pyspark读取hdfs文件并导入到hive中

01.创建对象,设定日志级别

from pyspark.sql import SparkSession
spark = SparkSession.builder.config("spark.driver.host","192.168.1.10")\
    .config("spark.ui.showConsoleProgress","false")\
    .appName("hdfs_hive").master("local[*]").enableHiveSupport().getOrCreate()
sc = spark.sparkContext
sc.setLogLevel("ERROR")

02.选定hive数据库，查看所有表

spark.sql("""
    use hive_test_one 
""")
spark.sql("""
    show tables
""").show()

​ 输出结果：

+-------------+-------------+-----------+
|     database|    tableName|isTemporary|
+-------------+-------------+-----------+
|hive_test_one|seeds_dataset|      false|
+-------------+-------------+-----------+

03.从hdfs加载数据

​ 查看结构是为了hive中创建表时使用

hdfs_df = spark.read.csv("hdfs://192.168.1.10:9000/HadoopFileS/DataSet/MLdataset/Customers.csv",inferSchema=True,header=True)
print(hdfs_df.count())
hdfs_df.printSchema()

​ 输出结果：

1000
root
 |-- Dummy_Id: string (nullable = true)
 |-- Email: string (nullable = true)
 |-- Address: string (nullable = true)
 |-- Avatar: double (nullable = true)
 |-- Avg Session Length: double (nullable = true)
 |-- Time on App: double (nullable = true)
 |-- Time on Website: double (nullable = true)
 |-- Length of Membership: double (nullable = true)
 |-- Yearly Amount Spent: string (nullable = true)

04.创建hive表，并查看

spark.sql("""
    create table IF NOT EXISTS Customers(
        Dummy_Id string ,
        Email string ,
        Address string,
        Avatar double ,
        Avg_Session_Length double ,
        Time_on_App double,
        Time_on_Website double ,
        Length_of_Membership double ,
        Yearly_Amount_Spent string 
    )
    ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' 
""")
spark.sql("show tables").show()
spark.sql("select * from Customers").show()

​ 输出结果：表已经创建好，但是没有数据

+-------------+-------------+-----------+
|     database|    tableName|isTemporary|
+-------------+-------------+-----------+
|hive_test_one|    customers|      false|
|hive_test_one|seeds_dataset|      false|
+-------------+-------------+-----------+

​

05.待导入数据注册为临时表

hdfs_df.createOrReplaceTempView("hdfs_df")
spark.sql("select * from hdfs_df").show()

​ 输出：

06.将数据插入hive表中

spark.sql("""
    insert into Customers select * from hdfs_df
""")
spark.sql("select * from Customers").show()

​ 输出结果：

07.删除hive中的表

# 删除hive中的表
spark.sql("drop table Customers")
spark.sql("show tables").show()

​ 输出结果：isTemporary列表示是否是临时表，比如之前注册的hdfs_df就是临时表