Re: Handling of charset

Hello Olly,

In our previous episode, Olly Betts said:
 
> However, Amaya seems to totally ignore the charset parameter on the HTTP
> Content-Type header and always assumes iso-8859-1.  This seems to contravene
> the HTTP/1.1 spec (RFC 2068), section 3.7.1:
> 
>   "HTTP/1.1 recipients MUST respect the charset label provided by the
>   sender; [...]"
> 

In its current state, Amaya doesn't interpret all of the HTTP headers
which describe the body, and in particular, the charset header. On the
other hand, Amaya interprets the charset when it's specified inside the
HTML document. We're currently upgrading our HTTP interface to improve
this.

> Admittedly Amaya's HTTP request says HTTP/1.0, but the documentation on
> www.w3.org appears to say it supports HTTP/1.1.

What kind of platform are you  using? Amaya does send HTTP/1.1 requests.
Perhaps you're going thru a proxy?

> Also, Amaya doesn't recognise "¨", and the warning message displayed to
> the console reports 'Invalid entity "&um"' (i.e. it misses off the 'l').
>

That's right! This bug has been fixed up and will disappear in the next
release.
  
> Here are some example pages to play with:
> 
> http://noxious.muscat.com/~olly/entity/
> 
> This has the same sample text available with the server configured to report
> different character sets.
>

Thanks for the URL :) I'll use in my tests.

Cheers,

-Jose

Received on Friday, 28 November 1997 08:57:23 UTC