W3C home > Mailing lists > Public > www-international@w3.org > April to June 1999

Re: Form response charset

From: Erik van der Poel <erik@netscape.com>
Date: Tue, 13 Apr 1999 12:56:53 -0700
Message-ID: <3713A185.D5685FE0@netscape.com>
To: Jesse Hall <jesse@Novonyx.COM>
CC: www-international@w3.org
Jesse Hall wrote:
> 
> I'm using UTF-8 for all the pages I send. The browsers I've tested with handle
> this properly. However, what I'm getting back from e.g. a Japanese browser (I've
> tried two) running on Japanese Windows is not UTF-8.

Netscape's client converts the data to the charset of the original HTML
form, and then submits it to the server. The charset of the original
form is determined using the usual method:

1. HTTP Content-Type header's charset parameter
2. HTML META tag's HTTP-EQUIV with charset
3. User's charset menu selection

If the form itself is being spewed out by a CGI, it is usually quite
easy to add the charset parameter to the Content-Type header. Otherwise,
you need to configure the server to spit it out. Failing that, put the
charset as early as possible in the HTML output. The name for UTF-8 is
"utf-8" (can be upper case if you wish). Note that it may still be too
early to rely on UTF-8, since the UTF-8 support in browsers is still
relatively new. Shift-JIS ("x-sjis") is a safer choice for Japanese.

Which browsers fail? OS version number? Browser version number?

Erik
Received on Tuesday, 13 April 1999 15:57:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 19:16:54 GMT