(j3.2006) C1247 and C1248

[Malcolm Cohen]

Van Snyder wrote:
> On Wed, 2009-09-09 at 05:08 -0700, Bill Long wrote:
>>
>> Van Snyder wrote:
>> > It seems that C1248 duplicates C1247, but less precisely.
>>
>>
>> C1248 and C1262 seem redundant as well.
>
> C1262 constrains a different statement from C1247/C1248, so I don't
> think it's redundant.

I agree.  They both constrain the initial statement of a separate module 
procedure (which perforce must use the keyword MODULE in it), but C1248 
constrains <prefix> and thus MODULE FUNCTION and MODULE SUBROUTINE (viz 
<function-stmt> and <subroutine-stmt>) whereas C1262 constrains MODULE PROCEDURE 
(viz <mp-subprogram-stmt>).

Actually, I don't think limiting C1248 to R1225 (<prefix>) is technically 
correct - it is too limited.  But the meaning is clear.  Ideally C1248 and C1262 
would be combined to form a single (non-syntax-rule-limited) constraint, but now 
is not the time for doing that (I would have a different opinion if they were 
not clear).

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