文章目录
读取文本文件
第一种方法:通过RDD配合case class转换DataFrame
步骤
一、创建测试所需的文本文件
在虚拟机的/export/servers/
目录下创建文本文件
cd /export/servers/
vim person.txt
1 zhangsan 20
2 lisi 29
3 wangwu 25
4 zhaoliu 30
5 tianqi 35
6 kobe 40
二、在spark-shell中执行以下操作
// 1.进入spark客户端
cd /export/servers/spark-2.2.0-bin-2.6.0-cdh5.14.0/
bin/spark-shell --master local[2]
// 2.读取创建好的文本文件,定义RDD为lineRDD,并对数据进行切割
scala> val lineRDD = sc.textFile("file:///export/servers/person.txt").map(x => x.split(" "))
lineRDD: org.apache.spark.rdd.RDD[Array[String]] = MapPartitionsRDD[7] at map at <console>:24
// 3.定义case class样例类
scala> case class Person(id: Int,name: String,age: Int)
defined class Person
// 4.关联RDD和case class
scala> val personRDD = lineRDD.map(x => Person(x(0).toInt,x(1),x(2).toInt))
personRDD: org.apache.spark.rdd.RDD[Person] = MapPartitionsRDD[8] at map at <console>:28
// 5.将RDD转换为DataFrame
scala> val personDF = personRDD.toDF
personDF: org.apache.spark.sql.DataFrame = [id: int, name: string ... 1 more field]
// 6.查看数据
scala> personDF.show
+---+--------+---+
| id| name|age|
+---+--------+---+
| 1|zhangsan| 20|
| 2| lisi| 29|
| 3| wangwu| 25|
| 4| zhaoliu| 30|
| 5| tianqi| 35|
| 6| kobe| 40|
+---+--------+---+
// tips! 将DataFrame转换为RDD直接调用rdd方法即可
scala> personDF.rdd.collect
res2: Array[org.apache.spark.sql.Row] = Array([1,zhangsan,20], [2,lisi,29], [3,wangwu,25], [4,zhaoliu,30], [5,tianqi,35], [6,kobe,40])
第二种方法:通过sparkSession构建DataFrame
// 1.直接读取文件即可
scala> val personDF2 = spark.read.text("file:///export/servers/person.txt")
personDF2: org.apache.spark.sql.DataFrame = [value: string]
// 2.查看数据
scala> personDF2.show
+-------------+
| value|
+-------------+
|1 zhangsan 20|
| 2 lisi 29|
| 3 wangwu 25|
| 4 zhaoliu 30|
| 5 tianqi 35|
| 6 kobe 40|
+-------------+
可以看到通过sparkSession直接读取的文本文件,查询数据发现每一行的数据都统一放到了一个字段,而通过第一种方法就会按照字段分开,所以读取文本文件时一般更推荐第一种方法
读取json文件
// 1.spark提供了json格式的example,可以直接读取
scala> val jsonDF = spark.read.json("file:///export/servers/spark-2.2.0-bin-2.6.0-cdh5.14.0/examples/src/main/resources/people.json")
jsonDF: org.apache.spark.sql.DataFrame = [age: bigint, name: string]
// 2.查看数据
scala> jsonDF.show
+----+-------+
| age| name|
+----+-------+
|null|Michael|
| 30| Andy|
| 19| Justin|
+----+-------+
读取parquet列式存储文件
// 1.spark也提供了parquet格式的example,可以直接读取
scala> val parquetDF = spark.read.parquet("file:///export/servers/spark-2.2.0-bin-2.6.0-cdh5.14.0/examples/src/main/resources/users.parquet")
parquetDF: org.apache.spark.sql.DataFrame = [name: string, favorite_color: string ... 1 more field]
// 2.查看数据
scala> parquetDF.show
+------+--------------+----------------+
| name|favorite_color|favorite_numbers|
+------+--------------+----------------+
|Alyssa| null| [3, 9, 15, 20]|
| Ben| red| []|
+------+--------------+----------------+