[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: An element's value is an invalid Unicode string ... how ca

• From: "G. Ken Holman" <gkholman@CraneSoftwrights.com>
• To: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
• Date: Fri, 04 Jan 2013 13:43:54 -0500

At 2013-01-04 18:26 +0000, Costello, Roger L. wrote:
>Consider this Spanish name: Martiñez
>
>Instead of using the ñ character, one can use 
>the (base) "n" character followed by a combining tilde (hex 303) character.

Perhaps one can do that for display purposes, but 
I would think linguists would challenge you.  The 
letter in the Spanish alphabet is a fully-formed 
letter on its own, it is not an accented "n", so 
if you compose it as an accented "n" then you 
aren't representing the Spanish letter.  There 
are 27 (or 29) letters in the Spanish alphabet:

   http://spanish.about.com/cs/forbeginners/a/beg_alphabet.htm

>So that Spanish name can be equivalently expressed as: Martiñez

I disagree, but let's continue with your discussion.

>Here is an XML document that uses the latter form:
>
><?xml version="1.0" encoding="utf-8"?>
><Name>Martin&#x303;ez</Name>
>
>I wrote a stylesheet that uses the substring() 
>function to extract the combining tilde character and onward:
>
>     <xsl:template match="/">
>             <Result>
>                     <xsl:value-of select="substring(Name, 7)" />
>             </Result>
>     </xsl:template>
>
>The output is:
>
><?xml version="1.0" encoding="UTF-8"?>
><Result>Þez</Result>
>
>I checked it for well-formedness and the XML Parser says it is well-formed.

Correct ... because there are no restrictions on 
the Unicode *characters* that can be included in 
an element that state an element cannot begin with a diacritic.

>According to the book, Fonts & Encodings (p. 61, first paragraph):
>
>     ... we select a substring that begins
>     with a combining character, this new
>     string will not be a valid string in
>      Unicode.
>
>The value of the <Result> element is not a valid 
>Unicode string, so how can it be a well-formed XML document?

Because well-formedness is not defined by strings, but by characters:

    Ref: [39] http://www.w3.org/TR/2008/REC-xml-20081126/#NT-element
    [39]    element    ::=    EmptyElemTag
                              | STag content ETag

    Ref: [43] http://www.w3.org/TR/2008/REC-xml-20081126/#NT-content
    [43]    content    ::=    CharData? ((element | Reference | CDSect |
                              PI | Comment) CharData?)*

    Ref: [14] http://www.w3.org/TR/2008/REC-xml-20081126/#NT-CharData
    [14]    CharData    ::=    [^<&]* - ([^<&]* ']]>' [^<&]*)

I see nothing there that states a diacritic 
cannot be the first character of character data.

Well-formedness is defined from the perspective 
of reading characters.  Well-formedness does not 
involve interpreting the content of an element represented by those characters.

It would be up to an application to complain that 
the content is not as expected, not the XML processor.

What if I wanted to have the following:

   <allowed-diacritics>
     <diacritic>&#303;</diacritic>
   </allowed-diacritics>

Why would an XML processor be obligated to tell 
me that that is not well-formed?

I hope this is helpful.

. . . . . . . . . Ken

--
Contact us for world-wide XML consulting and instructor-led training
Free 5-hour lecture: http://www.CraneSoftwrights.com/links/udemy.htm
Crane Softwrights Ltd.            http://www.CraneSoftwrights.com/x/
G. Ken Holman                   mailto:gkholman@CraneSoftwrights.com
Google+ profile: https://plus.google.com/116832879756988317389/about
Legal business disclaimers:    http://www.CraneSoftwrights.com/legal

• References:
  • An element's value is an invalid Unicode string ... how can it bewell-formed?
    • From: "Costello, Roger L." <costello@mitre.org>