#include <stdio.h>

int main(void)
{
FILE *fp;
/* char *line;  updated correction */
char line[256];   

if (mkfifo ("age-pipe", 0755) != 0)
{printf ("Could not make this fifo\n");}
else {printf ("FIFO was successfully made!\n");}

fp = fopen ("age-pipe", "r");

for(;;){
fgets (line,256, fp);
printf ("message: %s\n", line);
}
close (fp);
}
[download]

#!/usr/bin/perl
use strict;
use warnings;

open(FIFO, "> age-pipe") or die $!;

select FIFO; $|=1;
select STDOUT;$|=1;

my $input = 'a';
while(1){
print "Enter your age:\n";
#my $input = <STDIN>;
print FIFO "$input\n";
sleep(1);
$input++
}

close(FIFO);
[download]

When I run your example I get

~/tmp/cperl> ./a.out
FIFO was successfully made!
Segmentation fault
[download]

I get the segmentation fault (from the c program) when I start the perl script :(

Anyway, what is so different about this approach and simply using the number 63 ?

LuCa

Yeah, thats weird. I probably cheated on the way line is defined, and my compiler automagically fixed it. Instead of

char *line;
 /* try  */
char line[256];
[download]

C is very finicky about allocating space for strings, and will segfault without telling why. This is one of the primary reasons people like Perl over C.

Try this free book: Advanced Linux Programming and read the chapter on IPC.

The difference between my script and your's using &63, is that mine uses a named-pipe which can be accessed by different process. Using &63, you are using a shared filehandle referenced by it's fileno, and requires that both scripts share them..... that means 1 script must fork off the other, instead of them being started separately. Of course, I'm not an expert c programmer, so take that with a grain of salt. This isn't the place to discuss c anyways.

---

I'm not really a human, but I play one on earth. Cogito ergo sum a bum