All characters above U+007F are treated as parts of names, even though there are several reserved characters such as U+2028 and U+2029 which are logically whitespace.
Therefore, all namespace, class and property names are identifiers composed as follows:
Note that the Unicode specials range (U+FFF0...U+FFFF) are not legal for identifiers.
While the above sub-range of U+0080 . . . U+FFEF includes many diacritical characters which would not be useful in an identifier, as well as the Unicode reserved sub-range which has not been allocated, it seems advisable for simplicity of parsers to simply treat this entire sub-range as "legal" for identifiers.
Refer to RFC2279 (see ReferencedDocuments) as an example of a Universal Transformation Format that has specific characteristics for dealing with multi-octet characters on an application-specific basis.
U+FFFE is little endian.
All MOF keywords and punctuation symbols are as described in the MOF Syntax document and are not locale-specific. All such characters are composed of characters falling in the range U+0000...U+007F, regardless of the locale of origin for the MOF or its identifiers.
The supported delimiters are U+0027 or U+0022. Once a quoted string is started using one of these delimiters, the same delimiter is used to terminate it.
In addition, the digraph U+005C ("\") followed by U+0027 """ constitutes an embedded quotation mark, not a termination of the quoted string.
The characters permitted within the quotation mark delimiters just described may fall within the range U+0001 through U+FFEF.