Files
airflow/dags/jh/sample/samples_notice.py
T
2023-08-01 15:20:20 +08:00

103 lines
4.5 KiB
Python

from goojia_utils import ElasticSearchClient,dingutils,dateutils
import pandas as pd
import datetime
# 获取当前日期和2个月前的日期,统计2个月样本数据
now = datetime.datetime.today()
now_day = now.strftime('%Y%m%d')
last = dateutils.add_months(now,0,-2,0,0)
last_day = last.strftime('%Y%m%d')
# 区域维度,统计2个月内样本数据量
es206_client = ElasticSearchClient(host='192.168.0.206')
sql = 'select city,region,p_date,cat,root from seawater where cat = \'sell\' and p_date >='+str(last_day)+' and p_date<'+str(now_day)
df = es206_client.scan_by_sql_to_data_frame(sql)
df_region = pd.DataFrame(df.groupby(by=['city','region'])['_id'].count()).reset_index()
df_region = df_region.rename(columns={'_id':'count'})
df_region_res = df_region[df_region['count']<20]
if df_region_res.empty == False:
text = '【房源样本数量自动检测】数据已生成完毕,请点击 https://nextcloud.fangjia.com/s/KwxGsosMcnYxdoa 查看详情'
# 计算2个月的日期,用来计算网站数据量减少百分比
# 当前时间
now = datetime.datetime.today()
# 获取当前月份的第一天
first_day_of_month = datetime.datetime(now.year,now.month, 1)
# 获取1个月前的第一天
last_month = first_day_of_month - datetime.timedelta(days=1)
last_day_of_month = last_month.replace(day=1)
# 获取2个月前的第一天
last2_month = last_day_of_month - datetime.timedelta(days=1)
last2_day_of_month = last2_month.replace(day=1)
strat_time = last_day_of_month.strftime('%Y%m%d')
end_time = first_day_of_month.strftime('%Y%m%d')
strat_time2 = last2_day_of_month.strftime('%Y%m%d')
end_time2 = last_day_of_month.strftime('%Y%m%d')
# 房源网站维度,按当月数量,和当月数量相比上月减少百分比
sql1 = 'select city,region,p_date,cat,root from seawater where cat = \'sell\' and p_date >='+strat_time+' and p_date<='+end_time
sql2 = 'select city,region,p_date,cat,root from seawater where cat = \'sell\' and p_date >='+strat_time2+' and p_date<='+end_time2
df_root1 = es206_client.scan_by_sql_to_data_frame(sql1)
df_root2 = es206_client.scan_by_sql_to_data_frame(sql2)
df_root_group1 = pd.DataFrame(df_root1.groupby(by=['root'])['_id'].count()).reset_index()
df_root_group2 = pd.DataFrame(df_root2.groupby(by=['root'])['_id'].count()).reset_index()
df_root_group1 = df_root_group1.rename(columns={'_id':'count'})
df_root_group2 = df_root_group2.rename(columns={'_id':'count'})
# 房源网站,数量维度,数量少于100
df_root_res1 = df_root_group1[df_root_group1['count']<100]
if df_root_res1.empty == False:
if text == '':
text = '【房源样本数量自动检测】数据已生成完毕,请点击 https://nextcloud.fangjia.com/s/KwxGsosMcnYxdoa 查看详情'
# 房源网站,百分比维度,减少30%
df_root_merge = pd.merge(df_root_group1,df_root_group2,how='left',on=['root'])
df_root_res = df_root_merge.drop(df_root_merge[~((df_root_merge['count_x']<df_root_merge['count_y']) & ((df_root_merge['count_y']-df_root_merge['count_x'])/df_root_merge['count_y']>0.3))].index)
if df_root_res.empty == False:
if text == '':
text = '【房源样本数量自动检测】数据已生成完毕,请点击 https://nextcloud.fangjia.com/s/KwxGsosMcnYxdoa 查看详情'
path = "/home/airflow/data/jh/price"
file_name = f"{now_day}房源样本数量检测.xlsx"
air_file_name = path + '/' + file_name
def up_to_nextcloud():
"""
上传统计信息至nextcloud
"""
from nextcloud import NextCloud
NEXTCLOUD_URL = "https://nextcloud.fangjia.com"
NEXTCLOUD_USERNAME = "shenjian@fangjia.com"
NEXTCLOUD_PASSWORD = "goojia123456"
nextcloud_file_name = '/共享/建行/' + file_name
nxc = NextCloud(endpoint=NEXTCLOUD_URL, user=NEXTCLOUD_USERNAME, password=NEXTCLOUD_PASSWORD)
nxc.upload_file(air_file_name, nextcloud_file_name)
if text != '':
with pd.ExcelWriter(air_file_name) as writer:
if df_region_res.empty == False:
df_region_res.to_excel(writer, sheet_name="房源样本数量检测-区域维度", columns=["city","region",'count'], index=False)
if df_root_res1.empty == False:
df_root_res1.to_excel(writer, sheet_name="房源样本数量检测-房源网站维度(样本数)", columns=['root','count_x','count_y'], index=False)
if df_root_res.empty == False:
df_root_res.to_excel(writer, sheet_name="房源样本数量检测-房源网站维度(样本减少百分比)", columns=['root','count_x','count_y'], index=False)
up_to_nextcloud()
token = '1b6d0d782d3c316f2d3b6b18ea5144816170dd392befbf2042a24a01fb84b736'
dingutils.send_text(text,token)