ytjia

# -*- coding: utf-8 -*-

"""
春运发券数据统计
"""
def filter_cond():
    fcoupon = open(r'/Users/ytjia/Downloads/20w发券用户')
    fconsume = open(r'/Users/ytjia/Downloads/period_consume')
    fcouponbuy = open(r'/Users/ytjia/Downloads/consumed_coupon', 'w')
    fnormalbuy = open(r'/Users/ytjia/Downloads/consumed_normal', 'w')

ytjia

#include <math.h>  
  
const double x_pi = 3.14159265358979324 * 3000.0 / 180.0;  
  
void bd_encrypt(double gg_lat, double gg_lon, double &bd_lat, double &bd_lon)  
{  
    double x = gg_lon, y = gg_lat;  
    double z = sqrt(x * x + y * y) + 0.00002 * sin(y * x_pi);  
    double theta = atan2(y, x) + 0.000003 * cos(x * x_pi);  
    bd_lon = z * cos(theta) + 0.0065;  

ytjia

var data = sc.parallelize(Seq(("A", 2), ("A", 4), ("B", 2), ("Z", 0), ("B", 10)))
// data: org.apache.spark.rdd.RDD[(java.lang.String, Int)] = ParallelCollectionRDD[31] at parallelize at <console>:12

val avgValue = data.mapValues((_, 1)
        .reduceByKey((x, y) => (x._1 + y._1, x._2 + y._2))
        .mapValues{ case (sum, count) => (1.0 * sum) / count }
        .collectAsMap()
// avgValue: scala.collection.Map[java.lang.String,Double] = Map(Z -> 0.0, B -> 6.0, A -> 3.0)

ytjia

-- 通过地理位置过滤出特定高校的学生
-- 查询20140301前的地理信息数据，请将log.mobile_data_location替换为log.mobile_data_location_oldtb
-- v0.1北京大学20131101上报的终端
SELECT DISTINCT
    uuid
FROM
    log.mobile_data_location
WHERE
    dt = '20131101'
    AND marslat > 39.98714 AND marslat < 39.996889