All,
I have been looking even deeper into the unicode issues.
One aspect which matters for string operations is character
normalization. In unicode, for compatibility reasons, some characters
can be represented as either a single character, or a sequence of
characters which compose into a single character. Character
normalization means converting them into canonical form.
The friendly people at W3C have written the following document which is
intended to help, among others, writers of specifications:
http://www.w3.org/TR/2005/WD-charmod-norm-20051027/
I suggest that XACML 3.0 be made to conform to this.
Regards,
Erik