
>>
>>
>> _______________________________________________
>> Dune mailing list
>> Dune at dune-project.org
>> http://www.dune-project.org/cgi-bin/mailman/listinfo/dune
>>
>>
> 




More information about the Dune mailing list