W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1997

Editorial Nit in section 4.1 (new editorial issue GENERIC_MESSAGE)

From: Jim Gettys <jg@pa.dec.com>
Date: Sun, 16 Nov 1997 17:06:07 -0800
Message-Id: <9711170106.AA05351@pachyderm.pa.dec.com>
To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
For completeness.
				- Jim


attached mail follows:


I'm not sure what the status of this is r.e. the current draft,
but the BNF is correct.  The generic message parser needs to look
for zero or more header fields.  It may later complain about the
lack of Host or Date, but that is a different issue (different parser).

....Roy


------- Forwarded Message

Message-ID: <2FBF98FC7852CF11912A000000000001076C6A6D@DINO>
From: "Dean Justus (Exchange)" <deanj@exchange.microsoft.com>
To: "'fielding@ics.uci.edu'" <fielding@ics.uci.edu>
Subject: rfc2068
Date: Fri, 14 Nov 1997 11:15:25 -0800

Mr. Roy Fielding;

I do not know if this is an appropriate manner in handling this issue,
but in reading section 4.1 entitled "Message Types" in rfc2068 (Jan '97)
shouldn't:

 generic-message = start-line
			*message-header
                     CRLF
                     [ message-body ]

be written as:

 generic-message = start-line
			1*message-header
			CRLF
                     [ message-body ]


The paragraph above this describes the "message-header" as one or more
header fields.  Thus, in keeping with the *rule of the BNF, shouldn't it
be "1*message-header" instead of "*message-header?"

I apologize if I am aproaching this issue in an unappropriate manner and
would appreciate it if you could communicate to me the right way in
handling issues of this same nature.

Dean Justus

------- End of Forwarded Message
Received on Sunday, 16 November 1997 17:11:47 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:33:03 EDT