So I have some existing Perl 5.6 code that I am trying to port to Perl 5.8.

Actually, "port" is the wrong word, I'd like it to be "compatible" 
with both so the same code based works under both versions of Perl.

At a certain point in the program I have a variable that is "mostly 
ASCII", but occasionally contains binary trash.  Not UTF-16 or 
some  other alphabet . . . actual binary random garbage.

Under Perl 5.6 we did the following to suppress out "the bad stuff" 
and make the message printable:

    $errmsgbase =~ tr/[\200-\377]/./; # Kill UFT-8 Problem Characters
    $errmsgbase =~ s/\P{print}/./g;   # Doctor Unprintable Characters

However, under Perl 5.8 print.pm doesn't seem to exist any 
more.  I've read the page on perlunicode, but I cannot say I 
understand it.  It mentions new character classes, but nothing that 
seems similar to "print" that I can tell.

I am not wed to this way of doing things.

Given a string containing random binary garbage, does anyone have a 
way to stamp out the non-printable characters?

It needs to work under both Perl 5.6 and Perl 5.8.

Frank

_______________________________________________
Perl-Win32-Porters mailing list
Perl-Win32-Porters@[...].com
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs