printing two strings generated by internal write

The program below, which prints a string created by an internal write,
is compiled without warnings by Compaq Visual Fortran, Lahey/Fujitsu,
and g95. At run time, however, LF95 and g95 crash, whereas CVF gives
the output of "3" that I expect. Is the code ok? When I store
int_to_string(3) in a character variable and then print that variable,
all three compilers work fine.

module yy
implicit none
contains
pure function int_to_string(i) result(text)
integer, intent(in) :: i
character (len=10)  :: text
write (text,"(i6)") i
end function int_to_string
end module yy

program xx
use yy, only: int_to_string
implicit none
print*,int_to_string(3)
end program xx

In article <3064b51d.0411171728.79ecfdf5@posting.google.com>,
<beliavsky@aol.com> wrote:

>The program below, which prints a string created by an internal write,

Doesn't it break the "no recursive I/O" rule?

Works OK in PathScale's compiler...

-- greg

"Greg Lindahl" <lindahl@pbm.com> wrote in message
news:419bff8a$1@news.meer.net...

> In article <3064b51d.0411171728.79ecfdf5@posting.google.com>,
>  <beliavsky@aol.com> wrote:
 

> Doesn't it break the "no recursive I/O" rule?

> Works OK in PathScale's compiler...

Breaks the rule in f95, but not f03.  Some of my code did this
and worked in LF95 & CVF but many other compilers didn't like it.
I had to change it so that it would be more widely useable.

--
write(*,*) transfer((/17.392111325966148d0,6.5794487871554595D-85, &
6.0134700243160014d-154/),(/'x'/)); end

On 17 Nov 2004 17:28:16 -0800, beliavsky@aol.com wrote:

>The program below, which prints a string created by an internal write,
>is compiled without warnings by Compaq Visual Fortran, Lahey/Fujitsu,
>and g95. At run time, however, LF95 and g95 crash, whereas CVF gives
>the output of "3" that I expect. Is the code ok? When I store
>int_to_string(3) in a character variable and then print that variable,
>all three compilers work fine.

As others have noted, Fortran 95 prohibits starting an I/O operation as a
side-effect of a function call in another I/O operation.  F2003 relaxes this
.

Compaq and now Intel compilers currently allow nested I/O as long as differe
nt
units are used, and internal I/O is always allowed.


Steve Lionel
Software Products Division
Intel Corporation
Nashua, NH

User communities for Intel Software Development Products
http://softwareforums.intel.com/
Intel Fortran Support
http://developer.intel.com/software/products/support/