import os
import sys
import findspark  # 一定要在最前面导入

'''初始化spark环境'''
findspark.init()
# Path for spark source folder
os.environ['SPARK_HOME'] = "G:\Spark\Install\spark-2.4.3-bin-hadoop2.7"
# Append pyspark to Python Path
sys.path.append("G:\Spark\Install\spark-2.4.3-bin-hadoop2.7\python")

'''示例'''
from pyspark import SparkContext,SparkConf
from pyspark.sql import SparkSession
from pyspark.sql import Row

# SparkContext是spark功能的主要入口
sc = SparkContext("local", "app")
RawSalesDataRDD = sc.textFile("G:\\Spark\\作业\\taxi.csv")
print(RawSalesDataRDD.take(5))
salesRDD = RawSalesDataRDD.map(lambda line: line.split(","))
print(salesRDD.take(5))
taxi_Rows = salesRDD.map(lambda p:
                         Row(
                             id=p[0],
                             lat=p[1],
                             lon=p[2],
                             time=p[3]
                        ))

sqlContext = SparkSession.builder.getOrCreate()
taxi_df = sqlContext.createDataFrame(taxi_Rows)
print(taxi_Rows.take(5))

print('查看dataframe的字段名称和前5行数据：')
taxi_df.printSchema()
taxi_df.show(5)

'''使用SQL语句  操作表数据'''
# #创建临时表taxi_table
taxi_df.registerTempTable("taxi_table")

# 查询编号为 5 的出租车的 GPS 数据的前 10 行
taxi_df.filter("id='5'").show(10)
taxi_df.where("id='5'").show(10)
sqlContext.sql("select * from taxi_table where id='5'").show(10)