Win32 signals between parent/child

jrsimmon has asked for the wisdom of the Perl Monks concerning the following question:

Monks- The following code works perfectly well on UNIX (5.005_03for PA-RISC2.0), but fails on Win32 (5.8.8 for MSWin32-x86-multi-thread). On Win32, the signal handler is never called. Any suggestions?

$SIG{INT} = sub {syswrite(STDERR, "$$:\tChild has signaled.\n", 27);
};

$parent = $$;
print "$$:\tI am the parent.\n";

unless($pid = fork){
  print "$$:\tI am the child.\n";
  print "$$:\tSending SIGINT to $parent.\n";
  kill INT, $parent;
  exit;
}

print "$$:\tWaiting for the child to signal.\n";
print "." for(1..100);
print "Did I catch it?";
[download]

Comment on Win32 signals between parent/child Download Code

Replies are listed 'Best First'.
Re: Win32 signals between parent/child by TGI (Parson) on Aug 30, 2007 at 21:58 UTC
kill doesn't work the same on win32. See the entry in perlport. TGI says moo	[reply]
Re^2: Win32 signals between parent/child by jrsimmon (Hermit) on Aug 31, 2007 at 14:31 UTC
Ok, that explains that. Is there a recommended alternative for simple parent/child communication? I had hoped to avoid using sockets, but I'm not seeing any other alternatives...	[reply]
Re^3: Win32 signals between parent/child by BrowserUk (Patriarch) on Aug 31, 2007 at 15:16 UTC
As I pointed out above, the first problem with your OP code is that 'INT' is not defined anywhere, so your attempt to `kill INT, $pid;` is just a syntax error. Are you looking for a win32-only solution or must it be cross platform. Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice. "Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."	[reply] [d/l]
Re^4: Win32 signals between parent/child by jrsimmon (Hermit) on Aug 31, 2007 at 18:01 UTC
Re^5: Win32 signals between parent/child by BrowserUk (Patriarch) on Aug 31, 2007 at 18:24 UTC
Re^4: Win32 signals between parent/child by jrsimmon (Hermit) on Aug 31, 2007 at 20:08 UTC
Re^5: Win32 signals between parent/child by BrowserUk (Patriarch) on Aug 31, 2007 at 20:50 UTC
Re: Win32 signals between parent/child by BrowserUk (Patriarch) on Aug 30, 2007 at 22:54 UTC
Add `use strict;` to your snippet to see the cause of your problem. Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice. "Too many [] have been sedated by an oppressive environment of political correctness and risk aversion."	[reply] [d/l]
Re: Win32 signals between parent/child by suaveant (Parson) on Aug 30, 2007 at 18:25 UTC
I can't actually say as I am not a windows programmer, but I can offer a possible alternative... If you have an application where your child is simply waiting you could use a socketpair instead of a signal. Not only is it probably safer it is much more robust, and allows to to send info back and forth between the two. Obviously it doesn't help when you want to interrupt something, but often that is not necessary. - Ant - Some of my best work - (1 2 3)	[reply]