Re: [tlug] Re: font/char set question

["Birkir A. Barkarson" <birkirb@example.com>]

Edmund Edgar wrote:
> Just wondering: Do you, or does anyone else, maintain a publicly 
> available list of wierd hyphens or other Unicode characters that don't 
> strictly speaking map neatly back to anything in Shi(f)t-JIS, but in 
> practice can be converted to something that does? (Encapsulated in a 
> neat little class representing legacy-compatible-UTF8 strings would be 
> best...)

Don't know if it is relevant to your situation, but I recalled solving
this issue in java when working with shiftjis oracle databases.
You can see the relevant unicode names and codepoints below.

/**
 * Oracle の SJIS のＤＢテーブルのから Java Unicode のエンコディング変更直し
 *
 * @param data Oracle SJIS DB　example.com
 * @return　example.com
 */
public static String fixOracleConversions(String data)
{
	if(data.indexOf('\u301C') != -1)
	{
		// 文字　example.com (?) を　example.com@(～)　example.com
		return data.replace('\u301C','\uFF5E');
	}
	
	return data;
}