1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
"""
1. 上架日期
2. 分类id
"""
"""
author: 方星钧(ffman)
description: 基于ods_asin_detail历史表，计算出asin的历史数据指标(上架时间+bs分类id)
table_read_name: ods_asin_detail, , selection_off_line.dwd_bs_category_asin
table_save_name: dim_asin_history_info
table_save_level: dwd
version: 1.0
created_date: 2022-06-20
updated_date: 2022-06-20
"""
import os
import sys
from pyspark.storagelevel import StorageLevel
sys.path.append(os.path.dirname(sys.path[0])) # 上级目录
from utils.templates import Templates
# from ..utils.templates import Templates
# from AmazonSpider.pyspark_job.utils.templates import Templates
# 分组排序的udf窗口函数
from pyspark.sql.window import Window
from pyspark.sql import functions as F
from pyspark.sql.types import StringType, IntegerType
class DimAsinHistoryInfo(Templates):
def __init__(self, site_name="us", date_type="week", date_info="2022-1"):
super().__init__()
self.site_name = site_name
self.date_type = date_type
self.date_info = date_info
self.db_save = f"dim_asin_history_info"
self.spark = self.create_spark_object(app_name=f"{self.db_save} {self.site_name}, {self.date_info}")
self.df_date = self.get_year_week_tuple()
self.df_asin_detail = self.spark.sql(f"select 1+1;")
self.df_bs_category = self.spark.sql(f"select 1+1;")
self.df_bs_category_report = self.spark.sql(f"select 1+1;")
self.df_save = self.spark.sql(f"select 1+1;")
self.year_month = list(self.df_date.loc[self.df_date.year_week == f'{self.year_week}'].year_month)[0]
# self.reset_partitions_by()
self.partitions_by = ['site_name']
self.reset_partitions(20)
self.partitions_type = "dt"
self.u_year_week = self.spark.udf.register('u_year_week', self.udf_year_week, StringType())
@staticmethod
def udf_year_week(dt):
year, week = dt.split("-")[0], dt.split("-")[1]
if int(week) < 10:
return f"{year}-0{week}"
else:
return f"{year}-{week}"
def read_data(self):
print("1.1 读取ods_asin_detail和ods_keep_date表")
sql = f"select asin, launch_time, site_name, dt, 1 as type, rank, price, rating, total_comments from ods_asin_detail " \
f"where site_name='{self.site_name}' " \
f"union " \
f"select asin, launch_time, site_name, dt, 2 as type, null as rank, null as price, null as rating, null as total_comments from ods_keep_date " \
f"where site_name='{self.site_name}';"
self.df_asin_detail = self.spark.sql(sqlQuery=sql).cache()
self.df_asin_detail.show(10, truncate=False)
print("1.2 读取dwd_bs_category_asin表")
sql = f"select asin, cate_1_id, cate_current_id, dt from selection_off_line.dwd_bs_category_asin " \
f"where site='{self.site_name}' and dt !='9999-99';" # '9999-99'分区是去重不符合
self.df_bs_category = self.spark.sql(sqlQuery=sql).cache()
self.df_bs_category.show(10, truncate=False)
print("1.3 读取ods_one_category_report表")
sql = f"select cate_1_id, rank, orders from ods_one_category_report " \
f"where site_name='{self.site_name}' and dm='{self.year_month}';"
print("sql:", sql)
self.df_bs_category_report = self.spark.sql(sqlQuery=sql).cache()
self.df_bs_category_report.show(10, truncate=False)
def handle_data(self):
self.handle_data_duplicated()
self.handle_data_join()
self.handle_data_renamed()
self.df_save.show(10, truncate=False)
def handle_data_duplicated(self):
print("2.1 根据asin,dt去重")
self.df_asin_detail = self.df_asin_detail.withColumn(
"dt_sort", self.u_year_week(self.df_asin_detail.dt)
)
self.df_bs_category = self.df_bs_category.withColumn(
"dt_sort", self.u_year_week(self.df_bs_category.dt)
)
window = Window.partitionBy(['asin']).orderBy(
self.df_asin_detail.type.asc_nulls_last(),
self.df_asin_detail.dt_sort.desc()
)
self.df_asin_detail = self.df_asin_detail.withColumn("asin_dt_top", F.row_number().over(window=window))
self.df_asin_detail = self.df_asin_detail.filter("asin_dt_top=1")
# 合并keep_date的launch_time数据
# 这种写法会卡住
# asin_list = self.df_asin_detail.rdd.map(lambda x: x[0]).collect()
# self.df_keep_date = self.df_keep_date.filter(~self.df_keep_date.asin.isin(asin_list))
# self.df_asin_detail = self.df_asin_detail.unionByName(self.df_keep_date)
window = Window.partitionBy(['asin']).orderBy(
self.df_bs_category.dt_sort.desc()
)
self.df_bs_category = self.df_bs_category.withColumn("asin_dt_top", F.row_number().over(window=window))
self.df_bs_category = self.df_bs_category.filter("asin_dt_top=1")
self.df_asin_detail.groupby('dt_sort').count().show(30, truncate=False)
self.df_bs_category.groupby('dt_sort').count().show(30, truncate=False)
self.df_asin_detail = self.df_asin_detail.drop("asin_dt_top", "type", "dt_sort")
self.df_bs_category = self.df_bs_category.drop("asin_dt_top", "dt", "dt_sort")
def handle_data_join(self):
self.df_save = self.df_asin_detail.join(
self.df_bs_category, on=['asin'], how='left'
).join(
self.df_bs_category_report, on=['rank', 'cate_1_id'], how='left'
)
# self.df_save = self.df_save.withColumn(f"{self.partitions_type}", F.lit(self.date_info))
# quit()
def handle_data_renamed(self):
self.df_save = self.df_save.\
withColumnRenamed("launch_time", "asin_launch_time").\
withColumnRenamed("cate_1_id", "asin_bs_cate_1_id").\
withColumnRenamed("cate_current_id", "asin_bs_cate_current_id").\
withColumnRenamed("rank", "asin_rank").\
withColumnRenamed("price", "asin_price"). \
withColumnRenamed("rating", "asin_rating"). \
withColumnRenamed("total_comments", "asin_total_comments").\
withColumnRenamed("orders", "asin_bs_orders")
if __name__ == '__main__':
site_name = sys.argv[1] # 参数1：站点
# date_type = sys.argv[2] # 参数2：类型：week/4_week/month/quarter
date_info = sys.argv[2] # 参数3：年-周/年-月/年-季, 比如: 2022-1
# handle_obj = DwdAsinHistoryInfo(site_name=site_name, date_type=date_type, date_info=date_info)
handle_obj = DimAsinHistoryInfo(site_name=site_name, date_info=date_info)
handle_obj.run()