mr编程案例6——倒排索引创建
需求:有大量的文本文档,如下所示:
a.txt
hello tom
hello jim
hello kitty
hello rose
b.txt
hello jerry
hello jim
hello kitty
hello jack
c.txt
hello jerry
hello java
hello c++
hello c++
需要得到以下结果:
hello a.txt–>4 b.txt–>4 c.txt–>4
java c.txt–>1
jerry b.txt–>1 c.txt–>1
…
知识点
-
要点1:map方法中,如何获取所处理的这一行数据所在的文件名?
worker在调map方法时,会传入一个context,而context中包含了这个worker所读取的数据切片信息,而切片信息又包含这个切片所在的文件信息
那么,就可以在map中:
FileSplit split = context.getInputSplit();
String fileName = split.getpath().getName(); -
inputsplit输入切片:
用于描述每个maptask所处理的数据范围如果maptask如果 maptask读的是文件:划分范围应该用如下信息描述:文件路径、偏移量范围。
如果 maptask读的是数据库的数据呢?:划分任务范应该用如下信息描述:
库名.表名,行范
解题思路
Step1:
先写一个mr程序:统计出每个单词在每个文件中的总次数
hello-a.txt 4
hello-b.txt 4
hello-c.txt 4
java-c.txt 1
jerry-b.txt 1
jerry-c.txt 1
map:
切词和提取文件名,将单词和文件名拼接,作为key,value设做1.
reduce:
将单词和文件名拼接在一起的单词,做计数
Step2:
2、然后在写一个mr程序,读取上述结果数据:
map:
根据-切,以单词做key,后面一段作为value
reduce:
拼接values里面的每一段,以单词做key,拼接结果做value,输出即可
Step1代码
public class IndexStepOne {
public static class IndexStepOneMapper extends Mapper<LongWritable, Text, Text, IntWritable> {
// 产生 <hello-文件名,1>
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
// 从输入切片信息中获取当前正在处理的一行数据所属的文件
FileSplit inputSplit = (FileSplit) context.getInputSplit();
String fileName = inputSplit.getPath().getName();
String[] words = value.toString().split(" ");
for (String w : words) {
// 将"单词-文件名"作为key,1作为value,输出
context.write(new Text(w + "-" + fileName), new IntWritable(1));
}
}
}
public static class IndexStepOneReducer extends Reducer<Text, IntWritable, Text, IntWritable> {
@Override
protected void reduce(Text key, Iterable<IntWritable> values,
Reducer<Text, IntWritable, Text, IntWritable>.Context context)
throws IOException, InterruptedException {
int count = 0;
for (IntWritable value : values) {
count += value.get();
}
context.write(key, new IntWritable(count));
}
}
public static void main(String[] args) throws Exception{
Configuration conf = new Configuration(); // 默认只加载core-default.xml core-site.xml
System.out.println(conf.get("hdfs.p"));
System.out.println(conf.get("yy.p"));
Job job = Job.getInstance(conf);
job.setJarByClass(IndexStepOne.class);
job.setMapperClass(IndexStepOneMapper.class);
job.setReducerClass(IndexStepOneReducer.class);
job.setNumReduceTasks(3);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(IntWritable.class);
FileInputFormat.setInputPaths(job, new Path("F:\\mrdata\\index\\input"));
FileOutputFormat.setOutputPath(job, new Path("F:\\mrdata\\index\\out1"));
job.waitForCompletion(true);
}
}
Step2代码
public class IndexStepTwo {
public static class IndexStepTwoMapper extends Mapper<LongWritable, Text, Text, Text> {
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
String[] split = value.toString().split("-");
context.write(new Text(split[0]), new Text(split[1].replaceAll("\t", "-->")));
}
}
public static class IndexStepTwoReducer extends Reducer<Text, Text, Text, Text> {
// 一组数据: <hello,a.txt-->4> <hello,b.txt-->4> <hello,c.txt-->4>
@Override
protected void reduce(Text key, Iterable<Text> values,Context context)
throws IOException, InterruptedException {
// stringbuffer是线程安全的,stringbuilder是非线程安全的,在不涉及线程安全的场景下,stringbuilder更快
StringBuilder sb = new StringBuilder();
for (Text value : values) {
sb.append(value.toString()).append("\t");
}
context.write(key, new Text(sb.toString()));
}
}
public static void main(String[] args) throws Exception{
Configuration conf = new Configuration(); // 默认只加载core-default.xml core-site.xml
Job job = Job.getInstance(conf);
job.setJarByClass(IndexStepTwo.class);
job.setMapperClass(IndexStepTwoMapper.class);
job.setReducerClass(IndexStepTwoReducer.class);
job.setNumReduceTasks(1);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Text.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
FileInputFormat.setInputPaths(job, new Path("F:\\mrdata\\index\\out1"));
FileOutputFormat.setOutputPath(job, new Path("F:\\mrdata\\index\\out2"));
job.waitForCompletion(true);
}
}