RE: [xml-dev] hexBinary type

["Michael Kay" <mike@saxonica.com>]

It's unusual for Xerces to be wrong on something so basic, but I can't see any reason for rejecting this instance. The mapping from the lexical space to the value space is only informally described, but it's fairly clear to an intelligent reader that 0Fb8 and 0FB8 are different lexical representations of the same actual value, and this makes the instance valid. Certainly, that's the stance that Saxon takes.

 

Michael Kay

http://www.saxonica.com/

---

From: Eran Balter [mailto:E.Balter@F5.com]
Sent: 30 December 2007 18:12
To: xml-dev@lists.xml.org
Subject: [xml-dev] hexBinary type

Hi,

 

I am validating the following instance:

 

<hexBinary>0Fb8</hexBinary>

 

with the following schema:

 

<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">

        <xsd:element name="hexBinary">

                <xsd:simpleType>

                        <xsd:restriction base="xsd:hexBinary">

                                <xsd:enumeration value="0FB8"/>

                        </xsd:restriction>

         </xsd:simpleType>

        </xsd:element>

</xsd:schema>

 

The instance has the small “b” while the enumeration permits the capital “B”.

 

Is it a valid instance? (the standard claims: “The canonical representation for hexBinary is defined by prohibiting certain options from the Lexical Representation (§3.2.15.1). Specifically, the lower case hexadecimal digits ([a-f]) are not allowed.”). libxml2 claims it’s valid, while xerces claims it’s not.

 

TIA,

Eran