Idea: Google Summer of Code project?

[Christopher D. Rickett]

hi all,

i think that the iso_c_binding part is fairly complete too, though i could
be considered biased.  :-)  i did find a bug in renaming of symbols from
iso_c_binding, but i think i've fixed it.  i wasn't too familiar with the
idea of renaming so i overlooked it.  i hope to submit a patch soon.

i agree with Tobias that the IEEE module would be nice.  however, if there
are a number of outstanding issues with iso_c_binding, i would like to get
them out of the way.  procedure pointers would also be very nice.

Chris

> François-Xavier Coudert wrote:
>> In a last round of mail before I fly away: I have been wondering for a
>> while if it would be worth proposing a Google Summer of Code project
>> (http://code.google.com/soc/) for gfortran. Among the things that
>> might be worth thinking, it could be a project on ISO_C_BINDING,
>> depending on how much time we expect to have to work on ISO_C_BINDING
>> ourselves (mine is very close to zero) and how difficult we expect the
>> remaining tasks to be.
>
> I think the ISO C Bindings are rather complete in the
> Fortran-Experiments branch. I think the only bigger missing part is C
> interoperable procedural pointers, which should be rather simple to
> implement once we have support for procedural pointers in Fortran
> itself. (One application which uses C interoperable procedural pointers
> is CP2k.)
>
> Other nice things:
> The IEEE intrinsic module would also be nice. (I think FX was working on
> this?)
>
> Other than that one could consider the object-orientated enhancements of
> Fortran 2003.
> I'm especially thinking about type-bound procedures:
>
> type foo
> ...
> contains
>   procedure ::
> end type foo
> type, extends(foo) :: bar
> ...
> end type
>
> (Application for SoC begins March.)
>
> Tobias
>
>