A Data Stream View of Scientific Workflow - Grid Computing: Infrastructure, Service, and Applications - page 228

DTD; the content includes functionality, namespace sensitivity support,
and data exchange applications. Finally, the syntax of XML schema is
self-description, which stands for using an XML document to describe
the XML schema.
The XML schema is typically graph like, has a tree-like structure, and
can be described by automata, actually by a nondeterministic i nite autom-
aton (NFA) [22].

<schema...>
<!--ANY NUMBER OF FOLLOWING-->
<include…/>
<import>…</import>
<redefine>…</redefine>
<annotation>…</annotation>
<!--ANY NUMBER OF FOLLOWING DEFINITION-->
<simpleType>…</simpleType>
<complexType>…</complexType>
<element>…</element>
<attribute/>
<attributeGroup>…</attributeGroup>
<group>…</group>
<annotation>…</annotation>
</schema>

Above is an example of an XML schema for an XML document structure
and element content modeling. For element dei nitions, XML schema
“Element” is used to dei ne an element. The capability of “Element” can
offer more constraints than the XML DTD can do. The specii c power
of element dei nitions include simple content, complex content, empty
elements, child element content, namespace, occurrence options, choices,
group occurrence options, embedded groups, mixed content, and so on.
For a detailed description of XML schema application, see [ 23 ] .
There are also patterns in an XML schema. Due to the resemblance of
the workl ow schema and XML schema, the overlapping patterns
from both schemas can be found. This overlapping means that we can
reuse the pattern techniques from XML data processing, especially
XML data streams under the scientii c workl ow scheme, for data l ow
processing.

9.2.3.1 Pattern Analysis of the XML Schema

According to the work of van der Aalst, there are 20 workl ow patterns
[14]. In the XML schema, there are lots of pattern constraints and even
quantity and data range constraints. The following examples demonstrate
similarity, but are not exhaustive.