Re: [eigen] Initial implementation of tensor support

[Christoph Hertzberg <chtz@xxxxxxxxxxxxxxxxxxxxxxxx>]

On 30.10.2013 20:02, Jim Garrison wrote:
> On 10/30/2013 03:20 AM, Christoph Hertzberg wrote:
> > Generally, I agree that we should use more C++11 features in the future.
> > It does however bear the risk that projects using those features
> > immediately lose compatibility to older compilers (which unfortunately
> > are still common e.g. for embedded systems) -- that's especially the
> > case if we introduce new core functionality depending on C++11.
>
> Actually, this is one of the reasons I suggested using a #define to
> enable Eigen's C++11 features.  If one must manually define
> EIGEN_ENABLE_CXX11, it will remove the risk that a project begins using
> C++11 features without realizing it.

Yes, that's actually a good point. There were some points, where we need 
to detect C++11 to circumvent incompatibilities (something about 
allocators, IIRC).
So maybe an only user-defined EIGEN_ENABLE_CXX11 and an auto-detected 
EIGEN_HAS_CXX11.

> Another issue is that some compiler versions claim to support C++11
> through their __cplusplus macro, but do not support all C++11 features.

Yes, I'm afraid that some problems will appear, once more people start 
experimenting with C++11 and with more exotic compilers (AFAIK, no major 
compiler has it enabled by default as of now?)


BTW: Please just answer to the list, I'm getting every mail twice ...

Christoph



--
----------------------------------------------
Dipl.-Inf., Dipl.-Math. Christoph Hertzberg
Cartesium 0.049
Universität Bremen
Enrique-Schmidt-Straße 5
28359 Bremen

Tel: +49 (421) 218-64252
----------------------------------------------