spark_df = spark.createDataFrame(pd_df)

spark_df.write.mode('overwrite').format("hive").saveAsTable("dbname.tablename")

import pandas as pd
import numpy as np
from pyspark import SparkContext,SparkConf
from pyspark.sql import HiveContext,SparkSession
from pyspark.sql import SQLContext
pd_df = pd.DataFrame(np.random.randint(0,10,(3,4)),columns=['a','b','c'])
spark = SparkSession.builder.appName('pd_2_hive').master('local').enableHiveSupport().getOrCreate()
spark_df = spark.createDataFrame(pd_df)
#spark dataframe 有接口可以直接寫入到hive
spark_df.write.mode('overwrite').format("hive").saveAsTable("dbname.tablename")
'''
其中 overwrite 代表如果表中存在數(shù)據(jù)，那么新數(shù)據(jù)會(huì)將原來的數(shù)據(jù)覆蓋,此外還有append等模式,詳細(xì)介紹如下：
        * `append`: Append contents of this :class:`DataFrame` to existing data.
        * `overwrite`: Overwrite existing data.
        * `error` or `errorifexists`: Throw an exception if data already exists.
        * `ignore`: Silently ignore this operation if data already exists.
'''
#此外還可以將spark_df 注冊(cè)為臨時(shí)表，之后通過sql的方式寫到hive里
spark_df.registerTempTable('tmp_table')
tmp_sql = '''create table dbname.tablename as select * from tmp_table'''
spark.sql(tmp_sql)
spark.stop()

df.write().mode("overwrite").saveAsTable("tableName");
或
df.select(df.col("col1"),df.col("col2")) .write().mode("overwrite").saveAsTable("schemaName.tableName");
或
df.write().mode(SaveMode.Overwrite).saveAsTable("dbName.tableName");

df.createOrReplaceTempView("$tempTableName")
spark.sql("insert into table dbName.$hive_table_name PARTITION($partition_column) select * from $tempTableName")