This is the mail archive of the
mailing list for the GCC project.
Re: [Patch, Fortran] F2008: Implied-shape arrays
- From: Daniel Kraft <d at domob dot eu>
- To: Mikael Morin <mikael dot morin at sfr dot fr>
- Cc: Fortran List <fortran at gcc dot gnu dot org>, gcc-patches <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 13 Aug 2010 09:32:01 +0200
- Subject: Re: [Patch, Fortran] F2008: Implied-shape arrays
- References: <4C627E7E.firstname.lastname@example.org> <4C64AE9E.email@example.com>
Mikael Morin wrote:
Le 11.08.2010 12:42, Daniel Kraft a écrit :
Hi all,This is OK, yes.
while waiting review on my ASSOCIATE stuff (Tobias promised to do so
when he comes back), here's another patch implementing a small feature;
namely, Fortran 2008's implied-shape arrays. This seems particularly
useful in some places to me, although it is just a minor one (I would
have liked to use it in real-world code already myself).
Basically, everything is handled in the parser / matcher; an
implied-shape array gets transformed into an explicit one as soon as the
initializer is known. The only small complication is that when parsing
the array-spec in case of a DIMENSION attribute, we may not already know
whether this is a PARAMETER or not -- and thus, if (*) should be
implied-shape or assumed-size. But I think the handling in my patch
should be rather clear and still simple.
No regressions on GNU/Linux-x86-32. Ok for trunk?
Thanks for the review, committed as rev. 163221. I'll update the wiki
now as appropriate.
To go: Hea-Kni-Mon-Pri