SDO XML Data Access Service Functions

The following example shows how an XML document can be loaded from a file, altered, and written back.

An instance of the XML DAS is first obtained from the SDO_DAS_XML::create() method, which is a static method of the SDO_DAS_XML class. The location of the xsd is passed as a parameter. Once we have an instance of the XML DAS initialised with a given schema, we can use it to load the instance document using the loadFile() method. There is also a loadString() method if you want to load an XML instance document from a string. If the instance document loads successfully, you will be returned an object of type SDO_DAS_XML_Document, on which you can call the getRootDataObject() method to get the SDO data object which is the root of the SDO data graph. You can then use SDO operations to change the graph. In this example we alter the date, firstName, and lastName properties. Then we use the saveFile() method to write the changed document back to the file system. The saveFile method has an optional extra integer argument which if specified will cause the XML DAS to format the XML, using the integer as the amount to indent by at each change in level on the document.

This will write the following to letter-out.xml.

The previous example loaded the document from a file. This example shows how to create an SDO data graph in memory. In this example it is then saved to an XML string. Furthermore, because the letter contains both structured and unstructured content, it uses the Sequence API as well assignments to properties to construct the data graph.

This third example shows you how to set the XML version and encoding on the document object. These will be used when the XML is written out. If no XML declaration is wanted at all (perhaps you want to generate the XML as a string to embed in something) then you can use the setXMLDeclaration() method to suppress it.

The XML version and encoding are set in the XML declaration at the top of the XML document.

This fourth example illustrates the use of an SDO open type and the use of the createDataObject() method. For this example we use the following two schema:

Note the presence of the any element in the definition. This first schema defines the jungle complex type as containing a sequence of any other type. The other types that the example will use are defined in a second schema file:

Here is the example PHP code that uses these two schema files:

These two schema files are loaded into the XML DAS with first the create() and addTypes() methods. The createDataObject() method is used to create three separate data objects. In each case the namespaceURI and typename of the type are passed to the createDataObject() method: in this example the namespace URI is blank because no namespace is used in the schema. Once the three data objects - representing a bear, a panther and a snake - have been created, a document object is created with the createDocument() method. In this case there is no ambiguity about what the document element of the document should be - as the second schema file only defines complex types, the document element can only be the global jungle element defined in the first schema. This document will have a single root data object corresponding to an empty document element jungle. As this is an open type, properties can be added at will. When the first assignment is made to $do->bear, a property bear is added to the root data object: likewise for the next two assignments. When the document is written out by the saveString() method, the resulting document is:

This example is intended to illustrate how you can find the element name and namespace of the document element from the XML Document object, and the SDO type and namespace from the root data object of the XML data object, and how they relate to one another. This can be difficult to understand because there are four method calls: two can be made against the Document object, and two that can be made against any data object including the root data object. Because of the rules that define how the SDO model is derived from the XML model, when the data object concerned is the root object that represents the document object for the document, only three possible values can come back from these four method calls.

The two method calls that can be made against the document are getRootElementName() and getRootEelementURI(). These return the element name and namespace of the document element, respectively.

The two method calls that can be made against any data object are getTypeName() and getTypeNamespaceURI(). These return the SDO type name and type namespace of the data object, respectively.

Always, calling getRootElementURI() on the document object will return the same value as calling getNamespaceURI() on the root data object. Essentially, the information is all derived from the first few lines of the schema file, where there are three distinct pieces of information. For illustration, here again are the first few lines of the letter.xsd that we used above.

The three important values are:

It is part of the XML-SDO mapping rules that when the SDO model is built from the schema file, the typename and namespaceURI of the SDO types for the root element are taken from those of the complex type of the document element, where it exists. Hence in this example the typename of the root data object is FormLetter. In the event that there is no separate complex type definition for the document element, when the the type is defined inline and is anonymous, the SDO type name will be the same as the element name.

The following program loads the letter document and checks the return values from each of the four calls.

The output from this program is as follows:

The XML DAS provides a simple means to see what types and properties have been loaded. The php "print" or "echo" instruction will print out the types and properties.

The output from this program is as follows: