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
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
"""
author: 方星钧(ffman)
description: 清洗6大站点对应的 “ods_brand_analytics” 的表: 排名权重计算，用天补全周/30天/月，存储新增的关键词
table_read_name: ods_brand_analytics
table_save_name: ods_brand_analytics
table_save_level: ods
version: 1.0
created_date: 2022-11-21
updated_date: 2022-11-21
"""
import os
import sys
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 StringType
# 分组排序的udf窗口函数
from pyspark.sql.window import Window
from pyspark.sql import functions as F
class OdsBrandAnalytics(Templates):
def __init__(self, site_name='us', date_type="month", date_info='2022-01'):
super().__init__()
self.site_name = site_name
self.date_type = date_type
self.date_info = date_info
self.date_info2 = date_info
self.db_save = f'ods_brand_analytics'
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() # pandas的df对象
self.df_st = self.spark.sql(f"select 1+1;")
self.df_st_current = self.spark.sql(f"select 1+1;")
self.df_st_rank = self.spark.sql(f"select 1+1;")
self.df_save = self.spark.sql(f"select 1+1;")
self.partitions_num = 1
self.reset_partitions(partitions_num=self.partitions_num)
self.partitions_by = ['site_name', 'date_type', 'date_info']
self.get_year_week_tuple()
if self.date_type in ['4_week', "last30day"]:
print(f"date_type={self.date_type}, 无需导入数据")
else:
self.handle_st_import()
# if self.date_type == '4_week':
# self.date_info = '2022-12-17'
self.get_date_info_tuple()
def read_data(self):
if (self.date_type == 'week' and date_info >= '2023-21') or self.date_type == 'month_week':
# 周的搜索词排名从2023-21周开始出现大量重复, 需要动态判断, 决定是否根据id大小给出新的排名
pass
else:
if self.date_type == '4_week':
# if self.site_name in ['us']:
# params1 = f"date_type='day' and date_info in {self.date_info_tuple}"
# else:
# params1 = f"date_type='week' and date_info in {self.year_week_tuple}"
params1 = f"date_type='week' and date_info in {self.year_week_tuple} and rank <= 1500000"
params2 = f" limit 0"
elif self.date_type == 'week_old':
# 旧版周表导入之后直接退出
quit()
elif self.date_type in ['month_old']:
params1 = f"date_type='week_old' and date_info in {self.year_week_tuple} and rank <= 1500000"
params2 = f""
elif self.date_type in ['month']:
params1 = f"date_type='week' and date_info in {self.year_week_tuple} and rank <= 1500000"
params2 = f""
else:
params1 = f"date_type='day' and date_info in {self.date_info_tuple}"
params2 = f""
if self.date_type == "last30day":
params2 = f" limit 0"
print("1.1 读取ods_brand_analytics表")
# sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' " \
# f"and date_type='day' and date_info in {self.date_info_tuple};"
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' " \
f"and {params1};"
print("sql:", sql)
self.df_st = self.spark.sql(sql).cache()
self.df_st.show(10, truncate=False)
# if self.df_st.count() == 0:
# quit() # 此处停止会中断程序
# print("self.df_st:", self.df_st.drop_duplicates(['search_term']).count())
print("1.2 读取ods_brand_analytics表")
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' " \
f"and date_type='{self.date_type}' and date_info = '{self.date_info}' {params2};"
print("sql:", sql)
self.df_st_current = self.spark.sql(sql).cache()
self.df_st_current.show(10, truncate=False)
def handle_us_week_rank(self, year_week='2023-46'):
if self.date_type == 'month_week':
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' and date_type='week' and date_info = '{year_week}';"
else:
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' and date_type='{self.date_type}' and date_info = '{self.date_info}';"
print("sql:", sql)
self.df_st = self.spark.sql(sql).cache()
# 将读取的数据写入临时表
self.df_st.createOrReplaceTempView("temp_table")
self.df_st.unpersist() # 停止对表的读取操作， 从而可以进行覆盖写入
self.df_save = self.spark.sql("select * from temp_table").cache()
self.df_save.show(10, truncate=False)
st_count = self.df_save.count()
# st_max = self.df_save.rank.max(
# st_max = self.df_save.agg({"rank": "max"}).collect()[0][0]
# rate = st_max / st_count
if self.date_type != 'month_week':
st_max = self.df_save.agg({"rank": "max"}).collect()[0][0]
rate = st_max / st_count
if rate >= 0.95:
print("st_count, st_max, rate:", st_count, st_max, rate)
quit()
# elif st_count == 0:
# quit()
else:
if self.date_type == 'month_week':
# for year_week in self.year_week_tuple:
hdf_cmd = f"hdfs dfs -rm -f /home/big_data_selection/ods/ods_brand_analytics/site_name={self.site_name}/date_type=week/date_info={year_week}/*"
# pass # 无需删除
else:
hdf_cmd = f"hdfs dfs -rm -f /home/big_data_selection/ods/ods_brand_analytics/site_name={self.site_name}/date_type={self.date_type}/date_info={self.date_info}/*"
print("hdf_cmd:", hdf_cmd)
os.system(hdf_cmd)
window = Window.orderBy(
self.df_save.id.asc()
)
self.df_save = self.df_save.withColumn("rank", F.row_number().over(window=window))
# self.df_save.write.saveAsTable(name=self.db_save, format='hive', mode='overwrite', partitionBy=self.partitions_by)
# quit()
if self.date_type == 'month_week':
self.df_save = self.df_save.withColumn("date_type", F.lit('week'))
self.df_save = self.df_save.withColumn("date_info", F.lit(year_week))
self.df_save.show(10, truncate=False)
self.save_data()
def handle_data(self):
if self.date_type in ['week', 'month_week']:
if self.date_type == 'month_week':
for year_week in self.year_week_tuple:
self.handle_us_week_rank(year_week=year_week)
# pass
# 计算month_week
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' and date_type='week' and date_info in {self.year_week_tuple} and rank <= 1500000;"
print("sql:", sql)
self.df_st = self.spark.sql(sql).cache()
self.df_st.show(10, truncate=False)
# 将读取的数据写入临时表
self.df_st.createOrReplaceTempView("temp_table")
self.df_st.unpersist() # 停止对表的读取操作， 从而可以进行覆盖写入
self.df_save = self.spark.sql("select * from temp_table").cache()
sql = f"select * from ods_brand_analytics where site_name='{self.site_name}' " \
f"and date_type='{self.date_type}' and date_info = '{self.date_info}';"
print("sql:", sql)
self.df_st_current = self.spark.sql(sql).cache()
self.df_st_current.show(10, truncate=False)
self.handle_st_rank()
self.handle_st_duplicated()
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_info2))
self.df_save.show(10, truncate=False)
# # df 是您的DataFrame
# nan_count_df = self.df_save.select([F.count(F.when(F.isnan(c) | F.col(c).isNull(), c)).alias(c) for c in self.df_save.columns])
# nan_count_df.show()
self.save_data()
quit()
else:
self.handle_us_week_rank()
# elif self.site_name == 'us' and self.date_type == 'month' and self.date_info >= '2023-09':
# quit()
else:
self.handle_st_rank()
self.handle_st_duplicated()
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_info2))
self.df_save.show(10, truncate=False)
def handle_st_import(self):
print(f"导入关键词数据: {self.site_name}, {self.date_type}, {self.date_info}")
if self.date_type in ['month_week', 'month']:
# if self.date_type == 'month':
os.system(f"/mnt/run_shell/sqoop_shell/import/ods_brand_analytics.sh {self.site_name} {self.date_type} {self.date_info}")
for year_week in self.year_week_tuple:
os.system(f"/mnt/run_shell/sqoop_shell/import/ods_brand_analytics.sh {self.site_name} week {year_week}")
else:
os.system(f"/mnt/run_shell/sqoop_shell/import/ods_brand_analytics.sh {self.site_name} {self.date_type} {self.date_info}")
def handle_st_rank_old(self):
self.df_st_rank = self.df_st.select("search_term", "rank", "date_info")
self.df_st_current = self.df_st_current.withColumn("flag", F.lit(1))
self.df_st_rank = self.df_st_rank.join(
self.df_st_current.select("search_term", "flag"), on='search_term', how='left'
)
self.df_st_rank = self.df_st_rank.filter("flag is null")
self.df_st_rank.show(10, truncate=False)
def handle_st_rank(self):
self.df_st_rank = self.df_st.select("search_term", "rank", "date_info")
self.df_st_current = self.df_st_current.withColumn("flag", F.lit(1))
# self.df_st_rank.show(10, truncate=False)
self.df_st_rank = self.df_st_rank.join(
self.df_st_current.select("search_term", "flag"), on='search_term', how='left'
)
self.df_st_rank = self.df_st_rank.filter("flag is null")
self.df_st_rank.show(10, truncate=False)
# count = self.df_st_current.count() # 计算当前周/月关键词的数量
df_count = self.df_st.groupby(['date_info']).count()
# df_count.show(10, truncate=False)
df_count = df_count.toPandas()
date_dict = {date_info: count for date_info, count in zip(df_count.date_info, df_count['count'])}
print("date_dict:", date_dict)
self.df_st_rank = self.df_st_rank.groupby(['search_term']). \
pivot("date_info").agg(F.mean("rank"))
self.df_st_rank.show(10, truncate=False)
self.df_st_rank = self.df_st_rank.fillna(date_dict)
self.df_st_rank = self.df_st_rank.withColumn("rank_sum", F.lit(0))
for col in date_dict.keys():
print("col:", col)
self.df_st_rank = self.df_st_rank.withColumn(
"rank_sum", self.df_st_rank.rank_sum + self.df_st_rank[col]
)
self.df_st_rank = self.df_st_rank.withColumn(
"rank_sum_avg", self.df_st_rank.rank_sum / len(self.date_info_tuple)
)
print("1111==============")
self.df_st_rank.show(10, truncate=False)
window = Window.orderBy(
self.df_st_rank.rank_sum_avg.asc()
)
self.df_st_rank = self.df_st_rank.withColumn("rank_avg", F.row_number().over(window=window))
self.df_st_rank = self.df_st_rank.drop("rank_sum", "rank_sum_avg")
print("2222==============")
self.df_st_rank.show(10, truncate=False) # 这里都没有问题
for col in date_dict.keys():
self.df_st_rank = self.df_st_rank.drop(col)
# self.df_st_rank.show(10, truncate=False)
self.df_st_rank = self.df_st_rank.withColumnRenamed("rank_avg", "rank")
# self.df_st_rank = self.df_st_rank.withColumn("rank", self.df_st_rank.rank+F.lit(self.df_st_current.rank.count()))
df_max_rank = self.df_st_current.agg(F.max('rank').alias("max_rank"))
df_max_rank.show(10, truncate=False)
df_max_rank = df_max_rank.toPandas()
max_rank = list(df_max_rank.max_rank)[0] if self.date_type not in ['4_week', 'last30day'] else 0
max_rank = max_rank if self.df_st_current.count() != 0 else 0
if self.date_type == 'last30day':
self.df_st_rank = self.df_st_rank.fillna({'rank': 0})
self.df_st_rank = self.df_st_rank.withColumn("rank", self.df_st_rank.rank+F.lit(max_rank))
# print("self.df_st_rank:", self.df_st_rank.count())
self.df_st_rank.show(10, truncate=False)
def handle_st_duplicated(self):
# 默认取最新一天的关键词数据
window = Window.partitionBy(['search_term']).orderBy(
self.df_st.date_info.desc()
)
self.df_st = self.df_st.withColumn("rank_top", F.row_number().over(window))
self.df_st = self.df_st.filter("rank_top=1")
self.df_st = self.df_st.drop("rank_top", "rank")
self.df_save = self.df_st_rank.join(
self.df_st, on='search_term', how='left'
)
# print("self.df_save:", self.df_save.count())
# self.df_save.show(10, truncate=False)
if __name__ == '__main__':
site_name = sys.argv[1] # 参数1：站点
date_type = sys.argv[2] # 参数2：类型：day/week/4_week/month/quarter
date_info = sys.argv[3] # 参数3：年-月-日/年-周/年-月/年-季, 比如: 2022-1
handle_obj = OdsBrandAnalytics(site_name=site_name, date_type=date_type, date_info=date_info)
handle_obj.run()