W3C home > Mailing lists > Public > public-xsl-wg@w3.org > April 2016

Re: Test cases requiring DTD processing

From: Michael Kay <mike@saxonica.com>
Date: Wed, 27 Apr 2016 13:41:44 +0100
Cc: public-xsl-wg@w3.org
Message-Id: <04A38CAA-B057-47E1-85FA-AB2C5AF0CDE7@saxonica.com>
To: "TRON-DELTA.ORG" <info@tron-delta.org>
I'm not quite sure of the protocol here: you're on a list run by the W3C XSLT working group, but I don't think you are wanting a response from the group as such, and the discussion isn't really about the XSLT specifications; indeed much of it is about XML processing and not specifically XSLT at all. The xml-dev mailing list at www.xml.org/xml-dev might be a rather more suitable forum for these topics.

In answer to your first paragraph, XML Prague is indeed at xmlprague.cz, but the particular topic wasn't on the formal agenda and therefore is not covered in the formal proceedings. But there's a background article at http://dev.saxonica.com/blog/ that you might be interested in.

In response to your second paragraph, I was discussing the XML parsers supported by the browsers (for arbitrary XML vocabularies) and not their HTML or XHTML support.

Michael Kay
Saxonica


> On 27 Apr 2016, at 12:07, TRON-DELTA.ORG <info@tron-delta.org> wrote:
> 
> Hello,
> 
> I would like to hear more about that "XML Prague" convention. Is xmlprague.cz
> the official website? Is the "XSLT implementation", which was mentioned before,
> available as a document or something which can be reviewed?
> 
> Regarding the DTD processing issue I read an article from Mr. David Hammond,
> which tells that most browsers don't bother to implement real DTD-based parsing
> logic. In it I also read that page validation, according to XHTML DTD and XML's
> SGML-defined syntax *and* HTML DTD and `true` SGML-defined syntax, is different
> from each other. Also XML has additional well-formedness rules which is
> important with XHTML, of course.
> 
> I think important to note is that processing depends (1) on whether the XML or
> the SGML parser is utilized and (2) on the specific UA/browser implementation.
>> From what I understand, in terms of well-formedness and validity, UA/browsers
> silently deal with errors, and of course not every document processed is checked
> for well-formedness against the remote W3C DTD. Also no UA, that I am aware of
> at this time, does background validations (possible feature request?).
> 
> I hope everything I wrote here is correct; if not please correct me and let me
> know. Thanks in advance!
> 
> 
> Kind regards
> 
> Mathias Hollstein
> TRON-DELTA.ORG
> Nongovernmental Intelligence Organization
> --
> TRON-DELTA.ORG (NGO) | Jul. 15, 2015 - 17:00 PM | Sign. v5.3.10 in UTF-8
> I: #[T-D] (QuakeNet) | W: http://tron-delta.org | M: info@tron-delta.org
> 
> In case you are not the intended recipient of this message please delete
> this e-mail. Any unauthorized copying, disclosure or distribution of the
> material in it is strictly prohibited by international law.
> 
> Attachments with the extension .asc are PGP keys. See pgp.com, gnupg.org
> or pgpi.org for more information. The present PGP key ID is F7727720.
> 
> Please do not send us Microsoft Office/Apple iWork documents. Send Open-
> Office.org documents instead (PDF/A also accepted).
> 
> 
> 
Received on Wednesday, 27 April 2016 12:42:22 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 27 April 2016 12:42:22 UTC