[tei-council] CDATA islands in <egXML>

Lou Burnard lou.burnard at retired.ox.ac.uk
Thu Dec 8 17:54:52 EST 2011


On 08/12/11 22:46, Martin Holmes wrote:
> I'm looking at removing as many CDATA islands as possible from<egXML>
> blocks. In the MS chapter, there are examples of intentionally invalid
> TEI XML presented like this:
>
> [quote]
>
> ...neither of the following is valid:
> <egXML xmlns="http://www.tei-c.org/ns/Examples"><![CDATA[
> <physDesc>
> <objectDesc form="codex">
> <!-- ... -->
> </objectDesc>
> <p>Generic descriptive prose...</p>
> </physDesc>]]></egXML>
> <egXML xmlns="http://www.tei-c.org/ns/Examples"><![CDATA[
> <physDesc>
> <objectDesc form="codex">
>     <!-- ... -->
> </objectDesc>
> <p>Generic descriptive prose...</p>
> <!-- other specific elements here -->
> </physDesc>]]></egXML>
>
> [/quote]
>
> I would have thought that instead of using CDATA to protect the code
> from validation tests, @valid="false" could be used on<egXML>. Am I wrong?
>

Nope. Tho what Stormy's stylesheets do with intentionally invalid 
examples remains to be seen.



More information about the tei-council mailing list