- From: John Cowan <cowan@mercury.ccil.org>
- Date: Tue, 11 Sep 2012 12:39:06 -0400
- To: Uche Ogbuji <uche@ogbuji.net>
- Cc: public-microxml@w3.org
Uche Ogbuji scripsit: > Well in that case no MUST in any RFC would be valid, since all machines are > subject to be inability to cary out a requirement due to lack of resources. Indeed. After all, we don't say that a C implementation isn't conformant because it doesn't handle procedures that recurse infinitely deep (or even trillions of times), even though nothing in the language specification disallows it. We don't even call XML parsers non-conformant because they fail to handle the billion-laughs attack in accordance with the spec. Consequently, I've removed the paragraph as unnecessary; whatever is universally true doesn't need to be in the spec. > "For any sequence of bytes, a conforming MicroXML parser MUST be able to > report correctly whether it is a conforming MicroXML document, provided it > has sufficient computing resources to do so. If it is a conforming MicroXML > document, then a conforming MicroXML parser MUST be able to report the > correct abstract data model for the document." I see no need for this. > > or just weaken the MUST to a SHOULD: > > This is certainly not an option. No, it isn't. -- Henry S. Thompson said, / "Syntactic, structural, John Cowan Value constraints we / Express on the fly." cowan@ccil.org Simon St. Laurent: "Your / Incomprehensible http://www.ccil.org/~cowan Abracadabralike / schemas must die!"
Received on Tuesday, 11 September 2012 16:39:32 UTC