[Twisted-Python] Re: Twisted 1.0.4 - `html' .vs. `xhtml'

Brian Warner warner at lothar.com
Tue Apr 22 17:29:41 EDT 2003

Andrew Bennetts <andrew-twisted at puzzling.org> writes:
> Well, links could be more intelligent here, and it seems other browsers
> manage to figure this out just fine.  Neither the presence of a <!DOCTYPE
> ..> declaration or the opening <html> tag seem sufficient for links to
> realise that it is dealing with actual HTML.  Additionally, I wasn't able to
> successfully configure my local links to recognise .xhtml as being
> html/plain content, even though the menus seemed to suggest I could (it
> happily accepted the entry, but the behaviour was unchanged).

Forgive my confusion, but why isn't this sort of thing entirely up to the
server? Shouldn't the client merely obey the Content-Type: header?

I'm willing to believe it's more confusing than that, both because I've seen
it break and because I've seen the zillion mapping preferences in web
browsers, but I'd love to know how it's *supposed* to work before all the
band-aids got applied. In-line magic numbers like <html> or <!DOCTYPE> are
what MIME was supposed to get rid of.


More information about the Twisted-Python mailing list