orc文件格式对常用系统的支持

1、Hive支持

创建表时指定orc格式即可:

create table tmp.orc_test(id bigint, name string, age int) stored as orc TBLPROPERTIES('orc.compress'='SNAPPY')

压缩格式有"SNAPPY"和 "ZLIB"两种,需要哪种格式指定即可。

2、SPARK支持

Spark读:
df  = spark.read.orc("/tmp/test/orc_data")  # 读出来的数据是一个dataframe

Spark写:
df.write.format("orc").save("/tmp/test/orc_data2")

3、Hadoop Streaming支持

3.1、读orc文件,输出text

hadoop jar /usr/local/hadoop-2.7.0//share/hadoop/tools/lib/hadoop-streaming-2.7.0.jar \
-libjars /usr/local/hive-1.2.0/lib/hive-exec-1.2.0-SNAPSHOT.jar \
-mapper /bin/cat -reducer /bin/cat \
-input /tmp/test/orc_test1 \
-output /tmp/test/orc_streaming_test3 \
-inputformat org.apache.hadoop.hive.ql.io.orc.OrcInputFormat 

返回的数据:

null    {"name":"123","age":"456"}
null    {"name":"456","age":"789"}

3.2、读orc文件,写orc文件:

hadoop jar /usr/local/hadoop-2.7.0//share/hadoop/tools/lib/hadoop-streaming-2.7.0.jar \
-libjars orc_maprd_test.jar \
-D orc.mapred.output.schema="struct<id:string,name:string,sex:string,age:string>" \
-input /tmp/test/orc_streaming_test \
-output /tmp/test/orc_streaming_test2 \
-inputformat org.apache.orc.mapred.OrcInputFormat \
-outputformat org.apache.orc.mapred.OrcOutputFormat \
-mapper is.orc.MyMapper -reducer is.orc.MyReducer 

pom.xml

<dependencies>
  <dependency>
    <groupId>org.apache.orc</groupId>
    <artifactId>orc-mapreduce</artifactId>
    <version>1.1.0</version>
  </dependency>
  <dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-mapreduce-client-core</artifactId>
    <version>2.7.0</version>
  </dependency>
</dependencies>

mapper:

package is.orc;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.Writable;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.Mapper;
import org.apache.hadoop.mapred.OutputCollector;
import org.apache.hadoop.mapred.Reporter;
import org.apache.orc.mapred.OrcStruct;

import java.io.IOException;
import java.util.Random;

class MyMapper implements Mapper<NullWritable,OrcStruct,LongWritable,Text> {

    Random random = new Random();

    public void close() { }

    public void map(NullWritable nullWritable, OrcStruct orcStruct, OutputCollector<LongWritable, Text> outputCollector, Reporter reporter) throws IOException {
        StringBuffer str = new StringBuffer();
        str.append(orcStruct.getFieldValue(0).toString() + "\t");
        str.append(orcStruct.getFieldValue(1).toString() + "\t");
        str.append(orcStruct.getFieldValue(2).toString() + "\t");
        str.append(orcStruct.getFieldValue(3).toString() );
        
        //不知道为什么Mapper的OutputKey只能用LongWritable,用随机数生成一个key,防止读orc文件后单reduce的情况
        LongWritable key = new LongWritable(random.nextInt(5)); 
        outputCollector.collect(key, new Text(str.toString()));
    }

    public void configure(JobConf jobConf) {
        jobConf.setMapOutputKeyClass(Writable.class);
        jobConf.setMapOutputValueClass(Text.class);
    }
}

Reducer:

package is.orc;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.OutputCollector;
import org.apache.hadoop.mapred.Reducer;
import org.apache.hadoop.mapred.Reporter;
import org.apache.orc.TypeDescription;
import org.apache.orc.mapred.OrcStruct;
import java.io.IOException;
import java.util.Iterator;


class MyReducer implements Reducer<LongWritable, Text, NullWritable, OrcStruct> {
    //要创建的ORC文件中的字段类型
    private TypeDescription schema = TypeDescription.fromString(
            "struct<id:string," +
                    "name:string," +
                    "sex:string," +
                    "age:string>"
    );

    private OrcStruct pair = (OrcStruct)OrcStruct.createValue(schema);


    public void reduce(LongWritable text, Iterator<Text> iterator, OutputCollector<NullWritable, OrcStruct> outputCollector, Reporter reporter) throws IOException {

        while (iterator.hasNext()) {
            String[] lineSplit = iterator.next().toString().split("\t");
            pair.setFieldValue("name",new Text(lineSplit[0]));
            pair.setFieldValue("sex",new Text(lineSplit[1]));
            pair.setFieldValue("age",new Text(lineSplit[2]));
            pair.setFieldValue("id",new Text(lineSplit[3]));
            break;
        }

        outputCollector.collect(NullWritable.get(),pair);
    }

    public void close() throws IOException {

    }

    public void configure(JobConf jobConf) {

    }
}

4、MapReduce支持

读orc的mapper:

package is.orc;

import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.orc.mapred.OrcStruct;
import org.apache.hadoop.mapreduce.Mapper;
import java.io.IOException;

public class OrcFileReadMapper extends Mapper<NullWritable, OrcStruct, Text, NullWritable> {

    private Text outputKey = new Text();

    @Override
    protected void map(NullWritable key, OrcStruct value, Context context) throws IOException, InterruptedException {
        StringBuffer sb= new StringBuffer();
        if (value.getFieldValue(0) == null){
            sb.append("-1\t");
        }else{
            sb.append(value.getFieldValue(0).toString() + "\t");      //通过下标索引获取数据
        }


        sb.append(value.getFieldValue(1).toString()+ "\t");
        sb.append(value.getFieldValue(2).toString()+ "\t");
        sb.append(value.getFieldValue(3).toString());      //也可以通过字段名获取数据

        outputKey = new Text(sb.toString());

        context.write(outputKey, NullWritable.get());
    }
}

写orc的reduer:

package is.orc;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import java.io.IOException;

import org.apache.hadoop.mapreduce.Reducer;
import org.apache.orc.TypeDescription;
import org.apache.orc.mapred.OrcStruct;

import java.io.IOException;

public class OrcFileWriteReducer extends Reducer<Text,NullWritable,NullWritable,OrcStruct> {

    //要创建的ORC文件中的字段类型
    private TypeDescription schema = TypeDescription.fromString(
            "struct<id:string," +
                    "name:string," +
                    "sex:string," +
                    "age:string>"
    );

    private OrcStruct pair = (OrcStruct)OrcStruct.createValue(schema);

    @Override
    protected void reduce(Text key, Iterable<NullWritable> values, Context context) throws IOException, InterruptedException {
        String line = key.toString();
        String[] lineSplit = line.trim().split("\t");

        pair.setFieldValue("id",new Text(lineSplit[0]));
        pair.setFieldValue("name",new Text(lineSplit[1]));
        pair.setFieldValue("sex",new Text(lineSplit[2]));
        pair.setFieldValue("age",new Text(lineSplit[3]));

        context.write(NullWritable.get(),pair);
    }
}

job配置:

package is.orc;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.conf.Configured;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.util.Tool;
import org.apache.hadoop.util.ToolRunner;
import org.apache.orc.mapred.OrcStruct;
import org.apache.orc.mapreduce.OrcInputFormat;
import org.apache.orc.mapreduce.OrcOutputFormat;

import java.io.IOException;

/**
 * @author lyf
 * @since 2018/06/16
 */
public class OrcFileWriteJob extends Configured implements Tool {

    public int run(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
        Configuration conf = getConf();

        conf.set("orc.mapred.output.schema","struct<id:string,name:string,sex:string,age:string>");
        String input = "/dws/dd_read_d_v2/dt=20180809/000000_0";
        String output = "/tmp/test/test_mr_orc";

        Job job = Job.getInstance(conf);

        job.setJarByClass(OrcFileWriteJob.class);
        job.setMapperClass(OrcFileReadMapper.class);
        job.setReducerClass(OrcFileWriteReducer.class);

        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(NullWritable.class);

        job.setOutputKeyClass(NullWritable.class);
        job.setOutputValueClass(OrcStruct.class);

        job.setInputFormatClass(OrcInputFormat.class);
        job.setOutputFormatClass(OrcOutputFormat.class);

        FileInputFormat.addInputPath(job,new Path(input));
        FileOutputFormat.setOutputPath(job,new Path(output));

        boolean rt = job.waitForCompletion(true);
        return rt?0:1;
    }

    public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();
        int retnum = ToolRunner.run(conf,new OrcFileWriteJob(),args);
    }
}


最后编辑于
©著作权归作者所有,转载或内容合作请联系作者
  • 序言:七十年代末,一起剥皮案震惊了整个滨河市,随后出现的几起案子,更是在滨河造成了极大的恐慌,老刑警刘岩,带你破解...
    沈念sama阅读 214,444评论 6 496
  • 序言:滨河连续发生了三起死亡事件,死亡现场离奇诡异,居然都是意外死亡,警方通过查阅死者的电脑和手机,发现死者居然都...
    沈念sama阅读 91,421评论 3 389
  • 文/潘晓璐 我一进店门,熙熙楼的掌柜王于贵愁眉苦脸地迎上来,“玉大人,你说我怎么就摊上这事。” “怎么了?”我有些...
    开封第一讲书人阅读 160,036评论 0 349
  • 文/不坏的土叔 我叫张陵,是天一观的道长。 经常有香客问我,道长,这世上最难降的妖魔是什么? 我笑而不...
    开封第一讲书人阅读 57,363评论 1 288
  • 正文 为了忘掉前任,我火速办了婚礼,结果婚礼上,老公的妹妹穿的比我还像新娘。我一直安慰自己,他们只是感情好,可当我...
    茶点故事阅读 66,460评论 6 386
  • 文/花漫 我一把揭开白布。 她就那样静静地躺着,像睡着了一般。 火红的嫁衣衬着肌肤如雪。 梳的纹丝不乱的头发上,一...
    开封第一讲书人阅读 50,502评论 1 292
  • 那天,我揣着相机与录音,去河边找鬼。 笑死,一个胖子当着我的面吹牛,可吹牛的内容都是我干的。 我是一名探鬼主播,决...
    沈念sama阅读 39,511评论 3 412
  • 文/苍兰香墨 我猛地睁开眼,长吁一口气:“原来是场噩梦啊……” “哼!你这毒妇竟也来了?” 一声冷哼从身侧响起,我...
    开封第一讲书人阅读 38,280评论 0 270
  • 序言:老挝万荣一对情侣失踪,失踪者是张志新(化名)和其女友刘颖,没想到半个月后,有当地人在树林里发现了一具尸体,经...
    沈念sama阅读 44,736评论 1 307
  • 正文 独居荒郊野岭守林人离奇死亡,尸身上长有42处带血的脓包…… 初始之章·张勋 以下内容为张勋视角 年9月15日...
    茶点故事阅读 37,014评论 2 328
  • 正文 我和宋清朗相恋三年,在试婚纱的时候发现自己被绿了。 大学时的朋友给我发了我未婚夫和他白月光在一起吃饭的照片。...
    茶点故事阅读 39,190评论 1 342
  • 序言:一个原本活蹦乱跳的男人离奇死亡,死状恐怖,灵堂内的尸体忽然破棺而出,到底是诈尸还是另有隐情,我是刑警宁泽,带...
    沈念sama阅读 34,848评论 5 338
  • 正文 年R本政府宣布,位于F岛的核电站,受9级特大地震影响,放射性物质发生泄漏。R本人自食恶果不足惜,却给世界环境...
    茶点故事阅读 40,531评论 3 322
  • 文/蒙蒙 一、第九天 我趴在偏房一处隐蔽的房顶上张望。 院中可真热闹,春花似锦、人声如沸。这庄子的主人今日做“春日...
    开封第一讲书人阅读 31,159评论 0 21
  • 文/苍兰香墨 我抬头看了看天上的太阳。三九已至,却和暖如春,着一层夹袄步出监牢的瞬间,已是汗流浃背。 一阵脚步声响...
    开封第一讲书人阅读 32,411评论 1 268
  • 我被黑心中介骗来泰国打工, 没想到刚下飞机就差点儿被人妖公主榨干…… 1. 我叫王不留,地道东北人。 一个月前我还...
    沈念sama阅读 47,067评论 2 365
  • 正文 我出身青楼,却偏偏与公主长得像,于是被迫代替她去往敌国和亲。 传闻我的和亲对象是个残疾皇子,可洞房花烛夜当晚...
    茶点故事阅读 44,078评论 2 352