- From: DuCharme, Robert <Robert.DuCharme@moodys.com>
- Date: Fri, 23 Jun 2000 10:16:05 -0400
- To: "'www-xml-schema-comments@w3.org'" <www-xml-schema-comments@w3.org>
I was thinking about how XML schemas allow the specification of types for both elements and attributes, and how the old attribute types are retained for backward compatibility, and I realized that this allowed for the possibility of NMTOKEN elements (not to mention NMTOKENS elements). The bottom of the first table in 2.3 of the schema primer (http://www.w3.org/TR/xmlschema-0/#CreatDt) tells us that "to retain compatibility between XML Schema and XML 1.0 DTDs, the simple types ID, IDREF, IDREFS, ENTITY, ENTITIES, NOTATION, NMTOKEN, NMTOKENS should only be used in attributes." The normative spec doesn't mention this; in fact, an example there actually uses it: in section 4.3.3 of Part 1 of the spec, the definition of a complexType called "length2" includes a subelement ("unit") declared as being of type NMTOKEN. Section 3.3.3 of the XML spec (http://www.w3.org/TR/REC-xml#AVNormalize) describes specific responsibilities of an XML parser for, among other things, normalizing the space included with a non-CDATA attribute value. Without this normalization, there's not much point in using NMTOKEN. Have a parser's responsibilities regarding NMTOKEN element types been addressed anywhere? Bob DuCharme www.snee.com/bob <bob@ snee.com> "The elements be kind to thee, and make thy spirits all of comfort!" Anthony and Cleopatra, III ii
Received on Friday, 23 June 2000 10:15:42 UTC