SPUG: Understanding $<, $> and modules

[John Costello]

I'm puzzling over how Perl handles changes to $< and $> when run as root.

What I'm trying to do:  Drop privileges to a non-root user, do some 
things, return to root privileges.  There may be better ways to do all 
this, but I'm more interested in why Perl behaves the way it does.

If I run this from a perl script


	print "RUID $< and EUID $>\n";
	$< = $> = 8000;
	print "RUID now $< and EUID now $>\n";
	$< = $> = 0;
	print "RUID set to $< and EUID set to $>\n";

I get 

	RUID 0 and EUID 0
	RUID now 8000 and EUID now 8000
	RUID set to 8000 and EUID set to 8000

which is what I expect.

If I dump

	$< = $> = 8000;

into a module that I call from the main script, so that I now have this 
script

	print "RUID $< and EUID $>\n";
	$results = local::SetPerms->changed_ruid_euid();
        print "RUID now $< and EUID now $>\n";
        $< = $> = 0;
        print "RUID set to $< and EUID set to $>\n";

I get 

        RUID 0 and EUID 0
        RUID now 8000 and EUID now 8000
        RUID set to 0 and EUID set to 0

which I didn't expect, because I don't know what is going on under the 
hood.

So, do modules have their own $< and $>, copied from the main script?  Are 
modules run as separate processes?  Where would be a good spot to start 
reading (Programming Perl?)?

John