Re: backslash-issues

In article <1120155872.32886@xxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Bart Vandewoestyne <MyFirstName.MyLastName@xxxxxxxxxx> wrote:

> > character(len=*), parameter :: something= "\\"
> > character(len=*), parameter :: backslash=something(1:1)

> When I compile this with NAGWare Fortran 95 compiler Release 5.0(253) i get:
>
> test_backslash.f95:3:1: warning: unknown escape sequence: '\350'

That rather surprises me. The message makes no sense (there isn't an
octal 350 character around anywhere) and doesn't look like something I'd
expect from NAG. If I've got it right, octal 350 would be an "h" with
the high order bit set, which just doesn't seem to fit.

I just tried this (Nag f95 5.0(322) on OSX) and it worked as expected. I
even tried forcing nag to run fpp on it, just in case the problem was in
there. Still worked fine. Makes me wonder whether a stray character got
into the source file somehow.

You mentioned NAG 5.0(253), but I didn't notice which platform. NAG
supports quite a few and every once in a while problems are unique to a
specific port.

--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain | experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
.