Closed jbcaillau closed 1 year ago
@jbcaillau hi, I've just pushed that little request, thanks for the idea.
However, it's very limited what could be done without a fortran parser. For instance, if your source compile to several object, all of them will share all the source from the cell.
Thanks for the quick reply.
Yes, sharing all the source from the cell for every compiled can be misleading. Still, one can work piece by piece. All in all, I think it makes sense to keep track of the original code as it allows to iterate static operations on it (automatic differentiation, e.g.)
PS. A more general question: it is a bit strange to me that no one has made for C/C++
a magic as easy to use as the one you did for fortran
. (Of course there are cython
and other stuff...)
PS. A more general question: it is a bit strange to me that no one has made for
C/C++
a magic as easy to use as the one you did forfortran
. (Of course there arecython
and other stuff...)
this attempt exists https://github.com/D3f0/ipython_gcc
@mgaitan Hi, very nice piece of work.
One question: is there any easy way to retrieve the original fortran code, e.g. as a mere
string
, from the generated function ? Something likeThe motivation can be to reuse the source code for further transformations, such as automatic differentiation (e.g. with Tapenade online tool).