[Dune] Fwd: Re: [Dune-Commit] dune-common r6512 - trunk/dune/common

[Oliver Sander]

I agree with Markus in considering the change an interface change.
Therefore a little announcement or RFC before the commit would have
been nice.  I actually do think that the change could be quite
helpful.  Still, changing the interface without a RFC before is not
how Dune development should work.

best,
Oliver

Am 14.10.2011 12:17, schrieb Markus Blatt:
> On Thu, Oct 13, 2011 at 10:55:11AM +0100, Martin Nolte wrote:
>>
>>
>> I do not consider this an interface change, since all existing codes
>> will keep working unless you define row_reference to something other
>> than row_type&.
>
> Changing the signature of a function is IMHO an interface change even
> if it might change only future code.
>
>> And I do not intend to do such a thing to the
>> FieldMatrix, which is indeed so deeply in the core of DUNE that such a
>> change is nearly impossible.
>>
>> The reason for the change is to allow containers that store all values
>> in a single array (e.g., a FieldVector). If you do this, the row has to
>> be a dummy object that cannot be referenced (in the standard C++ sense).
>> However, all required information is so small that you can actually
>> store it in the reference itself, once this is allowed to be an object.
>>
>
> And one someone uses that container instead of FieldMatrix things
> might break and without this discussion it would be hard to find.
>
>> So, if you feel like voting on this, go ahead and create a FlySpray
>> entry on this. My vote is clear.
>
> I didn't even call for a vote. My only concern is to raise attention
> of developers and users before such changes. That would have been
> nice, would it not?
>
> The only still unanswered question is: Who will change all occurrences
> of row_type&  to row__reference and so on in the code to prevent
> future code breakage?
>
> Regards,
>
> Markus
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune