[J3] Clarification on F18/017

[Carlson, Neil]

On Mon, Jun 8, 2020, at 2:54 PM, Steve Lionel via J3 wrote:
> So... What is the technical reason for disallowing CBPA and how would
> you change the interp to satisfy the other requirements?

I think the better question is:  Does the order CBPA makes good language design sense? Does it make good sense for A to be finalized after P when its sibling component B is finalized before merely because one is allocatable and one is not?  I find that very surprising and non-intuitive.  Is there a rationale for that other than appealing to minimal disruption for some compilers? (Which I don't find to be a very compelling reason.)

There is no question here of breaking valid code by working toward an alternative solution that disallows CBPA.  The standard was conflicting and there could not have been any such thing as "valid code" in this circumstance; NAG produced CBPA, Intel CABP.

-Neil


---
Neil Carlson, PhD  ·  505.665.6386
ASC Telluride Project Lead
Computational Physics and Methods (CCS-2)
Los Alamos National Laboratory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.j3-fortran.org/pipermail/j3/attachments/20200608/52357a13/attachment.htm>