Characters Standart between Pc and Mac

Question

Hi,
My application which is parsing html pages and output some part of this pages has no problem to manage the different charset when used on a PC with J2re 1.4.1 .
But this application used on a Mac Os X with J2re 1.3 output many wrong characters even if I convert all the string output in ISO-8859-1 ( even if this conversion solves many problems). I format the String with this line :
test= new String(old_test.getBytes(), "ISO-8859-1");
The only way I've found is to isolate and convert the wrong characters one by one by this way :
test = test.replace('\u009c','\u0153'); // for example
Is there any global alternative solution ?
Thanks,
Thieum

Answer 1

First, thanks for your responses.
Few precisions :
I mainly handle French web pages, so UTF-8 is not efficient to manage all the different characters, and even Mac Roman is not able to do it. I think that Apple Java Virtual Machine doesn't manage the incoming data like the Windows one, because on PC, I don't have this charset problem.
Brumbaer, your proposition of adapting the conversion system to the charset is intersting, but I didn't found an efficient method to detect the charset of an HTML document that need to stay in HTML to be parsed. The problem is that the getEncoding method return null most of the time, because it is not required to specify the charset in an HTML document.
Maybe the only solution is to wait for an apple java virtual machine as efficient as the windows one to manage the charsets. But maybe there's another solution. to be continued...
Thanks,
Thieum