At 17:17 14-09-2004 -0400, A.D. wrote:
>Hello,
>
>Absoft seems to reject this:
>
>program test
>contains
>subroutine sub(i)
>integer, intent(out) :: i
>write(*,*,iostat=i) "HAHAHA!"
>end subroutine
>end program
>
>with:
>write(*,*,iostat=i) "HAHAHA!"
> ^
>cf90-1640 f90fe: ERROR SUB, File = test.f90, Line = 7, Column = 18
> Dummy arguments with the INTENT(OUT) attribute must be defined before
>use.
>
>Clearly this is a bug, right?
Yes, and my Absoft compiler (Windows, v 7.5) compiles and runs your
program correctly:
D:\Fortran\Error\Absoft>f95 -v iostat.f95
f90fe.exe "@iostat.mds" -Ab -F "D:\Tmp\Abs4F.bif" "iostat.f95"
tmod.exe -hmif=p -o "D:\Tmp\Abs4F.int" "D:\Tmp\Abs4F.bif"
del "D:\Tmp\Abs4F.bif"
ibe.exe -N2 -N76 -N104 -N105 -o "iostat.obj" -i "D:\Tmp\Abs4F.int"
del D:\Tmp\Abs4F.int
lnk.exe "@iostat.amh"
# link warning: file not needed for link - D:\ABSOFT\LIB\F90MATH.LIB
del iostat.obj
>I had no idea how buggy compilers were out there when it comes to new
>F90 features like pointers, derived types, generic resolution, etc. I
>have been using NAG and Lahey mostly and been blessed with stability. I
>cannot get another compiler to even compile correctly my support codes,
>yet alone the actual numerical stuff...
>
>Thanks,
>Aleksandar
\---
Best regards,
Jan van Oosterwijk | "They that can give up essential liberty
Netherlands | to obtain a little temporary safety deserve
| neither liberty nor safety."
| -- Benjamin Franklin (1759)
write(*,*)transfer((/778985834,1869504886,1702130543,1785296754 &
&,1635205227,1868849518,1819160175 /),(/'x'/)) ; end
mailto:Jan <dot> vanOosterwijk <at> wanadoo <dot> NL
http://huizen <dot> dto <dot> tudelft <dot> nl/vanOosterwijk/
|