SparkSql的语法一(调用方法)

调用方法（通过调用方法实现数据分析）

show：(以表格的形式展示数据集中前N行(20)记录)
select :(投影查询，指定查询的字段)
selectExpr :(支持表达式(基本运算或者别名)的投影查询)

df
.selectExpr("id+10", "name as username")

withColumn ：(添加额外列方法)
withColumnRenamed :(给列重命名方法 相当于sql中的as 别名)

val rdd = spark.sparkContext.makeRDD(List((1, "zs", true, 2000D), (2, "ls", false, 3000D)))

// rdd转换为ds或者df
import spark.implicits._
// scala隐式转换
val df = rdd.toDF("id", "name", "sex", "salary")

df
//.select("id", "name")
//.selectExpr("id+10", "name as username")
.withColumn("year_salary", $"salary" * 12)
.withColumnRenamed("name","username")
.show()
// select id,name,sex,salary,salary * 12 as  yearsalary from t_user

+---+----+-----+------+-----------+
| id|name|  sex|salary|year_salary|
+---+----+-----+------+-----------+
|  1|  zs| true|2000.0|    24000.0|
|  2|  ls|false|3000.0|    36000.0|
+---+----+-----+------+-----------+

+---+--------+-----+------+-----------+
| id|username|  sex|salary|year_salary|
+---+--------+-----+------+-----------+
|  1|      zs| true|2000.0|    24000.0|
|  2|      ls|false|3000.0|    36000.0|
+---+--------+-----+------+-----------+

printSchema ：(打印输出表的结果)

//结果类似与这样
root
 |-- id: integer (nullable = false)
 |-- username: string (nullable = true)
 |-- sex: boolean (nullable = false)
 |-- salary: double (nullable = false)
 |-- year_salary: double (nullable = false)

drop ：(用来删除特定列方法)
dropDuplicats :(特定列内容重复的数据只保留一个 (结果去重))

import spark.implicits._
val df = List(
  (1, "zs", false, 1, 15000),
  (2, "ls", false, 1,