(j3.2006) Storage association

[Van Snyder]

On Wed, 2016-09-07 at 19:24 -0700, Robert Corbett wrote:
> I did not suggest a change to the standard.  I said that if sequence 
> types had been deleted from the standard, then the change Van proposed 
> might be OK.  That is not a call for deleting sequence types from the 
> language.

I wasn't referring to "I might agree ... if sequence types had been
deleted...."  I was referring to "I think it would be preferable to say
it specifies 'the' storage sequence for objects of that type...."

If that were done, and what we presently call character association for
character arguments were described as sequence association (which it
actually is), could we print all of the storage association material in
obsolescent font?

> 
> Bob Corbett
> 
> On 9/7/2016 12:08 AM, Van Snyder wrote:
> > If we change the description in 4.5.2.3 of 16-007r1 as Bob suggests, 
> > and describe the form of storage association for character arguments 
> > as sequence association (which it is), can we print all of the storage 
> > association material in obsolescent font? On Wed, 2016-09-07 at 00:03 
> > -0700, Robert Corbett wrote:
> >> On 9/2/2016 11:06 AM, Van Snyder wrote:
> >>> On Fri, 2016-09-02 at 03:33 -0700, Robert Corbett wrote:
> >>>> On 9/2/2016 12:36 AM, Van Snyder wrote:
> >>>>> Is there a way to establish storage association other than by 
> >>>>> equivalence, common, or entry? 
> >>>> Yes. See, for example, paragraph 8 of Subclause 16.5.3.4 of 16-007r1. 
> >>> Before having this, I though all of the material on storage 
> >>> association and storage sequence could be printed in obsolescent 
> >>> font. Now it seems that only most of it can be printed in 
> >>> obsolescent font. For example, 19.5.3.2p2(1,2,6-8), and maybe 
> >>> 19.5.3.2p2(5), could be printed in obsolescent font. 
> >> I shall assume Clause 19 is the same as Clause 16. I might agree with 
> >> you if sequence types had been deleted from the standard. The second 
> >> sentence of paragraph 1 of Subclause 4.5.2.3 of 16-007r1 states The 
> >> order of the component definitions in a sequence type specifies a 
> >> storage sequence for objects of that type. I think it would be 
> >> preferable to say it specifies "the" storage sequence for objects of 
> >> that type, because I do not think the text anticipates there being 
> >> more than one storage sequence. Bob Corbett 
> >> _______________________________________________ J3 mailing list 
> >> J3 at mailman.j3-fortran.org 
> >> http://mailman.j3-fortran.org/mailman/listinfo/j3 
> > _______________________________________________ J3 mailing list 
> > J3 at mailman.j3-fortran.org 
> > http://mailman.j3-fortran.org/mailman/listinfo/j3
> 
> _______________________________________________
> J3 mailing list
> J3 at mailman.j3-fortran.org
> http://mailman.j3-fortran.org/mailman/listinfo/j3