[Bug fortran/34960] Statement starting in wrong column causes Unclassifiable Statement error in the preceding executable line

["kargl at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #4 from kargl at gcc dot gnu dot org  2008-01-25 04:04 -------
(In reply to comment #2)
> The example program was just to show a few variations. The program I was
> debugging only contained the second variant ("Unclassifiable statement error",
> followed by a lot of comment lines, followed by the actual line that contained
> the error). In other words, there were no cascading errors in the original
> code, since there was only one error.
> 
> If you edit the example to only contain one of the variations,
> you'll still get
> the associated error. If I were to do this to reflect my original error, I'd
> come up with this:
> 
>        FUNCTION TEST()
>        INTEGER IVAR1, IVAR2 
> 
>        IVAR1=1 ! "Unclassifiable statement" on this line
> C
> C      Many comment lines in between.
> C
>       IVAR2=0
> 
>        END

'Garbage in, garbage out'

        FUNCTION TEST()
        INTEGER IVAR1, IVAR2 
        IVAR1=1
       IVAR2=0
        test=1.
        END

The comment lines are a red herring.  You're asking the compiler to append
VAR2=0 to the preceding line.  The compiler can't classify 'ivar1=1var2=0'.
So, the compiler is giving up and tell you exact where you made your mistake.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34960