- From: James Clark <jjc@jclark.com>
- Date: Thu, 6 Sep 2012 08:49:28 +0700
- To: John Cowan <cowan@mercury.ccil.org>
- Cc: Liam R E Quin <liam@w3.org>, public-microxml@w3.org
Received on Thursday, 6 September 2012 01:50:16 UTC
On Wed, Sep 5, 2012 at 4:22 AM, John Cowan <cowan@mercury.ccil.org> wrote: > Liam R E Quin scripsit: > > > XML says, "For compatibility, the string " -- " (double-hyphen) must not > > occur within comments." > > > > 11a. Should we allow <!-- -- --> as a 琪ML comment? > > > > The XML spec does not say that compatibility constraints are not > > binding, so I believe the above not to be well-formed XML, and the > > answer should be that it is to be forbidden. > > Indeed, it says MUST NOT, so it is an error, though not a fatal one. > I don't know of any XML processors that treat non-fatal errors > differently from fatal ones. > "--" in comments are also disallowed by the XML 1.0 grammar Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' So it's a well-formedness error. > However, this raises the question of whether we should impose only > the looser XML specification of a comment or the tighter HTML5 > specification. This goes back to the issue of what we are trying to achieve as regards HTML5 friendliness, which also affects the issue of whether to allow a bare DOCTYPE. James
Received on Thursday, 6 September 2012 01:50:16 UTC