#配置本地spark的环境,必须放在最前面
import findspark
findspark.init()
print(findspark.find())
from pyspark.sql import SparkSession
#新建sparksession
sparksession = SparkSession.builder.master("local[*]").appName("hive_test_1") \
.config("hive.metastore.uris","thrift://158.158.4.49:9083") \
.enableHiveSupport().getOrCreate()
#从hive读数据
sql_1="select * from test.biao_4"
df1 = sparksession.sql(sql_1)
df1.show()
#数据写入到hive中
df2 = sparksession.createDataFrame((
(1, "asf"),
(2, "2143"),
(3, "rfds")
)).toDF("label", "sentence")
df2.write.mode("overwrite").saveAsTable("test.biao_6")
有些文章说,需要hdfs,hive的xml文件到项目中,奇怪,本人测试都不需要,非常轻松,比java spark本地操作hive要简单的多
java spark本地操作hive可查看博文
https://blog.csdn.net/qq_41712271/article/details/103206827