遍历ZIP（JAR）文件,动态加载,找出其中指定类型的类

想要动态加载指定文件夹下所有的JDBC驱动，用于建立DB连接。

有三个任务：

1.遍历文件夹,找出所有JAR文件:这个简单,先不管

2.遍历JAR文件中所有的项,找出CLASS文件

3.判断CLASS文件是否是Driver的子类

只剩2,3了.一步一步做:

遍历JAR文件中所有的项,找出CLASS文件

遍历是一回事,找出CLASS文件是另一件事,我还有可能遍历是为了解压缩,于是抽象遍历功能:

还有可能ZIP文件中还有ZIP.我要递归,于是加一个isRecursion判断一下

/** * walk throught a zip(jar)file * * @param file * zip(jar) file * @param processor * how to process every ZipEntry * @param isRecursion * is do recursion * @throws ZipException * throw not zip exception * @throws IOException * throw io exception */ public static final void process(File file, IZipEntryProcessor processor, boolean isRecursion) throws ZipException, IOException { if (file == null || !file.exists() || file.isDirectory()) { return; } IZipEntryProcessor localProcessor = processor; if (localProcessor == null) { localProcessor = new SimpleZipEntryProcessor(); } ZipFile zipFile = new ZipFile(file); Enumeration<? extends ZipEntry> zipEntrys = zipFile.entries(); while (zipEntrys.hasMoreElements()) { ZipEntry zipEntry = zipEntrys.nextElement(); localProcessor.processEntry(zipFile, zipEntry); if (isRecursion) { processEntry(zipFile, zipEntry, localProcessor); } } }

其中的IZipEntryProcessor就可以用来做不同的事情:

public interface IZipEntryProcessor { void processEntry(ZipFile file, ZipEntry entry); }

 

下一步:动态加载,找出指定的类,本着能抄就抄一点的态度,有了以下代码:(好多年前的代码了,来源:http://topic.csdn.net/t/20050601/15/4051576.html)

class FindMainClassProcessor implements IZipEntryProcessor { /** * change protected addURL to public ... */ MyClassLoader classLoader = new MyClassLoader(new URL[] {}, FindMainClassProcessor.class .getClassLoader()); @Override public void processEntry(ZipFile file, ZipEntry entry) { // check all class file if (entry.getName().endsWith(".class")) { try { // load jar to classpath classLoader.addURL(new File(file.getName()).toURI().toURL()); // get class full name(with package) String className = entry.getName().replace("/", "."); className = className.substring(0, className.length() - 6); // use reflect to get instance Class<?> mainClass = Class.forName(className, true, classLoader); // find main method Method mainMethod = mainClass.getMethod("main", new Class[] {String[].class }); // check is REAL main method if ((mainMethod.getModifiers() & (Modifier.STATIC | Modifier.PUBLIC)) == (Modifier.STATIC | Modifier.PUBLIC)) { // do something System.out.println(file.getName() + ":" + className); // call main method // mainMethod.invoke(null, new Object[] {new String[] {} }); } } catch (Throwable e) { } } } class MyClassLoader extends URLClassLoader { public MyClassLoader(URL[] urls, ClassLoader parent) { super(urls, parent); } @Override public void addURL(URL url) { super.addURL(url); } } }

 

最后,测试一下:

public static void main(String[] args) { JFileChooser chooser = new JFileChooser(); chooser.showOpenDialog(null); try { process(chooser.getSelectedFile(), new FindMainClassProcessor(), false); } catch (Exception e) { e.printStackTrace(); } }

 

正确运行,OK

源代码差不多就这些了,不上传了