W3C home > Mailing lists > Public > xsl-editors@w3.org > January to March 2003

RE: Elements output as XML with HTML output method

From: Kay, Michael <Michael.Kay@softwareag.com>
Date: Fri, 3 Jan 2003 18:53:51 +0100
Message-ID: <DFF2AC9E3583D511A21F0008C7E621060453DF0D@daemsg02.software-ag.de>
To: zongaro@ca.ibm.com, xsl-editors@w3.org

These are excellent questions, and I don't have any ready answers!

I will raise an issue (issue 164) against the 2.0 spec, which means it will
come up for discussion in the WG, and when we have decided what to do about
it in 2.0, that will provide guidance if we feel the urge to issue a
clarification for 1.0.

For the time being, I would say that there are many aspects of serialization
where the spec leaves the details to the implementor, and this is one of
them.

Michael Kay


> -----Original Message-----
> From: zongaro@ca.ibm.com [mailto:zongaro@ca.ibm.com] 
> Sent: 03 January 2003 16:40
> To: xsl-editors@w3.org
> Subject: Elements output as XML with HTML output method
> 
> 
> 
> Hello,
> 
>      According to Section 16.2 of the XSL Transformations (XSLT) 
> Recommendation Version 1.0 [1]:
> 
> [[
> The html output method should not output an element 
> differently from the 
> xml output method unless the expanded-name of the element has a null 
> namespace URI; an element whose expanded-name has a non-null 
> namespace URI 
> should be output as XML.
> ]]
> 
>      In the preceding, does "element" refer to an element node in the 
> result tree or to an element in the resulting HTML?  In other 
> words, does 
> the requirement that an element with a non-null namespace URI 
> be output as 
> XML apply only to the element node or to its children and associated 
> attribute and namespace nodes as well?
> 
> 
>      If it refers to the element node only, that leads to a number of 
> questions about output processing:
> 
> o Does it mean only that the special rules described for HTML 
> 4.0 empty 
> elements and for script, style and head elements are to be 
> ignored?  Or 
> does it further mean that the entire start and end tag (or XML empty 
> element tag, if appropriate) should be emitted in an XML style, which 
> means that the less than character (<) in an attribute value and any 
> ampersand character (&) preceding a left brace character ({) 
> should be 
> emitted as an entity reference?  Or does it mean that only 
> some portion of 
> the start and end tags (or XML empty element tag) should be 
> emitted in an 
> XML style, and if so, which portion?
> 
> o Should default values of the xsl:output element that are 
> appropriate for 
> XML processing apply for such elements?  For example, would a default 
> version of 4.0 make sense for emitting XML?  Does the indent 
> attribute 
> have a default value of "no" for that part of the output 
> processing, or 
> does the default of "yes" that is applicable to the XML output method 
> apply to that portion of the output?
> 
> 
>      If it refers to an element in the resulting HTML, the 
> same questions 
> apply as above, and in addition the following questions:
> 
> o Does the cdata-section-elements attribute apply?  It's not 
> mentioned in 
> the description of the HTML output method in section 16.2, 
> and according 
> to Section 16 of XSLT 1.0 [2], "If the semantics of an 
> attribute are not described for an output method, 
> then it is not applicable to that output method."  Does the 
> fact that cdata-section-elements is not described for the HTML 
> output method take precedence over the fact that it is 
> described for the 
> XML output method in this context?
> 
> Thanks,
> 
> Henry
> [1] http://www.w3.org/TR/xslt#section-HTML-Output-Method
> [2] http://www.w3.org/TR/xslt#output
> ------------------------------------------------------------------
> Henry Zongaro      Xalan development
> IBM SWS Toronto Lab   Tie Line 969-6044;  Phone (905) 413-6044
> mailto:zongaro@ca.ibm.com
> 
Received on Friday, 3 January 2003 12:54:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:53 GMT