Re^9: How to find perl line after segfault.

[BrowserUk]

But then when it segfaults I wont be able to manage the crash,

It would be no different if you used open3(). That also starts a separate process over which you would have no control.

[Steve_BZ]

Well, with open3(), I can use waitpid and then read (error); which allows me some control over what happens next, if I just did

system ("perl -d:CallTrace Foo.pl");

Wouldn't it just crash? Maybe not, I should probably try it.

Regards

Steve.

[BrowserUk]

Well, with open3(), I can use waitpid and then read (error); which allows me some control over what happens next, if I just did system ("perl -d:CallTrace Foo.pl"); Wouldn't it just crash?

It will crash either way, the only differences is that with Re: Catching errors (II)., you can get the error code (and, on *nix at least, the 'signal value'); and then read the output from stderr.

If you use my suggested system q[perl -d:CallTrace Foo.pl 2>trace.txt];, then you can also read back the stderr output, by reading the file.

You're also more likely to get the last few lines, because you won't have the double-buffering effect you get when you use pipes.

Also, if the exit code is important to you, then (from perldoc perlfunc system ):

If you'd like to manually inspect system's failure, you can check all 
+possible failure modes by inspecting $? like this:
    if ($? == -1) {
        print "failed to execute: $!\n";
    }
    elsif ($? & 127) {
        printf "child died with signal %d, %s coredump\n",
            ($? & 127),  ($? & 128) ? 'with' : 'without';
    }
    else {
        printf "child exited with value %d\n", $? >> 8;
    }