10.1 XML, DTDs, and XML Schemas

For an XML document to be used for information exchange, it must be parseable, and to be parseable, it must be well- formed . A well-formed document, among other rules, must have an end tag for every start tag, and different start and end tags may not be placed in such a matter as to cross each other's boundaries.

For instance, Example 10-1 is a well-formed XML document that contains information about a person named Jane Doe, born on January 1, 1980.

Example 10-1. An example of a well-formed XML document

 <?xml version="1.0"?> <person>   <name>     <last_name>Doe</last_name>     <first_name>Jane</first_name>   </name>   <birth_date>1980-01-01</birth_date>   <gender>Female</gender> </person> 

If a document is to be shared with more than one computer program, or shared between organizations, it should also be valid . A valid XML document is one that conforms to an agreed-upon format. An XML document is valid if it can be validated against a Document Type Definition (DTD) or an XML Schema.

DTD is the first schema standard developed for formally describing an XML document's format. A DTD is narrative document-centric, so while it provides a means to describe a document's structure, it doesn't include the ability to specify datatypes.

To continue with the Jane Doe example, Example 10-2 shows a DTD that can validate the XML document in Example 10-1.

Example 10-2. An example of a Document Type Definition

 <!ELEMENT person (name, birth_date, gender)> <!ELEMENT name (last_name, first_name)> <!ELEMENT last_name (#PCDATA)> <!ELEMENT first_name (#PCDATA)> <!ELEMENT birth_date (#PCDATA)> <!ELEMENT gender (#PCDATA)> 

XML Schema, a more recent standard, is data document-centric. Not only does it define document structure, it also supports user -defined datatypes and some object-oriented principles, such as inheritance. By validating an XML document against its XML Schema, you can validate its structure as well as the datatypes of its content. This is ideal for computer-to-computer data exchange.

Example 10-3 is an example of a possible XML Schema for the XML document in Example 10-1. If you compare this XML Schema with the DTD in Example 10-2, you'll see the additional specification of the datatypes for each element. Not only do XML Schemas add datatypes, but they also add complex constraints and namespace support, concepts that we will address in the sections that follow.

Example 10-3. An example of an XML Schema

 <?xml version="1.0"?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">   <xsd:element name="person">     <xsd:complexType>       <xsd:sequence>         <xsd:element name="name">           <xsd:complexType>             <xsd:sequence>               <xsd:element name="last_name"                  type="xsd:string" minOccurs="1"/>               <xsd:element name="first_name"                  type="xsd:string" minOccurs="1"/>             </xsd:sequence>           </xsd:complexType>         </xsd:element>         <xsd:element name="birth_date"            type="xsd:date"/>         <xsd:element name="gender"                type="xsd:string" minOccurs="1"/>       </xsd:sequence>     </xsd:complexType>   </xsd:element> </xsd:schema> 

A DTD, an XML Schema, or both can be used to define an XML document's vocabulary. This vocabulary, which is a formal specification, is called an XML application . This type of application is very different from our normal notion of an application, such as a set of programs coded in a programming language such as Java or PL/SQL. This distinction also brings to light the fact that XML itself isn't a programming language, a protocol, or a database management system.

Because of its capabilities and its promise as a universal data format, XML has suffered more than its fair share of hype. While XML is ideal for storing information in a universal format, making it a perfect choice for information exchange, it makes a poor database. Yes, it can be used as a database to store relatively small amounts of information in an organized manner, but beyond that, information stored as XML is slow to retrieve and/or manipulate when compared to information stored in an object-relational database such as Oracle.