(j3.2006) Should parent component have the accessibility of theparent type?

mark leair PGI mark.leair
Mon Jan 25 05:36:35 EST 2016


The PGI compiler does not currently allow This%B_t%A_t%i = 4 in the example below (but it does allow This%B_t%i = 4).

I agree that it would be nice to have this clarified in F2015.

-Mark


From: j3-bounces at mailman.j3-fortran.org [mailto:j3-bounces at mailman.j3-fortran.org] On Behalf Of Malcolm Cohen
Sent: Saturday, January 23, 2016 10:52 PM
To: fortran standards email list for J3
Subject: Re: (j3.2006) Should parent component have the accessibility of theparent type?

Hi Daniel,

I think this is referring to the accessibility of the parent type where it is defined, not where the extension is defined.  Otherwise, as you say, it does not make much sense.  That is certainly how I interpreted it (NAG compiles your example without complaint).  I just cannot imagine how it could make sense for an ancestor component to be popping into and out of visibility as you go down the type tree ? either it?s visible all the way down, or not visible at all.

Unfortunately the paper trail is not exactly clear ? paper 03-224 adds this specification as an editorial fixup with no indication of any supposed technical effect whatsoever.  Grrr.

It might be a good idea to clarify this in F2015.

Cheers,

From: Daniel C Chen<mailto:cdchen at ca.ibm.com>
Sent: Friday, January 22, 2016 11:26 AM
To: j3 at mailman.j3-fortran.org<mailto:j3 at mailman.j3-fortran.org>
Subject: (j3.2006) Should parent component have the accessibility of theparent type?


Hello,

Consider the following code:

Module A
private
Type A_t
Integer i
End type
Public A_t
End module

Module B
Use A
private
Type, extends(A_t) :: B_t
End type
Public B_t
End module

Module C
Use B
Type, extends(B_t) :: C_t
End type
Contains
Subroutine sub(this)
Type(C_t) :: this
This%B_t%A_t%i = 4 !! NOT allowed
This%B_t%i = 4 !! Allowed
End
End module

The standard states that the parent component has the accessibility of the parent type as follows: [16-007-p80: 2-6]
"An extended type has a scalar, nonpointer, nonallocatable, parent component with the type and type parameters
of the parent type. The name of this component is the parent type name. It has the accessibility of the parent
type. Components of the parent component are inheritance associated (16.5.4) with the corresponding components
inherited from the parent type. An ancestor component of a type is the parent component of the type or an
ancestor component of the parent component."

In the above example, "This%B_t%A_t%i" is not allowed because the parent component "A_t" of "This%B_t" has the accessibility of type A_t, which is private in Module B so not accessible in module C. On the other hand, "This%B_t%i" seems fine.

Is it the intention of the standard to disallow "This%B_t%A_t%i"?

Thanks,

Daniel

XL Fortran Development - IBM Toronto Software Lab
Phone: 905-413-3056
Tie: 969-3056
Email: cdchen at ca.ibm.com<mailto:cdchen at ca.ibm.com>
http://www.ibm.com/software/awdtools/fortran/xlfortran

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

________________________________
_______________________________________________
J3 mailing list
J3 at mailman.j3-fortran.org<mailto:J3 at mailman.j3-fortran.org>
http://mailman.j3-fortran.org/mailman/listinfo/j3

-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.j3-fortran.org/pipermail/j3/attachments/20160125/0728eea7/attachment-0001.html 



More information about the J3 mailing list