Hi all,
ping on this small one. What do you think? Should I continue working on
OPERATOR and just include it or can I commit this patch as fix to my
already committed OPERATOR support?
Thanks,
Daniel
-------- Original Message --------
Subject: [Patch, Fortran] Small fix for type-bound operators and NOPASS
Date: Tue, 11 Aug 2009 14:10:01 +0200
From: Daniel Kraft <d@domob.eu>
To: Fortran List <fortran@gcc.gnu.org>
CC: gcc-patches <gcc-patches@gcc.gnu.org>
Hi all,
I just started to work on calling type-bound operators, and discovered a
small bug with the last patch. For type-bound operator procedures,
there must be a passed-object dummy argument (makes sense) -- but that
was not checked, and so it was ok to define some procedure as type-bound
operator that has nothing to do with the type at all.
The attached patch fixes that. I'm not sure if we should do a seperate
check-in or just combine it with the large follow-up for calls; I think
it is nothing we need to fix for now, but on the other hand it doesn't
really belong to the calling-patch. So what do you think?
If I should commit this right away, I'll add a ChangeLog and run a
regression-test. Would it then be ok for trunk?
Yours,
Daniel