當前位置：首頁 > 编程资源 > 编程问答 >内容正文

编程问答

hdfs user 连接_通过API访问HDFS

發(fā)布時間：2025/3/20 编程问答 41 豆豆

生活随笔收集整理的這篇文章主要介紹了 hdfs user 连接_通过API访问HDFS 小編覺得挺不錯的,現(xiàn)在分享給大家,幫大家做個參考.

通過API操作HDFS

今天的主要內(nèi)容

HDFS獲取文件系統(tǒng)

HDFS文件上傳

HDFS文件下載

HDFS目錄創(chuàng)建

HDFS文件夾刪除

HDFS文件名更改

HDFS文件詳情查看

定位文件讀取

FileSystem類的學(xué)習(xí)

1. HDFS獲取文件系統(tǒng)

//獲取文件系統(tǒng)

@Test

public void initHDFS() throws Exception{

//1. 獲取文件系統(tǒng)

Configuration configuration = new Configuration();

FileSystem fileSystem = FileSystem.get(configuration);

//2. 打印文件系統(tǒng)到控制臺

System.out.println(fileSystem.toString());

}

2. HDFS文件上傳(測試參數(shù)優(yōu)先級)

@Test

public void putFileToHdfs() throws Exception{

Configuration conf = new Configuration();

conf.set("dfs.replication", "2"); //代碼優(yōu)先級是最高的

conf.set("fs.defaultFS", "hdfs://10.9.190.111:9000");

FileSystem fileSystem = FileSystem.get(conf);

//上傳文件

fileSystem.copyFromLocalFile(new Path("hdfs.txt"), new Path("/user/anna/hdfs/test.txt"));

//關(guān)閉資源

fileSystem.close();

}

參數(shù)優(yōu)先級：(1)客戶端代碼中設(shè)置的值 >(2)classpath 下的用戶自定義配置文件 > (3)然后是服務(wù)器的默認配置

3. HDFS文件下載

public void copyToLocalFile(boolean delSrc,Path src,Path dst,boolean useRawLocalFileSystem)

throws IOException

delSrc - whether to delete the src

src - path

dst - path

useRawLocalFileSystem - whether to use RawLocalFileSystem as local file system or not.

@Test

public void testCopyToLocalFile() throws Exception{

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.111:9000");

FileSystem fileSystem = FileSystem.get(conf);

///下載文件

fileSystem.copyToLocalFile(false,new Path("/user/anna/hdfs/test.txt"), new Path("test.txt"),true);

//關(guān)閉資源

fileSystem.close();

}

4. HDFS目錄創(chuàng)建

@Test

public void testMakedir() throws Exception{

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.111:9000");

FileSystem fileSystem = FileSystem.get(conf);

//目錄創(chuàng)建

fileSystem.mkdirs(new Path("/user/anna/test/hahaha"));

//關(guān)閉資源

fileSystem.close();

}

5. HDFS文件夾刪除

@Test

public void testDelete() throws Exception{

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.111:9000");

FileSystem fileSystem = FileSystem.get(conf);

//文件夾刪除

fileSystem.delete(new Path("/user/anna/test/hahaha"),true); //true表示遞歸刪除

//關(guān)閉資源

fileSystem.close();

}

6. HDFS文件名更改

@Test

public void testRename() throws Exception{

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.111:9000");

FileSystem fileSystem = FileSystem.get(conf);

//文件名稱更改

fileSystem.rename(new Path("/user/anna/test/copy.txt"), new Path("/user/anna/test/copyRename.txt"));

//關(guān)閉資源

fileSystem.close();

}

7. HDFS文件詳情查看

幾種實現(xiàn)方法

1. public abstract FileStatus[] listStatus(Path f) throws FileNotFoundException,IOException

* 返回FileStatus型數(shù)組

2. public FileStatus[] listStatus(Path f,PathFilter filter) throws FileNotFoundException,IOException

3. public FileStatus[] listStatus(Path[] files,PathFilter filter) throws FileNotFoundException,IOException

* 此時注意PathFilter是一個接口，里面只有一個方法:accept，本質(zhì)是對文件進行篩選

* Enumerate all files found in the list of directories passed in, calling listStatus(path, filter) on each one.

注意：以上方法返回的文件按照字母表順序排列

代碼：FileStatus[] listStatus(Path f)

//FileStatus[] listStatus(Path f)的使用

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//listStatus獲取/test目錄下信息

FileStatus[] fileStatuses = fileSystem.listStatus(new Path("/test"));

//遍歷輸出文件夾下文件

for(FileStatus fileStatus :fileStatuses) {

System.out.println(fileStatus.getPath() + " " + new Date(fileStatus.getAccessTime()) + " " +

fileStatus.getBlockSize() + " " + fileStatus.getPermission());

}

}catch(Exception e) {

e.printStackTrace();

}

在JDK1.8中輸出結(jié)果為：

----------------------------------------------------------------------------

hdfs://10.9.190.90:9000/test/hadoop-2.7.3.tar.gz 2012-07-26 134217728 rw-r--r--

hdfs://10.9.190.90:9000/test/hello.txt 2012-07-26 134217728 rw-r--r--

hdfs://10.9.190.90:9000/test/test2 1970-01-01 0 rwxr-xr-x

----------------------------------------------------------------------------

代碼：FileStatus[] listStatus(Path f,PathFilter filter)

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//列出目錄下后綴為.md的文件相關(guān)信息

FileStatus[] statuses = fileSystem.listStatus(new Path("/test/test2"), new PathFilter() {

@Override

public boolean accept(Path path) {

// TODO Auto-generated method stub

String string = path.toString();

if(string.endsWith(".md"))

return true;

else

return false;

}

});

//列出文件信息

for(FileStatus status : statuses) {

System.out.println("Path : " + status.getPath() + " Permisson : " + status.getPermission() +

" Replication : " + status.getReplication());

}

}catch(Exception e) {

e.printStackTrace();

}

7. 定位文件讀取

8. FileSystem類的學(xué)習(xí)

FileSystem的學(xué)習(xí)

今天的主要內(nèi)容

對照官方文檔進行FileSystem類的學(xué)習(xí)

FileSystem中的方法

* boolean exists(Path p)

* boolean isDirectory(Path p)

* boolean isFile(Path p)

* FileStatus getFileStatus(Path p)

* Path getHomeDirectory()

* FileStatus[] listStatus(Path path, PathFilter filter)

FileStatus[] listStatus(Path path)

FileStatus[] listStatus(Path[] paths, PathFilter filter)

FileStatus[] listStatus(Path[] paths)

* RemoteIterator[LocatedFileStatus] listLocatedStatus(Path path, PathFilter filter)

RemoteIterator[LocatedFileStatus] listLocatedStatus(Path path)

RemoteIterator[LocatedFileStatus] listFiles(Path path, boolean recursive)

* BlockLocation[] getFileBlockLocations(FileStatus f, int s, int l)

BlockLocation[] getFileBlockLocations(Path P, int S, int L)

* long getDefaultBlockSize()

long getDefaultBlockSize(Path p)

long getBlockSize(Path p)

* boolean mkdirs(Path p, FsPermission permission)

* FSDataOutputStream create(Path, ...)

FSDataOutputStream append(Path p, int bufferSize, Progressable progress)

FSDataInputStream open(Path f, int bufferSize)

* boolean delete(Path p, boolean recursive)

* boolean rename(Path src, Path d)

* void concat(Path p, Path sources[])

* boolean truncate(Path p, long newLength)

* interface RemoteIterator

boolean hasNext()

E next()

* interface StreamCapabilities

boolean hasCapability(capability)

準備工作

start-dfs.sh啟動hadoop集群

eclipse進行hdfs文件系統(tǒng)的訪問

導(dǎo)入相應(yīng)的jar包

創(chuàng)建與hdfs的連接并獲取FileSystem文件對象

第一種方式

* public static FileSystem get(Configuration conf) throws IOException

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000"); //namenode上的IP地址端口為：9000

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

第二種方式

* public static FileSystem get(URI uri,Configuration conf,String user)

throws IOException,

InterruptedException

URL.setURLStreamHandlerFactory(new FsUrlStreamHandlerFactory());

FileSystem fileSystem = FileSystem.get(new URI("hdfs://10.9.190.90:9000"),new Configuration(),"root");

//此時工作目錄會相應(yīng)更改為/user/root

兩種方式比較

第二種方式可能會拋出InterruptedException異常，因為

the static FileSystem get(URI uri, Configuration conf,String user) method MAY return a pre-existing instance of a filesystem client class—a class that may also be in use in other threads. The implementations of FileSystem shipped with Apache Hadoop do not make any attempt to synchronize access to the working directory field.(此時get方法可能會返回一個已經(jīng)存在FileSystem對象，也就是存在線程異步問題，所以我們盡量用前一種方式來完成FileSystem對象的創(chuàng)建)

org.apache.hadoop.fs.FileSystem簡介

The abstract FileSystem class is the original class to access Hadoop filesystems; non-abstract subclasses exist for all Hadoop-supported filesystems.(抽象基類FileSystem定義了對hadoop文件系統(tǒng)的操作)

All operations that take a Path to this interface MUST support relative paths. In such a case, they must be resolved relative to the working directory defined by setWorkingDirectory().(setWorkingDirectory()方法默認工作目錄)

FileSystem中的getWorkingDirector()返回當前系統(tǒng)的工作目錄

代碼

//獲得與hdfs文件系統(tǒng)的連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.90:9000");

//獲取文件系統(tǒng)對象

FileSystem fileSystem = FileSystem.get(conf);

//獲取當前工作目錄

System.out.println("=========獲取當前工作目錄=============");

System.out.println(fileSystem.getWorkingDirectory());

//設(shè)置新的工作目錄

//System.out.println("=========設(shè)置新的工作目錄=============");

fileSystem.setWorkingDirectory(new Path("hdfs://10.9.190.90:9000/user/anna")); //Path在hdfs中的作用和File作用類似，代表路徑

結(jié)果

=========獲取當前工作目錄=============

hdfs://10.9.190.90:9000/user/root

=========獲取設(shè)置后工作目錄=============

hdfs://10.9.190.90:9000/user/anna

FileSystem方法——判斷功能

預(yù)備知識

import org.apache.hadoop.fs.Path;類似于java.io.File代表hdfs的文件路徑

方法

public boolean exists(Path f) throws IOException

判斷文件是否存在

public boolean isDirectory(Path f) throws IOException

判斷是否為目錄

public boolean isFile(Path f) throws IOException

判斷是否為文件

練習(xí)

try {

//獲得與hdfs文件系統(tǒng)的連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.90:9000");

//獲取連接對象

FileSystem fileSystem = FileSystem.get(conf);

//判斷文件是否存在

System.out.println(fileSystem.exists(new Path("/test"))); //true

//判斷是否為目錄

System.out.println(fileSystem.isDirectory(new Path("/test"))); //true

//判斷是否為文件

System.out.println(fileSystem.isFile(new Path("/test"))); //false

}catch(Exception e) {

e.printStackTrace();

}

FileSystem方法——獲取功能—文件信息獲取

方法

public abstract FileStatus getFileStatus(Path f) throws IOException

Return a file status object that represents the path.

返回的是FileStatus對象類型

public Path getHomeDirectory()

Return the current user's home directory in this FileSystem. The default implementation returns "/user/$USER/".

返回當前用戶的home目錄

練習(xí)

try {

//獲得與hdfs文件系統(tǒng)的連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS", "hdfs://10.9.190.90:9000");

//獲取連接對象

FileSystem fileSystem = FileSystem.get(conf);

//獲取當前用戶的home目錄

System.out.println("========當前用戶的home目錄============");

Path path = fileSystem.getHomeDirectory();

System.out.println(path);

//獲取文件狀態(tài)對象

System.out.println("============文件信息===============");

FileStatus status = fileSystem.getFileStatus(new Path("/eclipse"));

System.out.println("Path : " + status.getPath());

System.out.println("isFile ? " + status.isFile());

System.out.println("Block size : " + status.getBlockSize());

System.out.println("Perssions : " + status.getPermission());

System.out.println("Replication : " + status.getReplication());

System.out.println("isSymlink : " + status.isSymlink());

}catch(Exception e) {

e.printStackTrace();

}

在JDK1.8中輸出結(jié)果為：

* ------------------------------------------------

* ========當前用戶的home目錄============

hdfs://10.9.190.90:9000/user/anna

============文件信息===============

Path : hdfs://10.9.190.90:9000/eclipse

isFile ? true

Block size : 134217728

Perssions : rw-r--r--

Replication : 3

isSymlink : false

------------------------------------------------

FileStatus中常用方法

public Path getPath()

public boolean isFile()

public boolean isSymlink()

public long getBlockSize()

public short getReplication()

public FsPermission getPermission()

FileSystem方法——獲取功能——文件夾遍歷1

方法

public abstract FileStatus[] listStatus(Path f) throws FileNotFoundException,IOException

返回FileStatus型數(shù)組

public FileStatus[] listStatus(Path f,PathFilter filter)

throws FileNotFoundException,IOException

public FileStatus[] listStatus(Path[] files,PathFilter filter)

throws FileNotFoundException,IOException

此時注意PathFilter是一個接口，里面只有一個方法:accept，本質(zhì)是對文件進行篩選

Enumerate all files found in the list of directories passed in, calling listStatus(path, filter) on each one.

注意：以上方法返回的文件按照字母表順序排列

練習(xí)1——FileStatus[] listStatus(Path f)的使用

//FileStatus[] listStatus(Path f)的使用

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//listStatus獲取/test目錄下信息

FileStatus[] fileStatuses = fileSystem.listStatus(new Path("/test"));

//遍歷輸出文件夾下文件

for(FileStatus fileStatus :fileStatuses) {

System.out.println(fileStatus.getPath() + " " + new Date(fileStatus.getAccessTime()) + " " +

fileStatus.getBlockSize() + " " + fileStatus.getPermission());

}

}catch(Exception e) {

e.printStackTrace();

}

在JDK1.8中輸出結(jié)果為：

----------------------------------------------------------------------------

hdfs://10.9.190.90:9000/test/hadoop-2.7.3.tar.gz 2012-07-26 134217728 rw-r--r--

hdfs://10.9.190.90:9000/test/hello.txt 2012-07-26 134217728 rw-r--r--

hdfs://10.9.190.90:9000/test/test2 1970-01-01 0 rwxr-xr-x

----------------------------------------------------------------------------

練習(xí)2——FileStatus[] listStatus(Path f,PathFilter filter)的使用

需求：列出/test/test2目錄下以.md結(jié)尾的問價信息

代碼：

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//列出目錄下后綴為.md的文件相關(guān)信息

FileStatus[] statuses = fileSystem.listStatus(new Path("/test/test2"), new PathFilter() {

@Override

public boolean accept(Path path) {

// TODO Auto-generated method stub

String string = path.toString();

if(string.endsWith(".md"))

return true;

else

return false;

}

});

//列出文件信息

for(FileStatus status : statuses) {

System.out.println("Path : " + status.getPath() + " Permisson : " + status.getPermission() +

" Replication : " + status.getReplication());

}

}catch(Exception e) {

e.printStackTrace();

}

注意問題

By the time the listStatus() operation returns to the caller, there is no guarantee that the information contained in the response is current. The details MAY be out of date, including the contents of any directory, the attributes of any files, and the existence of the path supplied.(listStatus()方法線程不安全)

FileSystem方法——獲取功能——文件夾遍歷2

方法

public org.apache.hadoop.fs.RemoteIterator listLocatedStatus(Path f)

throws FileNotFoundException, IOException

protected org.apache.hadoop.fs.RemoteIterator listLocatedStatus(Path f,PathFilter filter)

throws FileNotFoundException, IOException

注意：此方法是protected的，protected權(quán)限是：本類，同一包下(子類或無關(guān)類)，不同包下子類

注意：LocatedFileStatus是FileStatus的子類

使用

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//列出目錄下后綴為.md的文件相關(guān)信息

RemoteIterator iterator = fileSystem.listLocatedStatus(new Path("/test/test2"));

while(iterator.hasNext()) {

LocatedFileStatus status = iterator.next();

System.out.println("Path : " + status.getPath() + " Permisson : " + status.getPermission() +

" Replication : " + status.getReplication());

}

}catch(Exception e) {

e.printStackTrace();

}

* 在JDK1.8中輸出結(jié)果為：

* ---------------------------------------------------------------------------------------------

* Path : hdfs://10.9.190.90:9000/test/test2/Map.md Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/test2/biji.md Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/test2/haha.txt Permisson : rw-r--r-- Replication : 3

---------------------------------------------------------------------------------------------

* */

與listStatus(Path p)不同的是

listStatus返回的是FileStatus[]數(shù)組類型，遍歷時可通過數(shù)組for-each進行遍歷

listLocatedStatus(Path p)返回的是LocatedFileStatus類型的RemoteIterator集合，通過迭代器進行遍歷輸出

但是要注意的是listLocatedStatus()方法本質(zhì)上內(nèi)部還是listStatus(Path p)實現(xiàn)的

FileSystem方法——獲取功能——文件夾遍歷3

方法

public org.apache.hadoop.fs.RemoteIterator listFiles(Path f,boolean recursive)

throws FileNotFoundException,IOException

遞歸遍歷出文件夾內(nèi)容以及子文件夾中內(nèi)容

使用

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//列出目錄下后綴為.md的文件相關(guān)信息

RemoteIterator iterator = fileSystem.listFiles(new Path("/test"),true);

while(iterator.hasNext()) {

LocatedFileStatus status = iterator.next();

System.out.println("Path : " + status.getPath() + " Permisson : " + status.getPermission() +

" Replication : " + status.getReplication());

}

}catch(Exception e) {

e.printStackTrace();

}

* 在JDK1.8中輸出結(jié)果為：

* ---------------------------------------------------------------------------------------------------

* Path : hdfs://10.9.190.90:9000/test/hadoop-2.7.3.tar.gz Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/hello.txt Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/test2/Map.md Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/test2/biji.md Permisson : rw-r--r-- Replication : 3

Path : hdfs://10.9.190.90:9000/test/test2/haha.txt Permisson : rw-r--r-- Replication : 3

---------------------------------------------------------------------------------------------------

* */

FileSystem方法——獲取功能——獲取文件block的位置

方法

public BlockLocation[] getFileBlockLocations(Path p,long start,long len) throws IOException

public BlockLocation[] getFileBlockLocations(FileStatus file,long start,long len) throws IOException

使用

//查看/test/hadoop的block存放位置

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

FileStatus status = fileSystem.getFileStatus(new Path("/test/hadoop"));

BlockLocation[] locations = fileSystem.getFileBlockLocations(status, 0,status.getLen());

for(BlockLocation location : locations) {

System.out.println("host : " + location.getHosts() + " name : " + location.getNames() + " length : " + location.getLength());

}

}catch(Exception e) {

e.printStackTrace();

}

在JDK1.8中輸出結(jié)果為：

------------------------------------------------------------------------------

host : [Ljava.lang.String;@18ece7f4 name : [Ljava.lang.String;@3cce57c7 length : 134217728

host : [Ljava.lang.String;@1cf56a1c name : [Ljava.lang.String;@33f676f6 length : 79874467

------------------------------------------------------------------------------

FileSystem方法——獲取功能——獲取到某文件的輸出流

方法

public FSDataOutputStream create(Path f) throws IOException

public FSDataOutputStream create(Path f,boolean overwrite)

throws IOException

overwrite - if a file with this name already exists, then if true, the file will be overwritten, and if false an exception will be thrown.

public FSDataOutputStream create(Path f,

Progressable progress)

throws IOException

Create an FSDataOutputStream at the indicated Path with write-progress reporting. Files are overwritten by default.

public FSDataOutputStream create(Path f,boolean overwrite,int bufferSize)

throws IOException

public FSDataOutputStream create(Path f,boolean overwrite,int bufferSize, Progressable progress)throws IOException

FSDataOutputStream append(Path p, int bufferSize, Progressable progress)

使用——將本地E:/hzy.jpg上傳到hdfs的/1.jpg

public static void main(String[] args) {

BufferedInputStream in = null;

FSDataOutputStream out = null;

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//獲取本地文件輸入流

File file = new File("E:/hzy.jpg");

in = new BufferedInputStream(new FileInputStream(file));

final long fileSize = file.length();

//獲取到/test/hello.txt的輸出流

out = fileSystem.create(new Path("/1.jpg"),new Progressable() {

long fileCount = 0;

@Override

public void progress() {

// TODO Auto-generated method stub

fileCount++;

System.out.println("總進度：" + (fileCount/fileSize)*100 + " %");

}

});

//拷貝

int len = 0;

while((len = in.read()) != -1) {

out.write(len); //此時也可以用:IOUtils.copyBytes(in,out,conf);

}

in.close();

out.close();

}catch(Exception e) {

e.printStackTrace();

}finally {

if(in != null) {

try {

in.close();

} catch (IOException e) {

// TODO Auto-generated catch block

e.printStackTrace();

}

if (out != null) {

try {

out.close();

} catch (IOException e) {

// TODO Auto-generated catch block

e.printStackTrace();

}

FileSystem方法——獲取功能——獲取到某文件的輸入流——讀取文件

方法

public FSDataInputStream open(Path f) throws IOException

public abstract FSDataInputStream open(Path f,int bufferSize)throws IOException

使用——將hdfs中的1.jpg拷貝到本地E:/hzy2.jpg

try {

//創(chuàng)建與HDFS連接

Configuration conf = new Configuration();

conf.set("fs.defaultFS","hdfs://10.9.190.90:9000");

//獲得fileSystem

FileSystem fileSystem = FileSystem.get(conf);

//獲取hdfs文件輸入流

FSDataInputStream in = fileSystem.open(new Path("/1.jpg"));

//獲取本地輸出流

BufferedOutputStream out = new BufferedOutputStream(new FileOutputStream(new File("E:/hzyCopy.jpg")));

int len = 0;

byte[] bArr = new byte[1024*3];

while((len = in.read(bArr)) != -1) {

out.write(bArr,0,len);

}

in.close();

out.close();

}catch(Exception e) {

e.printStackTrace();

}

FileSystem方法——創(chuàng)建功能

public boolean mkdirs(Path f) throws IOException

FileSystem方法——刪除功能

public abstract boolean delete(Path f,boolean recursive) throws IOException

設(shè)計線程同步問題

FileSystem方法——重命名功能

public abstract boolean rename(Path src,Path dst)throws IOException

FileSystem其他方法

public void concat(Path trg,Path[] psrcs)throws IOException

Concat existing files together.

public boolean truncate(Path f,long newLength)throws IOException

interface RemoteIterator

定義

public interface RemoteIterator {

boolean hasNext() throws IOException;

E next() throws IOException;

}

The primary use of RemoteIterator in the filesystem APIs is to list files on (possibly remote) filesystems.

使用

//listLocatedFileStatus(Path f)

public org.apache.hadoop.fs.RemoteIterator listLocatedStatus(Path f)

throws FileNotFoundException,IOException

//listLocatedStatus(Path f,PathFilter filter)

protected org.apache.hadoop.fs.RemoteIterator listLocatedStatus(Path f,PathFilter filter)

throws FileNotFoundException,IOException

//listStatusIterator(Path p)

public org.apache.hadoop.fs.RemoteIterator listStatusIterator(Path p)

throws FileNotFoundException,IOException

//listFiles(Path f,boolean recursive)

public org.apache.hadoop.fs.RemoteIterator listFiles(Path f,boolean recursive)

throws FileNotFoundException,IOException

interface StreamCapabilities

方法

public interface StreamCapabilities {

boolean hasCapability(String capability);

}

使用

hadoop2.7.3中無此方法，在2.9.1中才有

總結(jié)

以上是生活随笔為你收集整理的hdfs user 连接_通过API访问HDFS的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。

上一篇：卸载一直在创建还原点_Win10电脑开机
下一篇：布尔运算_实例解析！布尔运算运用技巧！