Hello,
I am getting a transformation error stating that my file "contains 4 characters that should not be present in a file using the Codepage 1252 (Western) encoding". It claims the offending characters are "0x9D", but I am dubious of this character information as I do not find that string in the xml file.

In any case, it allows me to continue the translation, but it (XMLSpy ver 5 rel 2) insists on applying "substitution characters" where ever it finds the so called dis-allowed characters.

When it generates the output with the substituted characters, it clearly was hiccup-ing anywhere there was a "<" as the start of the html tag. I can tell this is the case because I am seeing the tail-end of every corresponding html tag (output as plain text since the browser never finds the "<" tag. As an example, I see td>, /td>, etc. all over the place.

Incidently, I am puzzled that when I perform the exact same transformation against the exact same XML/XSL files using a newer version of XMLSpy (2007 SP2), I never had this problem. If you are wondering why I don't simply stay with the newer version, its because I have the older version imposed upon me by the customer.

As a little bit of extra information, the input XML file presents "<" as "&lt;". For this reason, I had always been using d-o-e when outputting text. Again, always worked fine with XMLSpy 2007. Any help is greatly appreciated. Thanks

Can't help you at all from the information given. If it works with a newer release of Altova then it sounds as if it's a bug in the older release.

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

The offending sequence of characters that resulted in the character replacement was several places where double-quotes were nested within other double quotes. For instance, while performing transformation against input XML file that had:

<publish:property name="Body" publish:category="UML" publish:description="The Body of the Comment" publish:value=" user that, pursuant to 19, “is one who undertakes to eat fish[across the park].” Each use is assigned a code."/>

The xsl code where the replacement was forced to take place by XMLSpy (ver 5 rel 2) was:
[list]
                        <li>
                            <xsl:value-of select="ownedComment/@body" disable-output-escaping="yes"/>
                        </li>
                    </ul>

Again, I did not have similar issues with XMLSpy 2007 SP2. I sure wish I knew what to do here, short of stripping all input files of nested quotes...

>while performing transformation against input XML file that had:

<publish:property name="Body" publish:category="UML" publish:description="The Body of the Comment" publish:value=" user that, pursuant to 19, "is one who undertakes to eat fish[across the park]." Each use is assigned a code."/>

That's not well-formed XML, so no XSLT processor should ever accept it. You need to fix the process that's generating the data incorrectly.



Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference

My editor (XMLSpy) is telling me that the xml file is well-formed. Which part of the content is wrong exactly?

If an attribute is delimited by double quotes then it can't contain double-quotes, they should be escaped as &_quot; (ignore the underscore)

Michael Kay
http://www.saxonica.com/
Author, XSLT Programmer's Reference and XPath 2.0 Programmer's Reference