[Info-vax] Fortran

[Simon Clubley]

On 2018-12-05, Arne Vajhøj <arne at vajhoej.dk> wrote:
> On 12/5/2018 8:46 AM, Scott Dorsey wrote:
>> Simon Clubley  <clubley at remove_me.eisner.decus.org-Earth.UFP> wrote:
>>> Removing octal constant support was a bit strange unless there's
>>> something I am not understanding. Like you, I don't really see any
>>> reason why it could not have been left in place. This is different
>>>from the removal of Hollerith support where the removal of Hollerith
>>> support brings distinct safety benefits.
>> 
>> Hmm... could this possibly be due to a change in variable naming rules, that
>> what previously had been valid octal constants were now valid variable
>> names?
>
> My Fortran is getting a bit rusty, but as I recall it the syntax is
> like O'177', which is not a valid name and most likely never will be.
>

I've had a quick look at this as well, because it didn't seem
right to me either on further reflection as I took Dan's comment
as meaning literally that octal constant support had been removed
from the latest version of Fortran, which was something I was
not previously aware of.

It looks like the situation is more complicated.

Normal octal constant support is still in the latest version of
Fortran, at least according to the draft PDF posted as part of
the initial post in this thread.

However, there's a different type of Fortran octal constant, with
different semantics, designed for PDP-11 Fortran compatibility
support and I now suspect this (and the removal of this PDP-11
compatibility support) is what Dan was referring to.

Simon.

-- 
Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world