dwd_st_info.py
13.1 KB
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
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
"""
1. 计算上升词，热搜词，新出词
2. quantity_being_sold在售商品数
"""
import os
import sys
import pandas as pd
sys.path.append(os.path.dirname(sys.path[0])) # 上级目录
from pyspark.storagelevel import StorageLevel
from utils.templates import Templates
# from ..utils.templates import Templates
# from AmazonSpider.pyspark_job.utils.templates_test import Templates
from pyspark.sql.types import IntegerType
# 分组排序的udf窗口函数
from pyspark.sql.window import Window
from pyspark.sql import functions as F
class DwdStInfo(Templates):
def __init__(self, site_name='us', date_type="month", date_info='2022-1'):
super().__init__()
self.site_name = site_name
self.date_type = date_type
self.date_info = date_info
self.date_info_last = str()
self.date_info_last2 = str()
self.db_save = f'dwd_st_info'
self.spark = self.create_spark_object(app_name=f"{self.db_save}: {self.site_name}, {self.date_type}, {self.date_info}")
self.df_date = self.get_year_week_tuple()
self.df_save = self.spark.sql(f"select 1+1;")
self.df_st_info = self.spark.sql(f"select 1+1;")
self.df_repeat = self.spark.sql(f"select 1+1;")
self.df_st_asin_title = self.spark.sql(f"select 1+1;")
self.df_st_info_current = self.spark.sql(f"select 1+1;")
self.df_st_info_last = self.spark.sql(f"select 1+1;") # 上周/月/季度
self.df_st_info_last2 = self.spark.sql(f"select 1+1;") # 上上周/月/季度
self.df_st_info_duplicated = self.spark.sql(f"select 1+1;") # 不在当前周/月/季度
self.df_st_zr_page1_counts = self.spark.sql(f"select 1+1;")
self.date_info_tuple = tuple()
self.u_is_first = self.spark.udf.register("u_is_first", self.udf_is_first, IntegerType())
self.u_is_ascending = self.spark.udf.register("u_is_ascending", self.udf_is_ascending, IntegerType())
self.u_is_search = self.spark.udf.register("u_is_search", self.udf_is_search, IntegerType())
self.u_is_title_appear = self.spark.udf.register("u_is_title_appear", self.udf_is_title_appear, IntegerType())
self.reset_partitions(partitions_num=3)
self.partitions_by = ['site_name', 'date_type', 'date_info']
self.get_date_info_tuple()
@staticmethod
def udf_is_first(x):
"""针对flag字段判断是否为当前周新出的关键词"""
if x:
return 0
else:
return 1
@staticmethod
def udf_is_ascending(x):
if x >= 0.5:
return 1
else:
return 0
@staticmethod
def udf_is_search(x):
if x >= 0.8:
return 1
else:
return 0
@staticmethod
def udf_is_title_appear(search_term, title):
if search_term.lower() in title.lower():
return 1
else:
return 0
def read_data(self):
print("1.1 读取dim_st_info表")
if self.date_type == '4_week':
sql = f"select * from dim_st_info where (site_name='{self.site_name}' and date_type='month' and date_info in {self.date_info_tuple}) or " \
f"(site_name='{self.site_name}' and date_type='{self.date_type}' and date_info='{self.date_info}');"
else:
sql = f"select * from dim_st_info where site_name='{self.site_name}' and date_type='{self.date_type}' and date_info in {self.date_info_tuple};"
print("sql:", sql)
self.df_st_info = self.spark.sql(sql).cache()
self.df_st_info.show(10)
print("1.2 读取ods_rank_search_rate_repeat表")
sql = f"select rank as st_rank_avg, search_num as st_search_sum, rate as st_search_rate, search_sum as st_search_sum " \
f"from ods_rank_search_rate_repeat where site_name='{self.site_name}';"
self.df_repeat = self.spark.sql(sql).cache()
self.df_repeat.show(10)
print("1.3 读取ods_search_term_rank_zr和ods_asin_detail表")
# sql = f"select * from dim_st_asin_base_info left join" \
# f"where site_name='{self.site_name}' and dt in {self.year_week_tuple} and data_type='zr' and page=1 ;"
# self.df_st_asin_base_info = self.spark.sql(sql).cache()
# self.df_st_asin_base_info.show(10)
sql = f"""
SELECT a.search_term, a.asin, b.title FROM ods_search_term_rank_zr a left join
ods_asin_detail b on a.asin=b.asin and a.dt=b.dt where a.site_name ='{self.site_name}' and b.site_name ='{self.site_name}'
and a.page=1 and a.dt in {self.year_week_tuple} and b.dt in {self.year_week_tuple}"""
self.df_st_asin_title = self.spark.sql(sql).cache()
self.df_st_asin_title = self.df_st_asin_title.drop_duplicates(['search_term', 'asin'])
self.df_st_asin_title.show(10)
def handle_data(self):
self.handle_st_first()
self.handle_st_ascending()
self.handle_st_search()
self.handle_st_asin_title()
self.df_save = self.df_save.withColumn("site_name", F.lit(self.site_name))
self.df_save = self.df_save.withColumn("date_type", F.lit(self.date_type))
self.df_save = self.df_save.withColumn("date_info", F.lit(self.date_info))
self.df_save.show(10)
def get_date_info_tuple(self):
df_week_start = self.df_date.loc[(self.df_date.year_week == '2020-44')]
id_start = list(df_week_start.id)[0] if list(df_week_start.id) else 0
if self.date_type in ['week', '4_week']:
df_week_current = self.df_date.loc[self.df_date.year_week == self.date_info]
elif self.date_type == 'month':
df_week_current = self.df_date.loc[self.df_date.year_month == self.date_info]
elif self.date_type == 'quarter':
df_week_current = self.df_date.loc[self.df_date.year_quarter == self.date_info]
else:
print("date_type输入错误, 退出")
df_week_current = pd.DataFrame()
id_current_max = max(list(df_week_current.id)) if list(df_week_current.id) else 0
df_week_all = self.df_date.loc[(self.df_date.id >= id_start) & (self.df_date.id <= id_current_max)]
if self.date_type == 'week':
self.date_info_tuple = tuple(df_week_all.year_week)
if self.date_type == "4_week":
df_week_all = self.df_date.loc[(self.df_date.id >= id_start) & (self.df_date.id <= id_current_max - 5)]
self.date_info_tuple = tuple(set(df_week_all.year_month))
if self.date_type == 'month':
self.date_info_tuple = tuple(set(df_week_all.year_month))
if self.date_type == 'quarter':
self.date_info_tuple = tuple(set(df_week_all.year_quarter))
def handle_st_first(self):
print("新出词(当前周/4周/月/季度，第1次出现)")
# 匹配上周/月/季度
df_week_current = self.df_date.loc[(self.df_date.year_week == self.date_info)]
id_current = list(df_week_current.id)[0] if list(df_week_current.id) else 0
id_last = id_current - 1
id_last2 = id_current - 2
df_week_last = self.df_date.loc[self.df_date.id == id_last]
df_week_last2 = self.df_date.loc[self.df_date.id == id_last2]
self.date_info_last = list(df_week_last.year_week)[0] if list(df_week_last.year_week) else ''
self.date_info_last2 = list(df_week_last2.year_week)[0] if list(df_week_last2.year_week) else ''
self.df_st_info_current = self.df_st_info.filter(f"date_info='{self.date_info}'")
self.df_st_info_last = self.df_st_info.filter(f"date_info='{self.date_info_last}'").select("search_term", "st_rank_avg").withColumnRenamed("st_rank_avg", "st_rank_avg_last")
self.df_st_info_last2 = self.df_st_info.filter(f"date_info='{self.date_info_last2}'")
self.df_st_info_duplicated = self.df_st_info.filter(f"date_info!='{self.date_info}'")
self.df_st_info_duplicated = self.df_st_info_duplicated.select('search_term').dropDuplicates(
['search_term']).withColumn("st_is_first_text", F.lit(0))
self.df_st_info_current = self.df_st_info_current.join(
self.df_st_info_duplicated, on='search_term', how='left'
)
self.df_st_info_current = self.df_st_info_current.fillna(
{"st_is_first_text": 1}
)
self.df_st_info_current.show(10, truncate=False)
def handle_st_ascending(self):
print("上升词(相邻2周/月/季度，上升超过50%的排名)")
self.df_st_info_current = self.df_st_info_current.join(
self.df_st_info_last, on='search_term', how='left'
)
self.df_st_info_current = self.df_st_info_current.na.fill({'st_rank_avg_last': 0})
self.df_st_info_current = self.df_st_info_current.withColumn(
"st_is_ascending_text_rate",
(self.df_st_info_current.st_rank_avg_last - self.df_st_info_current.st_rank_avg) / self.df_st_info_current.st_rank_avg_last)
self.df_st_info_current = self.df_st_info_current.na.fill({'st_is_ascending_text_rate': -1})
self.df_st_info_current = self.df_st_info_current.withColumn(
"st_is_ascending_text", self.u_is_ascending(self.df_st_info_current.st_is_ascending_text_rate))
# self.df_st_info_current.select("search_term", "st_rank", "st_rank_last", "st_is_ascending_text_rate", "st_is_ascending_text").show(10, truncate=False)
self.df_st_info_current = self.df_st_info_current.drop("st_rank_avg_last")
self.df_st_info_current.show(10, truncate=False)
def handle_st_search(self):
print("热搜词(历史出现占比>=80%)")
df_counts = self.df_st_info.groupby(['search_term']).agg(F.count_distinct("date_info").alias("st_week_appear_counts"))
df_distinct = self.df_st_info.drop_duplicates(["date_info"])
df_distinct.select("search_term", "date_info").show(20, truncate=False)
self.df_st_info_current = self.df_st_info_current.join(
df_counts, on='search_term', how='left'
)
self.df_st_info_current = self.df_st_info_current.withColumn(f"st_week_counts", F.lit(len(df_distinct.to_pandas_on_spark().date_info.to_numpy())))
self.df_st_info_current.show(20, truncate=False)
self.df_st_info_current = self.df_st_info_current.withColumn(
"st_is_search_text_rate",
self.df_st_info_current[f"st_week_appear_counts"] / self.df_st_info_current[f"st_week_counts"])
self.df_st_info_current = self.df_st_info_current.withColumn(
"st_is_search_text", self.u_is_search(self.df_st_info_current.st_is_search_text_rate))
self.df_st_info_current.show(10, truncate=False)
def handle_st_quantity(self):
pass
def handle_st_asin_title(self):
# 只针对page=1的zr类型数据进行统计
self.df_st_asin_title = self.df_st_asin_title.withColumn(
"st_asin_in_tile_flag", self.u_is_title_appear(self.df_st_asin_title.search_term, self.df_st_asin_title.title)
)
df_st_zr_page1_counts = self.df_st_asin_title.groupby("search_term").count()
df_st_zr_page1_counts = df_st_zr_page1_counts.withColumnRenamed("count", "st_zr_page1_counts")
df_st_zr_page1_in_title_counts = self.df_st_asin_title.filter("st_asin_in_tile_flag=1").groupby("search_term").count()
df_st_zr_page1_in_title_counts = df_st_zr_page1_in_title_counts.withColumnRenamed("count", "st_zr_page1_in_title_counts")
self.df_st_zr_page1_counts = df_st_zr_page1_counts.join(
df_st_zr_page1_in_title_counts, on='search_term', how='left'
)
self.df_st_zr_page1_counts = self.df_st_zr_page1_counts.fillna(0)
self.df_st_zr_page1_counts = self.df_st_zr_page1_counts.withColumn(
"st_zr_page1_in_title_rate", self.df_st_zr_page1_counts.st_zr_page1_in_title_counts / self.df_st_zr_page1_counts.st_zr_page1_counts
)
self.df_st_zr_page1_counts.show(10, truncate=False)
self.df_save = self.df_st_info_current.join(
self.df_st_zr_page1_counts, on='search_term', how='left'
)
def handle_data_st_sold(self):
print("在售商品数")
if self.year >= 2022:
df_quantity = self.df_brand_week.filter("quantity_being_sold>0").select("search_term", "quantity_being_sold")
df_quantity = df_quantity.groupby(['search_term']).agg({"quantity_being_sold": "mean"})
df_quantity = df_quantity.withColumnRenamed("avg(quantity_being_sold)", "st_quantity_being_sold")
df_quantity = df_quantity.withColumn("st_quantity_being_sold", F.ceil(df_quantity.st_quantity_being_sold)) # 向上取整
self.df_brand_current = self.df_brand_current.join(df_quantity, on='search_term', how='left')
self.df_brand_current = self.df_brand_current.fillna({"st_quantity_being_sold": 0})
else:
self.df_brand_current = self.df_brand_current.withColumn("st_quantity_being_sold", F.lit(0))
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[3] # 参数3：年-周/年-月/年-季, 比如: 2022-1
handle_obj = DwdStInfo(site_name=site_name, date_type=date_type, date_info=date_info)
handle_obj.run()