- From: Bryan Rasmussen <brs@itst.dk>
- Date: Mon, 4 Apr 2005 12:56:28 +0200
- To: "'xmlschema-dev@w3.org'" <xmlschema-dev@w3.org>
I guess my earlier email must not have been clear, or at least I hope it was lack of clarity and not just a question so stupid as to be unanswerable without emotional pain, so here is a second attempt: given the following schema: <?xml version="1.0" encoding="UTF-8"?> <schema targetNamespace="http://rep.oio.dk/xkom.dk/xml/schemas/2004/02/02/" xmlns:xkom="http://rep.oio.dk/xkom.dk/xml/schemas/2004/02/02/" xmlns="http://www.w3.org/2001/XMLSchema" elementFormDefault="qualified" attributeFormDefault="unqualified"> <element name="EmailAddressIdentifier" type="xkom:EmailAddressIdentifierType"> <annotation> <documentation source="http://rep.oio.dk/xkom.dk/xml/schemas/2004/02/02/xkom_emailaddressid entifier.xsd.meta.xml"/> </annotation> </element> <simpleType name="EmailAddressIdentifierType"> <restriction base="token"> <pattern value="([^>\(\)\[\]\\,;:@\s]{1,90}@[^>@]{1,64})"/> </restriction> </simpleType> </schema> and the following instance <doc xmlns:xkom="http://rep.oio.dk/xkom.dk/xml/schemas/2004/02/02/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://rep.oio.dk/xkom.dk/xml/schemas/2004/02/02/ email.xsd" > <xkom:EmailAddressIdentifier>bry@itnisk.com</xkom:EmailAddressIdentifier> <xkom:EmailAddressIdentifier>bry@itnisk.com</xkom:EmailAddressIdentifier> <xkom:EmailAddressIdentifier>bry@itnisk.com</xkom:EmailAddressIdentifier> <xkom:EmailAddressIdentifier>bry@itnisk.com</xkom:EmailAddressIdentifier> </doc> is that instance valid? I thought it should be, according to XSV and msxml's SchemaCache it is okay, XSV in this case defaults to a lax processing, a coworker says he remembers this being optional according to the spec. Can anybody reference me to the part where this is defined as optional, or if not where it is explicitly allowed or disallowed? -----Oprindelig meddelelse----- Fra: Bryan Rasmussen Sendt: 31. marts 2005 16:00 Til: 'xmlschema-dev@w3.org' Emne: rules for handling document elements not in the target namespace? Hi, I have a schema in namespace X that validates one global element, in order to test that element I went ahead and put various versions of the element in a document with a document element in a different namespace, most processors reacted fine with that and told me that my proper examples were correct and my bad examples were incorrect. XMLSPY however got upset that the document element was not validated in my schema. as a second test I then created an instance consisting only of a document element not in the target namespace of the schema which XMLSPY also reported as being invalid because of the document element not being validated by the schema. A co-worker said that this was because doing so was optional by the spec, by my memory this was not one of the optional sections. I've looked but of course the spec is rather large, can someone point me to the optional part of the spec on this matter or give me some backup on its not being optional if indeed it isn't.
Received on Monday, 4 April 2005 11:03:43 UTC