Re: HTTP header 'Accept' should be sent by Validator

Tim Bagot wrote:
> At 2003-01-10T06:06-0000, Tom Pike wrote:-
>>The validator does not seem to send an 'Accept' HTTP header.
>>Therefore any code which checks the Accept header to determine what type
>>of document to send will assume that the validator doesn't accept
>>documents of the type that they are checking for (in my case
> It doesn't make much sense to me for the validator to use content
> negotiation. I'd normally want to validate a specific file (and typically
> each variant in turn), and the server should make each variant available
> on its own URI.
> Tim Bagot
Yes actually now that I think about it, it does make a lot more sense to 
do it that way.  I'm now doing things a little differently.

Also I wasn't aware that sending no Accept header is equivalent to 
sending "*/*;q=1.0" (didn't spot that in the HTTP specs) so this entire 
thread is somewhat irrevelant anyway.  Sorry about that.

Tom Pike

Received on Friday, 10 January 2003 12:56:11 UTC