DOM&&SAX

先说一下DOM和SAX的区别

Structure

DOM: Tree-based

SAX: Event-Based

Processing

DOM: Batch-Reads entire doc first. Application gets control after parsing completes and a tree is built in memory

SAX: Streaming-Fires event callbacks during the doc read. Application maintains full control during parsing

Ease of use:

DOM: Complex and more difficult  to use, but more flexible (Spec: 500 pages)

SAX: Easy to use, but does not actually do much

CRUD

DOM: All

SAX: Read Only

Resources

DOM: Memory and processor heavy

SAX: Less memory, fast and efficient

Typical use

DOM: Non-sequential processing and updating of structure (or read doc multiple times)

SAX: Scanning large documents

关于DOM，参考了JR上的入门文章

http://www.javaresearch.org/article/showarticle.jsp?column=5&thread=37929

顺便说一下现在才知道Factory Pattern应用之广：DOM中有DocumentFactory，Hibernate中有SessionFactory，至于其他太细节的东西，用的时候再学也不迟

关于SAX

分为三个部分：Parser Creation, Content Handler, Error Handler

private void init() {
 
  try {
   SAXParserFactory factory = SAXParserFactory.newInstance();
   factory.setValidating(false);
   javax.xml.parsers.SAXParser saxParser = factory.newSAXParser();
   XMLReader xmlReader = saxParser.getXMLReader();
   SAXHandler xmlHndlr = new SAXHandler();
   xmlReader.setContentHandler(xmlHndlr);
   SAXErrorHandler errHndlr = new SAXErrorHandler(System.err);
   xmlReader.setErrorHandler(errHndlr);
   xmlReader.parse("books.xml");
  
  } catch (ParserConfigurationException e) { 
   e.printStackTrace();
  } catch (SAXException e) {
   e.printStackTrace();
  } catch(IOException e){
   e.printStackTrace();
  }
 }

SAXHandler继承自DefaultHandler，中间包括很多方法，需要自己定义，比如

startDocument()

endDocument()

startElement(String uri, String localName, String qualifiedName, Attributes atts)

endElement(String uri, String localName, String qualifiedName)

characters(char ch[], int start, int length)

ErrorHandler是一个interface，SAXErrorHandler需要实现几个方法，比如：

public void error(SAXParseException exception) throws SAXException {
  throw new SAXException("Error: " +getExceptionInfo(exception));

 }

 private String getExceptionInfo(SAXParseException exception) {
  return "URL= " +exception.getSystemId()
  + "Line= " +exception.getLineNumber()+": " +exception.getMessage();
  }

 public void fatalError(SAXParseException exception) throws SAXException {
  throw new SAXException("Fatal: " +getExceptionInfo(exception));
 }

 public void warning(SAXParseException exception) throws SAXException {
  out.println("Warning: " +exception.getMessage());

}

SAX不将整个文档放入内存，而是以基于事件的方式来处理文档，因此在速度和性能上优于DOM。但是在可读性上，SAX却不如DOM操作清楚简单。因此在文档不是特别大的时候，还是采用DOM方法比较合适