- From: Grosso, Paul <pgrosso@ptc.com>
- Date: Sat, 18 Apr 2009 11:30:54 -0400
- To: <public-xml-core-wg@w3.org>
> -----Original Message----- > From: Simon Pieters [mailto:simonp@opera.com] > Sent: Saturday, 2009 April 18 5:49 > To: Grosso, Paul; public-xml-core-wg@w3.org > Subject: Re: xml-stylesheet issues--suggested resolutions > > On Fri, 17 Apr 2009 23:41:54 +0200, Grosso, Paul > <pgrosso@ptc.com> wrote: > > > Having reviewed what Arbortext does [1] and (thanks to Henry) > > what Saxon does [2], embedded herein are my suggestions for > > clarifications to the Associating Stylesheets Rec [3]. > > > > In general, we will need to decide if the various error > > cases are "it is an error (the xml-stylesheet processor > > may ignore the whole thing, may ignore what it doesn't > > understand and try to process the rest, etc.)" or "fatal > > xml-stylesheet error (a compliant xml-stylesheet processor > > must ignore the entire PI)" or "it is an error; the > > xml-stylesheet processor must recover by XXXX". > > > > In most cases, I'm tempted to say "is an error; the > > xml-stylesheet processor MAY ignore the entire PI; if > > it tries to recover, it SHOULD xxxx." Thoughts? > > I would prefer if for different errors it was either "is an > error: MUST ignore the entire PI" or "is an error: MUST > recover as follows: xxxx". We should certainly discuss this. My feeling is that too many MUSTs just means a lot of implementations will remain non-compliant. In particular, few implementations that currently accept a certain PI will want to change their behavior to ignore it--their uses would reasonably be upset that things that used to work suddenly stopped working. Remember that SHOULD implies that an implementation needs to have a "good reason" to do otherwise (where "good reason" is, of course, undefined), so while an implementation isn't non-compliant if it does otherwise, there is still some force of standardization to follow the recommendation. We can decide for each case what wording to use, but especially given the history/legacy involved in this situation as well as the fact the browsers are notoriously prone to be lenient in what they accept, I fear we'd just lose credibility if we said MUST in too many places. paul
Received on Saturday, 18 April 2009 15:33:11 UTC