W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1998

Re: HTTP 1.1 issue 07: 4.4 Message Length

From: Jim Gettys <jg@pa.dec.com>
Date: Wed, 11 Nov 1998 13:42:44 -0800
Message-Id: <9811112142.AA25335@pachyderm.pa.dec.com>
To: Jeffrey Mogul <mogul@pa.dec.com>
Cc: jg@pa.dec.com (Jim Gettys), http-wg@hplb.hpl.hp.com

> From: Jeffrey Mogul <mogul>
> Date: Wed, 11 Nov 98 13:36:22 PST
> To: jg@pa.dec.com (Jim Gettys)
> Cc: http-wg@hplb.hpl.hp.com
> Subject: Re: HTTP 1.1 issue 07: 4.4 Message Length
> -----
>     > In section 4.4 "Message Length", does the statement
>     >    "If a Content-Length header field (section 14.13) is present, its
>     >    decimal value in OCTETs represents both the entity-length and the
>     >    transfer-length. The Content-Length header field MUST NOT be used
>     >    if these two lengths are different (i.e., if a Transfer-Encoding
>     >    header field is present)."
>     > mean that a receiver should ignore Content-Length, or a sender should
>     > not send it?
> 
>     It means that the sender should not send it in the case of a
>     transfer encoded message, since the transfer-length is otherwise
>     encoded and would result in a contradiction in the lenth.
> 
>     I don't see an obvious rewrite to make this more obvious.  Unless
>     there is some concrete suggestion, I plan to leave this one alone.
> 
> "Use the Robustness Principle, Luke!"
> 
> It means both: it means that the sender MUST NOT send it
> and that (in the event of a non-compliant sender) the receiver
> MUST ignore it.  I agree that the verb here ("be used") is
> fuzzy.
> 
> So how about:
>     If a Content-Length header field (section 14.13) is present, its
>     decimal value in OCTETs represents both the entity-length and the
>     transfer-length. The Content-Length header field MUST NOT be sent
>     if these two lengths are different (i.e., if a Transfer-Encoding
>     header field is present).  If a message is received with both a
>     Transfer-Encoding header field and a Content-Length header field,
>     the latter MUST be ignored.
> 
> Maybe a little more rigid than formally necessary, but I think
> we need to take a strong stand against non-compliance in this
> area (as we did with the Host header), or else we could end up
> in a bad mess.
> 

Concrete suggestion gratefully accepted.  My light saber is getting limp
at this date on this spec... :-)
				- Jim
Received on Wednesday, 11 November 1998 21:42:54 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:33:25 EDT