pyspark读取es

本地运行把jar包放到spark目录下

方式一：sqlcontext

def readEs():
    conf = SparkConf().setAppName("es").setMaster("local[2]")
    sc = SparkContext(conf=conf)
    sqlContext = SQLContext(sc)
    df = sqlContext.read.format("org.elasticsearch.spark.sql") \
        .option("es.nodes.wan.only", "true") \
        .option("es.nodes","ip")\
        .option("es.port","port") \
        .option("es.net.http.auth.user", "user") \
        .option("es.net.http.auth.pass", "passwd") \
        .option("es.resource","metricbeat-6.8.10-2020.09.22/doc")\
        .load()
    df.registerTempTable("test")
    value = sqlContext.sql("select * from test")
    value.show()
    sc.stop()

方式二：SparkSession

def readEsThree():
    conf = SparkConf().setMaster("local[2]").setAppName("ES_test") \