Sorry people. Please forget I ever wrote this mail... Imagining a sequence of A's of length 3, followed by a length of sequence of length 4, we arrive at two sequences of A's with between 3 and 5 each, for a total indeed of 7 A's. Wasn't thinking straight this afternoon. Sorry again.
-----Original Message-----
From: James Bates
Sent: Fri 04/10/2002 17:29
To: [EMAIL PROTECTED]
Cc:
Subject: Schema validation
Considering this piece of schema:
<xs:complexType name="seqType">
<xs:sequence>
<xs:sequence minOccurs="2" maxOccurs="3">
<xs:element ref="A" minOccurs="3" maxOccurs="5" />
</xs:sequence>
</xs:sequence>
</xs:complexType>
I would expect an instance such as:
<root xsi:type="seqType">
<A/>
<A/>
....
</root>
to validate only if the number of A's can be expressed as a product of
2 or 3
with 3, 4 or 5, i.e. 6,8,9,10,12 or 15 A's. If there were 7,11,13 or
14 A's,
then the instance shouldn't validate. This is my understanding from
several
readings of paragraphs 3.8.4 and 3.9.4 part 3) of the XMLSchema
specification, part 1.
Using Xerces 2.2.0, it would however appear to validate any number of
A's
between 6 and 15. (MSXML4 demonstrates identical behaviour btw)
That's even with the
"http://apache.org/xml/features/validation/schema-full-checking"; feature
turned on. Is this a (known) bug? Or am I misreading the XML Schema's
spec?
Thanks,
James
As far as can understand paragraphs 3.8.4 and 3.9.4 part 3) of the
XMLSchema
specification, part 1, this is wrong, .
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<<winmail.dat>>
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
