'Effective Content' of a ComplexType in a Schema

[mahesh.kanakaraj]

Hi All,

I have a confusion in finding the 'effective content' of a complex
type
definition in a XML Schema.

I shall give you an example situation to clearly explain my problem.

Let's have an element declaration as follows:

<element name = "A" type="A_Type">

which is of type defined by the complexType definition as follows:

<complexType name="A_Type">
<complexContent >
<restriction base="anyType">
<choice>
<element name="One" type="string"/>
<element name="Two" type="string"/>
</choice>
</restriction>
</complexContent>
</complexType>

Now, based on the XML Schema specification [1], finding the effective
content for the complex type "A_Type" is what is really confusing me.

The specification says........
<quote>
[...]
2 [Definition:] Let the effective content be the appropriate case
among the following:
2.1 If one of the following is true
2.1.1 There is no <group>, <all>, <choiceor <sequence>
among the [children];
2.1.2 There is an <allor <sequenceamon g the [children]
with no [children] of its own excluding
<annotation>;
2.1.3 There is a <choiceamong the [children] with no
[children] of its own excluding <annotation>
whose
minOccurs [attribute] has the ·actual value· 0;
Then the appropriate case among the following:
2.1.4 If the ·effective mixed· is true, then A particle
whose properties are as follows:
{min occurs}
1
{max occurs}
1
{term}
A model group whose {compositor} is sequence
and
whose {particles} is empty.
2.1.5 otherwise empty
2.2 otherwise the particle corresponding to the <all>, <choice>,
<groupor <sequenceamon g the [children].
[...]
</quote>

Since for the complex type given above, the conditions 2.1 and 2.2
becomes false, and the condition 2.3 becomes true (since none of
the children has 'minoccurs' as '0'), so the calculation of the
effective content is done using 2.1.4 or 2.1.5 based
on the 'effective mixed' value.

For the complex type declared above 'effective mixed' is false, so
2.1.5 is chosen and the 'effective content' becomes empty.

And as per the specification (quote given below), calculating the
'conent type' of this complex type node......
<quote>
[....]
3 Then the value of the property is the appropriate case
among the following:
3.1 If the <restrictionalt ernative is chosen,
then the appropriate case among the following:
3.1.1 If the ·effective content· is empty ,
then empty;
3.1.2 otherwise a pair consisting of
3.1.2.1 mixed - if the ·effective mixed· is true,
otherwise elementOnly
3.1.2.2 The ·effective content·.
[....]
</quote>

Since the 'effective content' identified is empty, using 3.1.1,
the 'content type' of this complex type becomes empty.

My question comes here..........

Shouldn't the 'content type' of this complex type node be
a 'Choice'?

Is my interpretation of the specification
correct? Or where am I going wrong?

Any kind of help would be appreciated.
Thanks in Advance.

Mahesh K U

[1] - "http://www.w3.org/TR/xmlschema-1/#key-efm"

[usenet]

On 19 Jun, 08:34, mahesh.kanaka.. [e-mail] wrote:
....

<complexType name="A_Type">
<complexContent >
<restriction base="anyType">
<choice>
<element name="One" type="string"/>
<element name="Two" type="string"/>
</choice>
</restriction>
</complexContent>
</complexType>

....

2.1.3 There is a <choiceamong the [children] with no
[children] of its own excluding <annotation>
whose
minOccurs [attribute] has the ·actual value· 0;
Then the appropriate case among the following:
2.1.4 If the ·effective mixed· is true, then A particle

....

2.1.5 otherwise empty
2.2 otherwise the particle corresponding to the <all>, <choice>,
<groupor <sequenceamon g the [children].
[...]
</quote>

XML schema is an appallingly written spec. If I were you I would try
and make do with the primer, or a book such as Eric van der Vlist's.

In the case you mention, I think it's ambiguous in 2.1.3 as to what
the "whose" refers to. (In fact the spec if rife with these
ambiguities, often forcing you to work out what "it's" might refer to
from a number of possibilities!)

Clearly the construct you refer to is not empty. Therefore, the only
way to interpret 2.1.3 is to infer that the "whose" is referring to
the choice element. e.g. it would be better written as:

2.1.3 There is a <choiceamong the [children] whose minOccurs
[attribute] has the ·actual value· 0 and has no [children] of its own
excluding <annotation>.

Then the case you specify invokes 2.1.5 and 3.1.2.2.

The implication is that:

<choice minOccurs="0"></choiceis legal, but
<choice></choiceis illegal.

HTH,

Pete.
--
=============== =============== ===============
Pete Cordell
Codalogic Ltd
for XML Schema to C++ data binding visit
http://www.codalogic.com/lmx/
=============== =============== ===============