Jim Ley wrote:
"Jonas Sicking" <[EMAIL PROTECTED]>
Jim Ley wrote:
"Jonas Sicking" <[EMAIL PROTECTED]>
Though the usecase for Accept-Charset seems fairly weak. Why
couldn't the author simply filter responseText?
How would an author do that? Or do you mean the ASCII use case, and
not might accessibility tool use case?
The only use case I've heard so far is the ability to request a
pure-ascii response. Or I guess more generally requesting an encoding
that only contains a specific subset of the unicode map. However I'm
not sure what value this gives over simply looping over responseText
and testing if all chars are in the desired subset.
As part of the W3's EARL work, it's likely I'll be resurrecting some
tools of mine which consume EARL reports about a site and present
information to the user about the problems, these tools either run in
the browser as a script extension, or by being installed on the site
that's tested. To be able to show the error, I need to be able to
repeat the request that was tested, this means sending the same headers
- so I need to be able to achieve that. Of course if the encoding is
not supported then I cannot do anything and expect it to fail, but in
the majority situation lots of encodings are supported so changing
preferences will work.
I'm not sure I understand the use case. Do you need to be able to send
any combination of headers as was repeated earlier by another UA? If so
I don't think we can ever archive that functionality since there will
always be some headers that we want to limit.
/ Jonas