org.apache.commons.io.DirectoryWalker如何递归遍历

Question

类

 public class DirectoryTracker extends DirectoryWalker<String> {
    private int bufferSize=100;
    private final List<String> workedBuffer=Collections.synchronizedList(new ArrayList<String>());
    private static final Logger log = LoggerFactory.getLogger(DirectoryTracker.class);

    public DirectoryTracker(FileFilter filter) {
        super(filter,-1);
        // TODO Auto-generated constructor stub
    }
    /**
     * 设置一次处理的结果集大小
     * @param size
     */
    public void setBufferSize(int size){
        this.bufferSize=size;
    }

    @Override
    protected boolean handleDirectory(File directory, int depth, Collection<String> results) throws IOException {
        // TODO Auto-generated method stub
        //这理怎么写
        return true;
    }

    @Override
    protected void handleFile(File file, int depth, Collection<String> results) throws IOException {
        // TODO Auto-generated method stub
        //super.handleFile(file, depth, results);
        results.add(file.getAbsolutePath());
    }

    private List<String> getAllFile(File startDirectory) throws IOException{
        List<String> result=new ArrayList<>(bufferSize);
        walk(startDirectory,result);
        return result;
    }
    //ETC
}

指定一个开始的目录(例:D:\Documents\Pictures),这个目录下有图片也有子目录,子目录的情况如父目录一样,现在这个类不进入子目录,怎样作到无限递归遍历

Answer 1

// TODO Auto-generated method stub
//这理怎么写
if(depth==-1)return;
File[] files=directory.listFiles();
for(File file:files){
if(file.isDirectory()){
handleDirectory(file,depth-1,results);
}else if(filter.accept(file)){
handleFile(file,depth-1,results);
}
}

            手写的，没有编译调试过，大概是这个意思吧