On May 14, 7:18 am, GaryScott <garylsc...@[EMAIL PROTECTED]
> wrote:
> On May 14, 1:36 am, robert.corb...@[EMAIL PROTECTED]
wrote:
>
>
>
> > On May 13, 8:05 pm, robert.corb...@[EMAIL PROTECTED]
wrote:
>
> > > On May 13, 6:33 pm, nos...@[EMAIL PROTECTED]
(Richard Maine) wrote:
> > > It does look as if I got it wrong. Here is an example. Given
> > > the program
>
> > > PROGRAM MAIN
> > > OPEN (10, FILE='XXX')
> > > REWIND 10
> > > WRITE (10, '(A)', ADVANCE='NO') 'ABCDEF'
> > > WRITE (10, '(T1, A)') 'GHI'
> > > END
>
> > When I compile and run the program with a handful of different
> > compilers, ifort 8.0 appears to be the odd man out. Do more
> > recent versions of ifort still write
>
> > GHIDEF
>
> > to the file XXX? Do other implementations?
>
> > Bob Corbett
>
> CVF 6.6 produces GHIDEF also. Removing the "T1," above produces
> ABCDEFGHI.
>
> So in the "correct" implentation, how do you produce GHIDEF in a
> standard conforming way?
I could write that way in the first place. If the records
are short enough, I could buffer the output myself and then
write it when it is complete. I would have to know in advance
what the size of the data written would be, which is painful
but not impossible.
Bob Corbett
|
