[PATCH] fortran/20869 -- INTRINSIC and EXTERNAL conflict

[Steve Kargl]

On Tue, Jan 17, 2006 at 06:53:11AM +0100, Paul Thomas wrote:
> Steve,
> 
> >The attached has been bootstrapped and regression tested on
> >i386-*-freebsd.  There were no new regression.  Unless someone
> >objects, I'll commit the patch to trunk and mainline tomorrow.
> >
> >2006-01-16  Steven G. Kargl  <kargls@comcast.net>
> >
> >	PR fortran/20869
> >	* resolve.c (resolve_symbol): Check for EXTERNAL and INTRINSIC 
> >	conflict.
> >	* gfortran.dg/intrinsic_external.f90: New test.
> >
> symbol.c(check_conflict) already contains a test for EXTERNAL/INTRINSIC 
> clashes.  From a very quick look, something is very wrong with the 
> matching of EXTERNAL and INTRINSIC.   None of the checking invoked by 
> decl.c(gfc_match_intrinsic, gfc_match_external) seems to be done at all 
> - for example, USE association is not picked up.

gfc_match_intrinsic and gfc_match_external work for attributes
in a single statement, ie.,

real, intrinsic, external :: x

will trigger an error.  OTOH, the INTRINSIC and EXTERNAL statements
do not (apparently) go through the check_conflict routine.  I spent
a few hours today trying to see how these statements work and simply
haven't figured out how either is parsed with respect to setting
the attributes.

> It seems to me that resolve.c is the wrong place for checking the 
> atrributes of the same symbol; especially, when this is, in principle, 
> already done for you.

Well, it is obviously not done for us, or I won't have added the
code to resolve.c :-)  I'll look at it some more tomorrow.  Time for
some sleep.

-- 
Steve