scala 代码实例——customer_extract

import org.apache.spark.SparkConfimport org.apache.spark.SparkContextimport org.apache.spark.sql.{DataFrame, SQLContext, SparkSession}import org.apache.spark.rdd.RDDimport util.control.Breaks._object customer_extract {  def main(args: Array[String]) {    val spark = SparkSession      .builder()      .appName("Spark SQL ")      .config("spark.some.config.option", "some-value")      .getOrCreate()    import spark.implicits._    val df = spark.read.json("/spark_data/spark_sql_data.json")    df.createOrReplaceTempView("data")    // Global temporary view is tied to a system preserved database `global_temp`    //获取所有用户Mac地址    val distinct_mac_DF = spark.sql("SELECT  DISTINCT mac FROM data")    val mac_array   = distinct_mac_DF.collect()    var i = 0    var result_string = ""    //对每一个用户（Mac）进行循环    while(i < mac_array.length){      var mac = mac_array(i)      var sql = "SELECT  `time` from data where mac = '"+mac+"' order by `time`  limit 1"      var min_time = (spark.sql(sql).collect())(0).toString.toInt      sql = "SELECT  `time` from data where mac = '"+mac+"' order by `time` desc limit 1"      var max_time  = (spark.sql(sql).collect())(0).toString.toInt      //第一层过滤，过滤掉 只检测到一次的用户      if(min_time == max_time)        {          break()        }      var now_time = min_time      var j = 0      var old_num = 0      var num  = 0      var flag = 0  // 0 为 正常降序,1 为 另一个 start_time      var start_time = 0      var leave_time = 0      /* 在最小时间和最大时间按照时间间隔从小到大循环*/      while(now_time <( max_time+ 300)){        now_time = min_time +(j * 120)        j = j + 1        sql = "SELECT  count(*) num from data where mac ='"+mac+"' and `time`>"+ now_time+""        num = (spark.sql(sql).collect())(0).toString.toInt        if (j == 1) {          old_num = num;          start_time = min_time;        } else {          if ((old_num > num) && (flag == 0)) {            old_num = num;          } else if ((num == old_num) && (flag == 0)) {            leave_time = now_time - 180;            //添加到结果集            result_string = result_string + """{ "mac":"""+mac + ","+ """ "start_time": """+start_time +","+ """  "leave_time": """+leave_time+"}\n"            //重置参数，准备记录下一次访问参数            start_time = 0;            leave_time = 0;            flag = 1;          } else if ((old_num > num) && (flag == 1)) {            start_time = now_time;            flag = 0;          }        }      }      i=i+1    }    //将结果集 存入 文件    import java.io._    val writer = new PrintWriter(new File("visit_records.txt" ))    writer.write(result_string)    writer.close()  }}