SparkSql 常见使用

1. 使用sparkSession代替sparkContext
2. sparkSession读各种源:

• 读文件:val peopleDf = spark.read.textFile("E:\\07-hadoop\\01-spark\\sql\\people.txt")

1. df转rdd : val peopleDs = peopleDf.rdd
2. 将rdd变换为一行一行:
   val peopleRdd = peopleDs.map(line => Row(line.split(",")(0), line.split(",")(1).toInt))
3. 通过rdd和schema 创建临时表 spark.createDataFrame(peopleRdd,schema).createOrReplaceTempView("people")
4. 使用sql查询临时表:val peopleSqlDf = spark.sql("select * from people")
5. 两个dateFrame之间join等操作:

val rdd =

      locationSqlDf.join(peopleSqlDf,locationSqlDf("name") === peopleSqlDf("name")).drop(locationSqlDf("name")).toDF()