RE: No specification for querystrings and fragment identifiers

From: Dave J Woolley []
Sent: Tuesday, November 23, 1999 10:40 AM
> 	Some further information.
> 	Section 2.7 of the XML specification specifically
> 	forbids interpretation of character entitities in
> 	CDATA, quoting < and > as examples, but section
> 	2.4 makes specific exceptions for " and '
> 	in CDATA attributes.  Looks like there is no escape 
> 	convention to represent literal " in XML!  Oops!

Section 2.7 only applies to CDATA sections, of the form
<![CDATA[ data data...    ]]>

The attribute type CDATA, perversely, is something different.

Inside a CDATA section, nothing counts as markup.
In a CDATA attribute, there is markup.  The < character
is not allowed; character references and entity references
are processed.

Section 3.3.3 of the XML spec, "Attribute-Value Normalization",
specifies that "[b]efore the value of an attribute is passed to
the application or checked for validity, the XML processor
must normalize it...."  This "normalization" includes the usual
treatment of character references and entity references.  This
happens regardless of the declared "type" of the attribute.

The value &quot; could be represented by "&amp;quot;".


Received on Tuesday, 23 November 1999 12:28:26 UTC