19. Working with XML¶
XML is a powerful, flexible, open-standards-based method of data storage. Its format makes it human readable, while remaining easy to parse for programs. It is a simple hierarchical markup language. Tags are used to mark off sections of content and attributes are used to add metadata to the content.
Example:
<?xml version="1.0"?>
<books>
<book id="1">
<title>Les miserables</title>
<author>Victor Hugo</author>
</book>
</books>
One problem with semantic markup is the possibility for confusion. What is the meaning of title (book title or author title?). This is why namespaces have been introduced. A namespace provides a frame of reference for tags and has a unique ID (a URL) and a prefix to apply to tags from that namespace. In the following example, the URL is http://server.domain.tld/NameSpaces/Books and with a prefix of book: . The document would look like this:
<?xml version=”1.0”?>
<books:books owner='John Q. Reader' xmlns:lib=”http://server.domain.tld/NameSpaces/Library”>
<books:book id="1">
<books:title>Les miserables</title>
<books:author>Victor Hugo</author>
</books:book>
</books:books>
Todo
xslt, schema, xpath, ... dtd