从HDFS读取文件，把记录存到Hbase的java API操作

从HDFS读取文件，文件的数据格式类似如下

computer,xuzheng,54,52,86,91,42
computer,huangbo,85,42,96,38
english,zhaobenshan,54,52,86,91,42,85,75
english,liuyifei,85,41,75,21,85,96,14
algorithm,liuyifei,75,85,62,48,54,96,15

数据的意义：课程名 姓名 多次考试的分数

首先主类extends Configured implements Tool

public class ReadFromHdfsToHbase extends Configured implements Tool{
	private static final String ZK_CONNECT="hadoop02:2199,hadoop03:2199";
	private static final String TABLE_NAME="Student_Score";
	public static void main(String[] args) throws Exception {
		int run=ToolRunner.run(new ReadFromHdfsToHbase(), args);
		System.exit(run);
	}

	@Override
	public int run(String[] arg0) throws Exception {
		//设置环境变量
		Configuration conf=HBaseConfiguration.create();
		/**
		 * 思路
		 * 1.建立与hdfs的连接(设定zookeeper法定人数)
		 * */
		conf.set("hbase.zookeeper.quorum",ZK_CONNECT);
		System.setProperty("HADOOP_USER_NAME", "root");
		
		conf.addResource("config/core-site.xml");
		conf.addResource("config/hdfs-site.xml");
		conf.set("fs.defaultFS", "hdfs://myha01/");
		/**
		 * 创建一张family叫chengji的表
		 * 获取admin对象
		 * */
		HBaseAdmin haAdmin=new HBaseAdmin(conf);
		HTableDescriptor htd=new HTableDescriptor(TABLE_NAME);
		HColumnDescriptor family=new HColumnDescriptor("chengji".getBytes());
		//添加一个family名字是chengji到整个表的描述中
		htd.addFamily(family);
		//判断这张表是否存在
		if(haAdmin.tableExists(TABLE_NAME)){//如果存在这张表 先删除
			haAdmin.disableTable(TABLE_NAME);
			haAdmin.deleteTable(TABLE_NAME);
		}
		haAdmin.createTable(htd);
		//-----------
		//对job的设置
		Job job=Job.getInstance(conf);
		job.setJarByClass(ReadFromHdfsToHbase.class);
		job.setMapperClass(ReadFromHdfsToHbase_Mapper.class);
		//设置reduce
		TableMapReduceUtil.initTableReducerJob(TABLE_NAME,
				ReadFromHdfsToHbase_Reducer.class,
				job,
				null,
				null,
				null,
				null,
				false);
		// 设置数据读取组件
		job.setInputFormatClass(TextInputFormat.class);
		// 给mapper和reducer指定输出的key-value的类型
		job.setMapOutputKeyClass(Text.class);
		job.setMapOutputValueClass(NullWritable.class);
		job.setOutputKeyClass(NullWritable.class);
		job.setOutputValueClass(Put.class);
		
		// 指定输入数据的路径
		FileInputFormat.setInputPaths(job, new Path("/aaa/input1/"));
		boolean waitForCompletion = job.waitForCompletion(true);
		return waitForCompletion?0:1;
	}
}

配置文件建议创建一个source Folder,命名为config

用来存放hdfs-site.xml和core-site.xml

因为已经配好HA高可用集群，所以根据加载的配置文件，可以自动识别myha01代表的是两个namenode，系统会自动找到active的一个执行程序

public static class ReadFromHdfsToHbase_Mapper extends Mapper{
		@Override
		protected void map(LongWritable key, Text value, Mapper.Context context)
				throws IOException, InterruptedException {
			context.write(value, NullWritable.get());
		}
	}
	/**
	 * english,huanglei,85,75,85,99,66,88,75,91

rowkey computer family chengji qualify:name  qualify:score
	 * */
	public static class ReadFromHdfsToHbase_Reducer extends TableReducer{
		@Override
		protected void reduce(Text key, Iterable value,Context context)
				throws IOException, InterruptedException {
			
				String line=key.toString();
				String[] strs=line.split(",");
				String rowkey=strs[0]+"---"+strs[1];
				String family="chengji";
				String qualify1=strs[1];
				String qualify2=strs[2];
				Put put=new Put(rowkey.getBytes());//获取rowkey
				put.addColumn(family.getBytes(), "name".getBytes(), qualify1.getBytes());
				put.addColumn(family.getBytes(), "score".getBytes(), qualify2.getBytes());
				/**
				 * 输出格式
				 *  computer name:huangleu
				 *  computer score:99
				 * */
				ImmutableBytesWritable rowkeystr=new ImmutableBytesWritable(rowkey.getBytes());
				context.write(NullWritable.get(), put);
			
			
		}
	}

注意TableReducer的outputValue强制是mutation

mutation的实现类有如下四个，分别对应的是Hbase的添加删除