通过JAVA—API访问HDFS 上的文件

1.  通过对core-site.xml配置文件进行配置。

   配置项：hadoop.tmp.dir表示命名节点上存放元数据的目录位置，对于数据节点则为该节点上存放文件数据的目录。   

   配置项：fs.default.name表示命名的IP地址和端口号,缺省值是file:///，对于JavaAPI来讲，连接HDFS必须使用这里的配置的URL地址，对于数据节点来讲，数据节点通过该URL来访问命名节点。

2 利用java api来访问hdfs上的文件，并进行修改。

    * /**上传文件到HDFS上去*/

 private static void uploadToHdfs() throws FileNotFoundException,IOException {
  String localSrc = "d://123.txt";
  String dst = "hdfs://localhost:9000/";
  InputStream in = new BufferedInputStream(new FileInputStream(localSrc));
  Configuration conf = new Configuration();
  
  FileSystem fs = FileSystem.get(URI.create(dst), conf);
  OutputStream out = fs.create(new Path(dst), new Progressable() {
   public void progress() {
    System.out.print(".");
   }
  });
  IOUtils.copyBytes(in, out, 4096, true);
 }

  * /**从HDFS上读取文件*/

 private static void uploadToHdfs() throws FileNotFoundException,IOException {
 
  String localSrc = "d://123.txt";
  String dst = "hdfs://localhost:9000/";
 InputStream in = new BufferedInputStream(new FileInputStream(localSrc)); Configuration conf = new Configuration(); FileSystem fs = FileSystem.get(URI.create(dst), conf);
 OutputStream out = fs.create(new Path(dst), new Progressable() { public void progress() { System.out.print("."); } }); IOUtils.copyBytes(in, out, 4096, true); }

  * /**从HDFS上读取文件*/


private static void readFromHdfs() throws FileNotFoundException,IOException {
  String dst = "hdfs://localhost:9000/";
 Configuration conf = new Configuration(); FileSystem fs = FileSystem.get(URI.create(dst), conf); FSDataInputStream hdfsInStream = fs.open(new Path(dst)); OutputStream out = new FileOutputStream("d:/qq-hdfs.txt"); byte[] ioBuffer = new byte[1024];
 int readLen = hdfsInStream.read(ioBuffer); while(-1 != readLen){ out.write(ioBuffer, 0, readLen); readLen = hdfsInStream.read(ioBuffer); } out.close(); hdfsInStream.close(); fs.close(); }


 * /**以append方式将内容添加到HDFS上文件的末尾;注意：文件更新，需要在hdfs-site.xml中添<property><name>dfs.append.support</name><value>true</value></property>*/
private static void appendToHdfs() throws FileNotFoundException,IOException {
  
   String dst = "hdfs://localhost:9000/"; 
 Configuration conf = new Configuration(); FileSystem fs = FileSystem.get(URI.create(dst), conf); FSDataOutputStream out = fs.append(new Path(dst)); int readLen = "zhangzk add by hdfs java api".getBytes().length; while(-1 != readLen){ out.write("zhangzk add
 by hdfs java api".getBytes(), 0, readLen); } out.close(); fs.close(); }

 * /**从HDFS上删除文件*/
 private static void deleteFromHdfs() throws FileNotFoundException,IOException {
  
  String dst = "hdfs://localhost:9000/";
 Configuration conf = new Configuration(); FileSystem fs = FileSystem.get(URI.create(dst), conf); fs.deleteOnExit(new Path(dst)); fs.close(); }

 /**遍历HDFS上的文件和目录*/


 private static void getDirectoryFromHdfs() throws FileNotFoundException,IOException {
  String dst = "hdfs://localhost:9000/"; 
  Configuration conf = new Configuration();  
  FileSystem fs = FileSystem.get(URI.create(dst), conf);
  FileStatus fileList[] = fs.listStatus(new Path(dst));
  int size = fileList.length;
  for(int i = 0; i < size; i++){
  System.out.println("name:" + fileList[i].getPath().getName() + "/t/tsize:" + fileList[i].getLen());
  }
  fs.close();
 } 

}


 * 最后直接设置配置文件

Configuration conf = new Configuration();
		conf.addResource(new Path(
				"/project/hadoop-0.20.2/conf/hdfs-site.xml"));
		conf.addResource(new Path(
				"/project/hadoop-0.20.2/conf/core-site.xml"));
		conf.addResource(new Path(
				"/project/hadoop-0.20.2/conf/mapred-site.xml"));
		conf.addResource(new Path(
				"/project/hadoop-0.20.2/src/hdfs/hdfs-default.xml"));
		conf.set(UnixUserGroupInformation.UGI_PROPERTY_NAME, name);

参考文茂