From: NAME: Jeremy Dudley
FUNC: Process Technology
TEL: 01793 511 711 x 2005 <DUDLEY@SATURNA1@TITAN>
To: SMTP%"[log in to unmask]"@SATURN@MRGATE@TITAN
The textbooks I used for FORTRAN 77
warned that default SAVE was not part
of the standard, but that so many compilers
supported it, and so many programs made use
of it, that it was unlikely that any new compilers
would change to a non-default SAVE. But it did
advise using SAVE wherever required, to 'future proof'
code and to simplify moving between different compilers/
environments.
Having grown up with DATA I find
REAL :: A = 0
having the defined behaviour logical - I would find
it less intuitive if it had the meaning
REAL :: A
A = 0
because for that I would expect to write the
code as that, making it clear what I was after.
Following the line of my old textbook, I would
actually have gone for
REAL, SAVE:: A = 0
despite the language rules.
I don't think you can say of one contributor
'The standard did not support SAVE of everything -
that was your failure to read the standard'
and then go on to say
'I expected that REAL:: A = 0 should behave in a different
way, and I was suprised to read the standard and
find I was wrong.'
Regards to all, and perhaps now is the time
for the universities to remind us of their
programming courses!
Jeremy Dudley
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|