CSS2.1 i18n and bidi tests for review

Bert Bos wrote:

I don't know how well the converter deals with invalid XML, but as far
as I know it's not actually validating, so an attribute that is invalid
but well-formed probably survives.

Right.

the converter could remove the CDATA section, the same way it already
does for HTML. ( almost the same way: for HTML, it is enough to
remove the "<![CDATA[" and "]]>", for XHTML it would also have to
escape any <, and &.)

The converter parses the XHTML source file into a document tree
and prints output based on that. Whether the source file used
CDATA or escapes is inconsequential.

~fantasai