Re: Fortran 77 alphabet character $



On May 29, 6:40 pm, har...@xxxxxxxxxxxxx (John Harper) wrote:
In article <1ihp0t1.1wwsw6bist6d2N%nos...@xxxxxxxxxxxxx>,

Richard Maine <nos...@xxxxxxxxxxxxx> wrote:
Herman D. Knoble <SkipKnobleL...@xxxxxxxxxxxxxxx> wrote:

What  does the F77 Standard say about using the character, $ for the
beginning of a variable name?

Nobody seems yet to have actually answered this question. The f77
standard (and all other versions of the standard) doesn't allow it.

I'm afraid I don't have any brilliant suggestions about how to fix such
nonstandard code.

Because f95 allows longer names than f77, if the underscore wasn't in
the professor's character set, you could compile with f95 by changing
the offending $ into D_ at the cost of
(a) continuation lines wherever the code now went past space 72,
or (b) changing to free source form, which IMHO would need more work.


Even better (in the scheme of lesser of evils) would be I_ as a
replacement for a leading $. That way all of these would default to
type INTEGER - removing the need for IMPLICIT INTEGER ($).

- e

.



Relevant Pages

  • Re: Fortran 77 alphabet character $
    ... The f77 ... standard doesn't allow it. ... nonstandard code. ... you could compile with f95 by changing ...
    (comp.lang.fortran)
  • Re: On writing negative zero - with or without sign
    ... Given the present state of the standard, which as we have seen allows ... with gfortran, Sun f95 and Intel f95, ... to make an IEEE-compliant processor write a zero real value ...
    (comp.lang.fortran)
  • NAG, Sun, Compaq possible f95 bug
    ... instead of compiling it. ... NAG say they will fix that bug. ... The certain bug, which Sun and Compaq f95 don't detect, but I don't ... In the f95 standard section ...
    (comp.lang.fortran)
  • Re: upgrading
    ... standard, It has been rewritten in parts to f90 standard. ... Most compilers that are actively developed and supported are Fortran ... so a lack of compilers is not reason to avoid F95. ...
    (comp.lang.fortran)
  • Re: Viewing Double Prescision reals as Integers
    ... I would need to go read section 13 of the standard. ... objects of the specified type and type parameters, ... Both f95 and f2003 then say the result is processor dependent. ... of a model for the representation and behavior of numbers on a processor. ...
    (comp.lang.fortran)