W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1996

Re: Question on Chunk-Size sect 3.6, 8/12/96

From: John C. Mallery <jcma@ai.mit.edu>
Date: Fri, 30 Aug 1996 00:32:26 -0400
Message-Id: <v0300780dae4c1cf26e63@[128.52.39.15]>
To: "Roy T. Fielding" <fielding@liege.ICS.UCI.EDU>
Cc: http working group <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>
At 7:56 PM -0700 1996-08-29, Roy T. Fielding wrote:
>  Under the circumstances, the spec could say
>
> #1       chunk          = chunk-size [ chunk-ext ] *(SP | HT) CRLF
>                        chunk-data CRLF
>
>and explicitly disallow any other LWS between chunk-size and chunk-data CRLF
>(as it does for the other cases where CRLF is acting as a delimiter).
>

Your description would require the ";" to immediately follow chunk-size 
and padding to appear between  the ";" and CRLF.

The present known server and client implementations are handling this:

#2       chunk          = chunk-size *(SP | HT) [ chunk-ext ] CRLF
                        chunk-data CRLF

Although one reference client doesn't parse [ chunk-ext ], and so
might need a small hack to detect ";" and skip *(SP | HT) to the CRLF.

Another alternative could be:

#3       chunk          = chunk-size *(SP | HT) [ chunk-ext ] *(SP | HT) CRLF
                        chunk-data CRLF

This would seem most robust, if a little longer.
Received on Thursday, 29 August 1996 21:36:25 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:32:09 EDT