Perl's handling of foreign (Polish) characters?

[CaptainPinko]

I am try to write a script to anglicise Polish text by replacing accented characters with their english counterparts but I am encountering so really odd behaviour.

This code (coloured to look like Kate):

sub anglicise ($){
	$_ = $_[0];
	print  "ang: $_\n" ;
	tr/?????Ó????????ó???/ACELNOSZZacelnoszz/ ;
	print  "anged: $_\n" ;
	return;
}  #end sub anglicise ()
print  "returned: " .anglicise ("?????ó???");
print  "\n" ; exit(0);

produces this:

ang: ?????ó???
anged: AzAzAzSzSCczSzSzSz
returned:

[edit]

<rant>
*sigh* this site doesn't seem to like foreign characters either... but then again this is not surprising since it doesn't even specify a character set:

<meta http-equiv="MSThemeCompatible" content="Yes">

That "Yes" part made me laugh if it wasn't so sad.
</rant>


The garbled part of the tr// is accented versions ACELNOSZZ, the upper case followed by the lower case. The arguments to anglescise() are just the lower letters. The string starting with "ang: " is all the lower case letters correctly displayed. The string starting with "anged: " is how it actually appears and there is nothing after the "returned: " strangely.