HDFS JAVA API
|字数总计:1.7k|阅读时长:8分钟|阅读量:|
一、 简介
想要使用 HDFS API,需要导入依赖 hadoop-client
。如果是 CDH 版本的 Hadoop,还需要额外指明其仓库地址:
<?xml version="1.0" encoding="UTF-8"?> <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion>
<groupId>com.ihadu</groupId> <artifactId>hdfs-java-api</artifactId> <version>1.0</version>
<properties> <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding> <hadoop.version>2.6.0-cdh5.15.2</hadoop.version> </properties>
<!---配置 CDH 仓库地址--> <repositories> <repository> <id>cloudera</id> <url>https: </repository> </repositories>
<dependencies> <!--Hadoop-client--> <dependency> <groupId>org.apache.hadoop</groupId> <artifactId>hadoop-client</artifactId> <version>${hadoop.version}</version> </dependency> <dependency> <groupId>junit</groupId> <artifactId>junit</artifactId> <version>4.12</version> <scope>test</scope> </dependency> </dependencies>
</project>
|
二、API的使用
2.1 FileSystem
FileSystem 是所有 HDFS 操作的主入口。由于之后的每个单元测试都需要用到它,这里使用 @Before
注解进行标注。
private static final String HDFS_PATH = "hdfs://192.168.0.106:8020"; private static final String HDFS_USER = "root"; private static FileSystem fileSystem;
@Before public void prepare() { try { Configuration configuration = new Configuration(); configuration.set("dfs.replication", "1"); fileSystem = FileSystem.get(new URI(HDFS_PATH), configuration, HDFS_USER); } catch (IOException e) { e.printStackTrace(); } catch (InterruptedException e) { e.printStackTrace(); } catch (URISyntaxException e) { e.printStackTrace(); } }
@After public void destroy() { fileSystem = null; }
|
2.2 创建目录
支持递归创建目录:
@Test public void mkDir() throws Exception { fileSystem.mkdirs(new Path("/hdfs-api/test0/")); }
|
2.3 创建指定权限的目录
FsPermission(FsAction u, FsAction g, FsAction o)
的三个参数分别对应:创建者权限,同组其他用户权限,其他用户权限,权限值定义在 FsAction
枚举类中。
@Test public void mkDirWithPermission() throws Exception { fileSystem.mkdirs(new Path("/hdfs-api/test1/"), new FsPermission(FsAction.READ_WRITE, FsAction.READ, FsAction.READ)); }
|
2.4 创建文件,并写入内容
@Test public void create() throws Exception { FSDataOutputStream out = fileSystem.create(new Path("/hdfs-api/test/a.txt"), true, 4096); out.write("hello hadoop!".getBytes()); out.write("hello spark!".getBytes()); out.write("hello flink!".getBytes()); out.flush(); out.close(); }
|
2.5 判断文件是否存在
@Test public void exist() throws Exception { boolean exists = fileSystem.exists(new Path("/hdfs-api/test/a.txt")); System.out.println(exists); }
|
2.6 查看文件内容
查看小文本文件的内容,直接转换成字符串后输出:
@Test public void readToString() throws Exception { FSDataInputStream inputStream = fileSystem.open(new Path("/hdfs-api/test/a.txt")); String context = inputStreamToString(inputStream, "utf-8"); System.out.println(context); }
|
inputStreamToString
是一个自定义方法,代码如下:
private static String inputStreamToString(InputStream inputStream, String encode) { try { if (encode == null || ("".equals(encode))) { encode = "utf-8"; } BufferedReader reader = new BufferedReader(new InputStreamReader(inputStream, encode)); StringBuilder builder = new StringBuilder(); String str = ""; while ((str = reader.readLine()) != null) { builder.append(str).append("\n"); } return builder.toString(); } catch (IOException e) { e.printStackTrace(); } return null; }
|
2.7 文件重命名
@Test public void rename() throws Exception { Path oldPath = new Path("/hdfs-api/test/a.txt"); Path newPath = new Path("/hdfs-api/test/b.txt"); boolean result = fileSystem.rename(oldPath, newPath); System.out.println(result); }
|
2.8 删除目录或文件
public void delete() throws Exception {
boolean result = fileSystem.delete(new Path("/hdfs-api/test/b.txt"), true); System.out.println(result); }
|
2.9 上传文件到HDFS
@Test public void copyFromLocalFile() throws Exception { Path src = new Path("D:\\BigData-Notes\\notes\\installation"); Path dst = new Path("/hdfs-api/test/"); fileSystem.copyFromLocalFile(src, dst); }
|
2.10 上传大文件并显示上传进度
@Test public void copyFromLocalBigFile() throws Exception {
File file = new File("D:\\kafka.tgz"); final float fileSize = file.length(); InputStream in = new BufferedInputStream(new FileInputStream(file));
FSDataOutputStream out = fileSystem.create(new Path("/hdfs-api/test/kafka5.tgz"), new Progressable() { long fileCount = 0;
public void progress() { fileCount++; System.out.println("上传进度:" + (fileCount * 64 * 1024 / fileSize) * 100 + " %"); } });
IOUtils.copyBytes(in, out, 4096);
}
|
2.11 从HDFS上下载文件
@Test public void copyToLocalFile() throws Exception { Path src = new Path("/hdfs-api/test/kafka.tgz"); Path dst = new Path("D:\\app\\");
fileSystem.copyToLocalFile(false, src, dst, true); }
|
2.12 查看指定目录下所有文件的信息
public void listFiles() throws Exception { FileStatus[] statuses = fileSystem.listStatus(new Path("/hdfs-api")); for (FileStatus fileStatus : statuses) { System.out.println(fileStatus.toString()); } }
|
FileStatus
中包含了文件的基本信息,比如文件路径,是否是文件夹,修改时间,访问时间,所有者,所属组,文件权限,是否是符号链接等,输出内容示例如下:
FileStatus{ path=hdfs://192.168.0.106:8020/hdfs-api/test; isDirectory=true; modification_time=1556680796191; access_time=0; owner=root; group=supergroup; permission=rwxr-xr-x; isSymlink=false }
|
2.13 递归查看指定目录下所有文件的信息
@Test public void listFilesRecursive() throws Exception { RemoteIterator<LocatedFileStatus> files = fileSystem.listFiles(new Path("/hbase"), true); while (files.hasNext()) { System.out.println(files.next()); } }
|
和上面输出类似,只是多了文本大小,副本系数,块大小信息。
LocatedFileStatus{ path=hdfs://192.168.0.106:8020/hbase/hbase.version; isDirectory=false; length=7; replication=1; blocksize=134217728; modification_time=1554129052916; access_time=1554902661455; owner=root; group=supergroup; permission=rw-r--r--; isSymlink=false}
|
2.14 查看文件的块信息
@Test public void getFileBlockLocations() throws Exception {
FileStatus fileStatus = fileSystem.getFileStatus(new Path("/hdfs-api/test/kafka.tgz")); BlockLocation[] blocks = fileSystem.getFileBlockLocations(fileStatus, 0, fileStatus.getLen()); for (BlockLocation block : blocks) { System.out.println(block); } }
|
块输出信息有三个值,分别是文件的起始偏移量 (offset),文件大小 (length),块所在的主机名 (hosts)。
这里我上传的文件只有 57M(小于 128M),且程序中设置了副本系数为 1,所有只有一个块信息。
原文:https://github.com/heibaiying/BigData-Notes