> > Right.... apart from that tiny little nagging problem being that the > charset parameter on application/* is ILLEGAL.
Perhaps... Anyway, the point is that "url-encoded" implies US-ASCII... Its common for user agents to ignore that, but since there is no mechanism for specifying the character set of a URL it is pretty much mandated. So if a form is REALLY URL encoded, then it must be ASCII, mustn't it? > > > > Except that this is not possible, at least not if you intend to do > > > anything meaningful with the data. > > > > That might be overstating it a bit. As long as you don't have to convert > > it to any other charset you can still use the information. Honestly I > > wonder how big a problem this is for most applications. I cannot recall > > having serious problems with accepting fairly simple form input wrt > > encodings. > > And how do you use textual data without knowing which charset it is in? You > can't, you don't. In fact, without knowing the charset you're really > dealing with raw binary data, and all the power of Perl can't help much. > I've had those problems, they exist. Well, I have yet to have a form submitted to me and I wasn't able to process it. Doubtless it has happened somewhere to someone, but I've built and deployed at a rough guess 40 web applications over the last 8 years. Maybe I'm just lucky ;o). --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
