- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Wed, 16 Sep 2009 15:08:46 +0200
- To: Mark Nottingham <mnot@mnot.net>
- CC: "Roy T. Fielding" <fielding@gbiv.com>, HTTP Working Group <ietf-http-wg@w3.org>
Mark Nottingham wrote:
> So, strict BNF with a note in prose that old implementations may quote
> other characters? Works for me.
> ...
Not convinced yet.
1) This seems to be inconsistent with how we treated other cases, such
as when we introduced BWS ("bad whitespace") as grammar production.
2) We currently use quoted-pair both for quoted strings and comments. So
this would need to change, and both would have to exclude different
characters.
3) In my experience as programmer, un-escaping usually treats all
escaped characters the same way.
4) Also, there's a chance that producers currently use a single
escaping/quoting implementation both for comments and quoted-string,
which would then escape '(', ')', '\' and '"'.
Thus, my proposal is to just disallow escapes, and warn about anything
else in prose.
BR, Julian
Received on Wednesday, 16 September 2009 13:11:25 UTC