The first step to solving the issue would be to compose a self-contained snippet of code which shows how you've tried to use Encode (which I suppose you're referring to). Then post it here. That way we won't have to guess what you might have done wrong...

Well ... the first thing to realize is that all iso-8859-2 characters can be converted to utf-8 but the reverse is not true (for example, japanese glyphs cannot be converted to iso-8859-2).

I'd recommend reading Perl Programming/Unicode UTF-8. It's mostly about iso-8859-1 <-> UTF-8 in a CGI::Application/Apache stack but it's probably the best written article about encoding issues and I'm sure you can glean some relevant tips.

well, I do just

$ret = from_to($ret, "iso-8859-2", "utf8");
[download]

no magic. I fetch webpage, cut sime content and want to display it. but this one does not work : /

Apart from the fact that you haven't provided sample data or enough code to demonstrate what you mean by "does not work", you either didn't read or didn't understand what the Encode man page says about the "from_to()" function: from_to() returns the length of the converted string (in octets), after doing the conversion "in-place" (i.e. replacing the original content of a string variable with the converted content).

This one line of code you showed us is assigning the return value (length in octets) to the same variable that you are trying to convert. That obliterates whatever string you just converted, replacing it with a number.

...no magic. I fetch webpage, cut sime content and want to display it. but this one does not work : /

That is not adequate to replicate your problem. How (Not) To Ask A Question

Show code :)