Spring底层使用SAX解析XML配置文件
最近在研究Spring源代码，在梳理spring-beans时故意只保留org.springframework.beans.factory.xml包下的xsd的3.1版本，然后在看Reference时，上面有一个简单的XML配置文件，如下
?
<?xml version="1.0" encoding="UTF-8"?><beans xmlns="http://www.springframework.org/schema/beans" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans-3.0.xsd"> <!-- beans --></beans>
?
?注意到这是一个3.0的xsd，然后测试时发现依然可以正常，只是变慢了，有时甚至卡住了，于是怀疑Spring底层代码访问了网址http://www.springframework.org/schema/beans/spring-beans-3.0.xsd，然后将返回的结果作为校验xml的xsd。
抽取了一部分ClassPathXmlApplicationContext加载过程的代码，仅仅解析XML配置文件，代码如下：
?
package test.temp;import java.io.InputStream;import javax.xml.parsers.DocumentBuilder;import javax.xml.parsers.DocumentBuilderFactory;import org.apache.commons.logging.Log;import org.apache.commons.logging.LogFactory;import org.springframework.beans.factory.xml.DefaultDocumentLoader;import org.springframework.beans.factory.xml.DelegatingEntityResolver;import org.springframework.beans.factory.xml.DocumentLoader;import org.springframework.core.io.Resource;import org.springframework.core.io.support.EncodedResource;import org.springframework.core.io.support.PathMatchingResourcePatternResolver;import org.springframework.core.io.support.ResourcePatternResolver;import org.springframework.util.xml.SimpleSaxErrorHandler;import org.w3c.dom.Document;import org.xml.sax.EntityResolver;import org.xml.sax.ErrorHandler;import org.xml.sax.InputSource;public class Test2 {private static Log logger = LogFactory.getLog(Test2.class);public static void main(String[] args) throws Exception {String location = "a/a1.xml";////DefaultListableBeanFactory beanFactory = new DefaultListableBeanFactory(null); //XmlBeanDefinitionReader beanDefinitionReader = new XmlBeanDefinitionReader(beanFactory);//int counter = beanDefinitionReader.loadBeanDefinitions(location, null);//ResourcePatternResolver resourceLoader = new PathMatchingResourcePatternResolver();Resource[] resources = ((ResourcePatternResolver) resourceLoader).getResources(location);for (Resource r : resources) {EncodedResource encodedResource = new EncodedResource(r);InputStream inputStream = encodedResource.getResource().getInputStream();InputSource inputSource = new InputSource(inputStream);// Resource resource = encodedResource.getResource();// // Key 2EntityResolver entityResolver = new DelegatingEntityResolver(null);ErrorHandler errorHandler = new SimpleSaxErrorHandler(logger);System.out.println("begin");DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();factory.setValidating(true); // Key 1factory.setNamespaceAware(true);factory.setAttribute("http://java.sun.com/xml/jaxp/properties/schemaLanguage", "http://www.w3.org/2001/XMLSchema");if (logger.isDebugEnabled()) {logger.debug("Using JAXP provider [" + factory.getClass().getName() + "]");}System.out.println("waiting...");DocumentBuilder docBuilder = factory.newDocumentBuilder();if (entityResolver != null) {docBuilder.setEntityResolver(entityResolver);}if (errorHandler != null) {docBuilder.setErrorHandler(errorHandler);}System.out.println("builder:[" + docBuilder + "]");// FIMXEDocument document = docBuilder.parse(inputSource);/*DocumentLoader documentLoader = new DefaultDocumentLoader();Document doc = documentLoader.loadDocument(inputSource, entityResolver, errorHandler, 3, false);*/System.out.println("end");}}}?
?然后就开始一段痛苦而又折磨人的DEBUG之旅，使用Eclipse，设置端点，Spring使用SAX解析XML，SAX底层代码很复杂，调用极多，所以我就在F5，F6之间迷失了...还好，最终定位到
?
package com.sun.org.apache.xerces.internal.impl;...public class XMLDocumentScannerImpl extends XMLDocumentFragmentScannerImpl{... public void setInputSource(XMLInputSource inputSource) throws IOException { fEntityManager.setEntityHandler(this); //this starts a new entity and sets the current entity to the document entity. fEntityManager.startDocumentEntity(inputSource); // fDocumentSystemId = fEntityManager.expandSystemId(inputSource.getSystemId()); setScannerState(XMLEvent.START_DOCUMENT); } // setInputSource(XMLInputSource)?和
package com.sun.org.apache.xerces.internal.impl ;...public class XMLEntityManager implements XMLComponent, XMLEntityResolver {...public String setupCurrentEntity(String name, XMLInputSource xmlInputSource, boolean literal, boolean isExternal) throws IOException, XNIException { // get information final String publicId = xmlInputSource.getPublicId(); String literalSystemId = xmlInputSource.getSystemId(); String baseSystemId = xmlInputSource.getBaseSystemId(); String encoding = xmlInputSource.getEncoding(); final boolean encodingExternallySpecified = (encoding != null); Boolean isBigEndian = null; // create reader InputStream stream = null; Reader reader = xmlInputSource.getCharacterStream(); // First chance checking strict URI String expandedSystemId = expandSystemId(literalSystemId, baseSystemId, fStrictURI); if (baseSystemId == null) { baseSystemId = expandedSystemId; } if (reader == null) { stream = xmlInputSource.getByteStream(); if (stream == null) { URL location = new URL(expandedSystemId); URLConnection connect = location.openConnection(); if (!(connect instanceof HttpURLConnection)) { stream = connect.getInputStream(); } else { boolean followRedirects = true; // setup URLConnection if we have an HTTPInputSource if (xmlInputSource instanceof HTTPInputSource) { final HttpURLConnection urlConnection = (HttpURLConnection) connect; final HTTPInputSource httpInputSource = (HTTPInputSource) xmlInputSource; // set request properties Iterator propIter = httpInputSource.getHTTPRequestProperties(); while (propIter.hasNext()) { Map.Entry entry = (Map.Entry) propIter.next(); urlConnection.setRequestProperty((String) entry.getKey(), (String) entry.getValue()); } // set preference for redirection followRedirects = httpInputSource.getFollowHTTPRedirects(); if (!followRedirects) { setInstanceFollowRedirects(urlConnection, followRedirects); } } stream = connect.getInputStream(); // REVISIT: If the URLConnection has external encoding // information, we should be reading it here. It's located // in the charset parameter of Content-Type. -- mrglavas if (followRedirects) { String redirect = connect.getURL().toString(); // E43: Check if the URL was redirected, and then // update literal and expanded system IDs if needed. if (!redirect.equals(expandedSystemId)) { literalSystemId = redirect; expandedSystemId = redirect; } } } }...? 总算是验证的我的猜想 :)
另外在还在频繁的debug中发现，拔网线会产生异常，但抛出的异常并不是想象中的IOException，而是经过包装后的，乍一看是不知道网络连接失败导致的。
再另，如果classpath下有3.0的xsd，Spring就会直接使用它。
?
?
?