(j3.2006) statement labels and scope

[Bill Long]

On Jan 14, 2016, at 5:19 PM, Robert Corbett <robert.corbett at oracle.com> wrote:

> On 01/14/16 12:56, Bill Long wrote:
>> In 3.2.5 Statement labels we say that a statement label shall not be given to more than one statement ?in its scope?.  Since BLOCK constructs are scopes, if this is taken as stated, the following
>> 
>> subroutine sub
>> 10 continue
>> 
>> block
>> 10 continue
>> end block
>> 
>> end subroutine sub
>> 
>> should be conforming.  In, instead, we actually meant ?inclusive scope?, then it would not be.  Personally, I think that allowing this is preferred, and would not require a wording change.   But I?m checking because I see inconsistent behavior with different compilers.  (And this is in one of my clauses to read this time).
> 
> We did mean inclusive scope.  The section on scopes makes this clear.
> 
> Subclause 16.1 of 10-007r states
> 
>       The scope of ... a local identifier is its inclusive scope.
> 
> Subclause 16.3.1 says that statement labels are local identifiers.

unless they are construct entities. 

> Subclause 16.4 does not say that statement labels can be statement
> or construct entities.

So, we conclude that either 3.2.5 or 16.4 is wrong.   

Note, unrelated, for the group reading 16:  It looks like SELECT RANK got overlooked in several places in 16.4.

Cheers,
Bill




> 
> Bob Corbett
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3

Bill Long                                                                       longb at cray.com
Fortran Technical Support  &                                  voice:  651-605-9024
Bioinformatics Software Development                     fax:  651-605-9142
Cray Inc./ Cray Plaza, Suite 210/ 380 Jackson St./ St. Paul, MN 55101