[Fortran, patch] PR 52158 - Regression on character function with gfortran 4.7

[Janus Weil]

Hi Alessandro,

> my name is Alessandro, I'm a newbie of GCC and helped by Tobias Burnus
> and Paul Thomas I'll try to add support for final subroutines.

since Tobias already reviewed your patch, I just want to say: Welcome
to the team :)

We're certainly looking forward to your contributions! Having final
subroutines in gfortran will be great progress.

Btw, one thing that you may want to look into on your way to FINAL is
proper deallocation of polymorphic variables(cf.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46321). This is one thing
which is still missing in our OOP implementation, and it's serious in
the sense that almost any sensible OOP program produces memory leaks
with gfortran right now. It is not exactly a prerequisite to FINAL,
but at least similar in spirit. Maybe both features can even be
implemented with similar techniques. I actually had some plans to
implement it myself, but currently I'm not able to spend large amounts
of time on gfortran (unfortunately).

In any case, I wish you all the best for your efforts ...

Cheers,
Janus