This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: [Patch, Fortran, F08] PR 54756: Should reject CLASS, intent(out) in PURE procedures
- From: Dominique d'HumiÃres <dominiq at lps dot ens dot fr>
- To: Janus Weil <janus at gcc dot gnu dot org>
- Cc: gfortran <fortran at gcc dot gnu dot org>
- Date: Sun, 21 Dec 2014 15:29:52 +0100
- Subject: Re: [Patch, Fortran, F08] PR 54756: Should reject CLASS, intent(out) in PURE procedures
- Authentication-results: sourceware.org; auth=none
- References: <20141221121617 dot 34CFD105 at mailhost dot lps dot ens dot fr> <CAKwh3qhgseGSiCWRhHas4fPBcELg0A56PBTNesoxHTHNU-AE6A at mail dot gmail dot com>
I have not yet tested your patch (for my lay eyes it is in the  obvious  category). Iâll test it as soon as I am done with testing your patch in pr60357!
Cheers,
Dominique
> Le 21 dÃc. 2014 Ã 14:50, Janus Weil <janus@gcc.gnu.org> a Ãcrit :
>
> Hi Dominique,
>
>> Your patch seems related to pr59103 if I am not mistaken.
>
> yes, that PR is essentially about the same thing as PR 54756. However,
> the former lists a few additional cases that should be checked for.
> Those checks are not included in my current patch, but could be
> treated in a follow-up. Rejecting polymorphic DEALLOCATE statements in
> pure procedures should be rather straightforward, but catching "any
> statement that might result in a deallocation" might be a bit harder.
>
> Given the number of testcases that are affected already by the
> INTENT(OUT) part, I'd like to get that out of the way before going
> further. If you have any comments to the patch itself, please let me
> know (I think it's pretty straightforward).
>
> Cheers,
> Janus