(j3.2006) (j3-members.2006) procedure pointer components in sequence types?

[Malcolm Cohen]

I think Bob sent this to the members-only list by mistake...

...but anyway, the existing text in 4.5.2.4 seems pretty much ok to me.  As in, 
components have to agree in "attributes", and attributes covers practically 
everything that can be said about something.  I mean, attribute is
  "property of an entity that determines its uses"
which seems to cover everything one could think of.

I see no immediately-obvious problem caused by procedure pointer components 
here.  If Bob (or anyone else) thinks there is something ambiguous or unclear, 
we can maybe improve the wording, but I'd like to see a problem before I try to 
fix it.

Cheers,

-----Original Message----- 
From: Robert Corbett
Date: ?? 25?11?2? 8:02
To: j3-members at j3-fortran.org
Subject: (j3-members.2006) procedure pointer components in sequence types?

I find nothing in the standard that prohibits sequence types from having
components that are procedure pointers.  A derived type definition that defines
a sequence type cannot have a type-bound procedure part, but procedure pointer
components are declared in the component part.  Did I miss a restriction?

If a sequence type can have a component that is a procedure pointer, the
definition of what it means for two sequence types to be the same type needs
some work.

Bob Corbett
_______________________________________________
j3-members mailing list
j3-members at mailman.j3-fortran.org
http://mailman.j3-fortran.org/mailman/listinfo/j3-members

________________________________________________________________________
This e-mail has been scanned for all viruses by Star.
________________________________________________________________________

-- 
................................Malcolm Cohen, Nihon NAG, Tokyo.