Sark SQL****编程初级实践
一、实验环境
操作系统:Ubunt 16.04.
Spark 版本:2.4.0。
数据库:MySQL。
Python 版本:3.4.3.
from pyspark import SparkContext, SparkConf
from pyspark.sql import SparkSession
import os
os.environ["JAVA_HOME"]="/home/spark021/servers/jdk"
os.environ["PYSPARK_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
os.environ["PYSPARK_DRIVER_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
spark= SparkSession.builder.appName("employee021").config(conf=SparkConf()).getOrCreate()
二、 实验内容与完成情况
1. Spark SQL基本操作
为 employee.json 创建 DataFrame,并编写 Python 语句完成下列操作:
(1)查询所有数据;
(2)查询所有数据,并去除重复的数据;
(3)查询所有数据,打印时去除id 字段;
(4)筛选出 age>30 的记录;
(5)将数据按 age 分组;
(6)将数据按name升序排列;
(7)取出前3行数据;
(8)查询所有记录的 name 列,并为其取别名为 userame;
(9)查询年龄 age的平均值;
(10)查询年龄 age 的最小值。
(1) 加载 JSON 文件并查询所有数据
df = spark.read.json("file:///home/spark021/data/employee021.json")
df.show()
(2) 去除重复的数据
unique_df = df.dropDuplicates()
unique_df.show()
(3) 去除 id 字段
df_no_id = df.selectExpr(*[c for c in df.columns if c != 'id'])
df_no_id.show()
(4) 筛选出 age > 30 的记录
older_than_30 = df.filter(df['age'] > 30)
older_than_30.show()
(5) 按 age 分组
grouped_by_age = df.groupBy('age')
grouped_by_age.count().show()
(6) 按 name 升序排列
sorted_by_name = df.orderBy('name')
sorted_by_name.show()
(7) 取出前3行数据
first_three = df.limit(3)
first_three.show()
(8) 重命名 name 列为 username
renamed_df = df.withColumnRenamed('name', 'username')
renamed_df.show()
(9)使用聚合函数avg()计算年龄的平均值
average_age = df.agg({"age": "avg"}).collect()0
print("年龄平均值:", average_age)
(10)使用聚合函数min()计算年龄的最小值
min_age = df.agg({"age": "min"}).collect()0
print("年龄最小值:", min_age)
关闭 SparkSession
spark.stop()
2.编程实现将RDD转换为DataFrame
from pyspark import SparkConf
from pyspark.sql import SparkSession,Row
from pyspark.sql.types import *
import os
os.environ["JAVA_HOME"]="/home/spark021/servers/jdk"
os.environ["PYSPARK_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
os.environ["PYSPARK_DRIVER_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
spark = SparkSession.builder.appName("RDD转换成DataFrame").config(conf=SparkConf()).getOrCreate()
*#**配置Spark应用程序
- employee021 = spark.sparkContext.
textFile("file:///home/spark021/data/employee021.txt").
map(lambda line: line.split(",")).
map(lambda x: Row(id=x[0],name=x[1],age=x[1]))
employee021.foreach(print)
*#**创建DataFrame(具有模式)
*** schemaPeople = spark.createDataFrame(employee021)
** *#**注册为临时表,才可以通过sql查询
- schemaPeople.createOrReplaceTempView("people")
peopleDF = spark.sql("select * from people ")
peopleDF.show()
peopleRDD = peopleDF.rdd.map(lambda p: "id"+":"+p.id+","+"name"+":"+p.name+","+"age"+":"+p.age)
peopleRDD.foreach(print)
3.编码实现利用DataFrame读写Mysql的数据
(1)创建数据库
创建表
create table employee021 (id int(4), name char(20), gender char(4), age int(4));
插入数据
insert into employee021 values(1,'Alice','F',22);
insert into employee021 values(2,'John','M',25);
查看表
select * from employee021;
(2)进入saprk,配置 Spark通过 JDBC 连按数据库 MySQL,编程实现利用 DataFrame 插入数据到 MySQL 中,最后打印出 age 的最大值和 age 的总和。
from pyspark.sql import Row
from pyspark.sql.types import *
from pyspark import SparkContext, SparkConf
from pyspark.sql import SparkSession
import os
os.environ["JAVA_HOME"]="/home/spark021/servers/jdk"
os.environ["PYSPARK_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
os.environ["PYSPARK_DRIVER_PYTHON"]="/home/spark021/servers/anaconda3/envs/pyspark/bin/python3.6"
spark=SparkSession.builder.config("spark.jars", "/home/spark021/servers/spark-local/jars/mysql-connector-java-5.1.40-bin.jar").getOrCreate()
#配置 Spark通过 JDBC 连按数据库 MySQL
employeeDF = spark.read
.format("jdbc")
.option("driver","com.mysql.jdbc.Driver")
.option("url","jdbc:mysql://localhost:3306/sparktest")
.option("dbtable","employee021")
.option("user","spark021")
.option("password","123456")
.load()
employeeDF.show()
#编程实现利用 DataFrame 插入数据到 MySQL 中
#设置模式信息
schema =StructType([StructField("id", IntegerType(),True),
StructField("name",StringType(),True),
StructField("gender",StringType(),True),
StructField("age",IntegerType(),True)])
#设置两条数据,表示两个员工的信息
employeeRDD =spark
.sparkContext
.parallelize(["3 Mary F 26","4 Tom M 23"])
.map(lambda x:x.split(" "))
#创建 Row对象,每个Row 对象都是 rowRDD 中的一行
rowRDD = employeeRDD.map(lambda p:Row(int(p[0].strip()),p[1].strip(),p[2].strip(),int(p[3].strip())))
#建立 Row 对象和模式之间的对应关系,也就是把数据和模式对应起来
employeeDF = spark.createDataFrame(rowRDD,schema)
#写人数据库
prop ={}
prop['user']='spark021'
prop['password']='123456'
prop['driver']="com.mysql.jdbc.Driver"
employeeDF.write.jdbc("jdbc:mysql://localhost:3306/sparktest",'employee021','append', prop)
employeeDF.show()
#最后打印出 age 的最大值和 age 的总和。
jdbcDF=spark.read.format("jdbc")
.option("driver","com.mysql.jdbc.Driver") .option("url","jdbc:mysql://localhost:3306/sparktest")
.option("dbtable","employee021") .option("user","spark021")
.option("password","123456")
.load()
jdbcDF.createOrReplaceTempView("employee021")
spark.sql("select MAX(age) from employee021").show()
spark.sql("select Sum(age) from employee021").show()
版权归原作者 丢不掉的喜欢 所有, 如有侵权,请联系我们删除。