package com.zhen.hdfs;
import java.io.IOException;
import java.io.OutputStream;
import java.net.URI;
import java.net.URISyntaxException;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
/**
* @author FengZhen
* @date 2018年8月12日
*
*/
public class FileSystemStatusAPI {
/**
* 文件元数据:FileStatus
* 任何文件系统的一个重要特征都是提供其目录结构浏览和检索它所在文件和目录相关信息的功能。
* FileStatus类封装了文件系统中文件和目录的元数据,包括文件长度、块大小、复本、修改时间、所有者以及权限信息
*/
private static FileSystem fileSystem;
public static void main(String[] args) {
//setUp();
//fileStatusForFile();
//tearDown();
//globbing();
pathFilter();
}
public static void setUp() {
String uri = "/user/hdfs/MapReduce/data/test_status";
Configuration configuration = new Configuration();
try {
fileSystem = FileSystem.get(new URI(uri), configuration);
OutputStream outputStream = fileSystem.create(new Path(uri));
outputStream.write("content".getBytes("UTF-8"));
outputStream.close();
} catch (IOException e) {
e.printStackTrace();
} catch (URISyntaxException e) {
e.printStackTrace();
}
}
public static void tearDown() {
if (fileSystem != null) {
try {
fileSystem.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
/**
* path=/user/hdfs/MapReduce/data/test_status
isDir=false
length=7
modificationTime=1534080334126
replication=3
blockSize=134217728
owner=FengZhen
group=hdfs
permissions=rw-r--r--
*/
public static void fileStatusForFile() {
Path file = new Path("/user/hdfs/MapReduce/data/test_status");
try {
FileStatus fileStatus = fileSystem.getFileStatus(file);
String path = fileStatus.getPath().toUri().getPath();
System.out.println("path="+path);
Boolean isDir = fileStatus.isDirectory();
System.out.println("isDir="+isDir);
long length = fileStatus.getLen();
System.out.println("length="+length);
long modificationTime = fileStatus.getModificationTime();
System.out.println("modificationTime="+modificationTime);
int replication = fileStatus.getReplication();
System.out.println("replication="+replication);
long blockSize = fileStatus.getBlockSize();
System.out.println("blockSize="+blockSize);
String owner = fileStatus.getOwner();
System.out.println("owner="+owner);
String group = fileStatus.getGroup();
System.out.println("group="+group);
String permissions = fileStatus.getPermission().toString();
System.out.println("permissions="+permissions);
} catch (IOException e) {
e.printStackTrace();
}
}
/**
* 文件模式
* 在单个操作中处理一批文件是一个很常见的需求。
* 在一个表达式中使用通配符来匹配多个文件是比较方便的,无需列举每个文件和目录来指定输入,该操作称为"通配"(globbing)。
* Hadoop为执行通配提供了两个FileSystem方法
* public FileStatus[] globStatus(Path pathPattern) throws IOException {
return new Globber(this, pathPattern, DEFAULT_FILTER).glob();
}
public FileStatus[] globStatus(Path pathPattern, PathFilter filter) throws IOException {
return new Globber(this, pathPattern, filter).glob();
}
globStatus方法返回与其路径匹配于指定模式的所有文件的FileStatus对象数组,并按路径排序。
PathFilter命令作为可选项可以进一步对匹配结果进行限制
*/
public static void globbing() {
String uri = "/user/hdfs/MapReduce/data";
Configuration configuration = new Configuration();
try {
fileSystem = FileSystem.get(new URI(uri), configuration);
// /2018/08/12 /2017/08/11
FileStatus[] fileStatus = fileSystem.globStatus(new Path("/user/hdfs/MapReduce/data/*/*/{11,12}"));
// 1./user/hdfs/MapReduce/data/201[78](201[7-8] 、 201[^01234569]) hdfs://fz/user/hdfs/MapReduce/data/2017 hdfs://fz/user/hdfs/MapReduce/data/2018
// 2./user/hdfs/MapReduce/data/*/*/11 hdfs://fz/user/hdfs/MapReduce/data/2017/08/11
// 3./user/hdfs/MapReduce/data/*/*/{11,12} hdfs://fz/user/hdfs/MapReduce/data/2017/08/11 hdfs://fz/user/hdfs/MapReduce/data/2018/08/12
for (FileStatus fileStatus2 : fileStatus) {
System.out.println(fileStatus2.getPath().toString());
}
fileSystem.close();
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* PathFilter
* 通配符模式并不总能够精确地描述我们想要访问的字符集。比如,使用通配格式排除一个特定文件就不太可能。
* FileSystem中的listStatus和globStatus方法提供了可选的pathFilter对象,以编程方式控制通配符
*/
public static void pathFilter() {
String uri = "/user/hdfs/MapReduce/data";
Configuration configuration = new Configuration();
try {
fileSystem = FileSystem.get(new URI(uri), configuration);
// /2018/08/12 /2017/08/11 新增一个/2017/08/12
FileStatus[] fileStatus = fileSystem.globStatus(new Path("/user/hdfs/MapReduce/data/201[78]/*/*"), new RegexExcludePathFilter("^.*/2017/08/11$"));
//FileStatus[] fileStatus = fileSystem.globStatus(new Path("/user/hdfs/MapReduce/data/2017/*/*"), new RegexExcludePathFilter("/user/hdfs/MapReduce/data/2017/08/11"));
for (FileStatus fileStatus2 : fileStatus) {
System.out.println(fileStatus2.getPath().toString());
}
fileSystem.close();
} catch (Exception e) {
e.printStackTrace();
}
}
}