diff --git a/dags/mt/supply/data_stat.py b/dags/mt/supply/data_stat.py index 24f4f03..2aa2a1f 100644 --- a/dags/mt/supply/data_stat.py +++ b/dags/mt/supply/data_stat.py @@ -5,6 +5,7 @@ import pandas as pd from goojia_utils import dateutils, ClickHouseClient import datetime import sys +import numpy as np path = "/home/airflow/data/mt" @@ -13,7 +14,7 @@ clickhouse_client = ClickHouseClient(database='fangjia_stat') output_date = sys.argv[1] if sys.argv[1] else dateutils.format_date(datetime.datetime.now(), '%Y%m') print("输出日期" + output_date) -result_df = pd.read_csv(path + '/FJW_COMM_' + output_date, sep='@', encoding='utf-8') +result_df = pd.read_csv(path + '/FJW_COMM_' + output_date + ".csv", sep='@', encoding='utf-8') city_df = pd.read_excel('/opt/airflow/dags/repo/dags/mt/supply/data/美团19城市列表.xlsx') cities = city_df['城市列表'].values.tolist() @@ -38,6 +39,42 @@ aim_file_name = path + "/" + output_date + '覆盖率统计.xlsx' df.to_excel(aim_file_name, index=False) +def deal_nan_val(x): + if np.isnan(x['avg_price']): + x['avg_price'] = None + if np.isnan(x['rent_price']): + x['rent_price'] = None + if np.isnan(x['lng']): + x['lng'] = None + if np.isnan(x['lat']): + x['lat'] = None + if np.isnan(x['car_site_count']): + x['car_site_count'] = None + if np.isnan(x['households']): + x['households'] = None + if np.isnan(x['all_area']): + x['all_area'] = None + if np.isnan(x['build_area']): + x['build_area'] = None + if np.isnan(x['house_area']): + x['house_area'] = None + if np.isnan(x['deal_num']): + x['deal_num'] = None + if np.isnan(x['market']): + x['market'] = None + if np.isnan(x['subway']): + x['subway'] = None + if np.isnan(x['bank']): + x['bank'] = None + if np.isnan(x['bank']): + x['bank'] = None + if np.isnan(x['hospital']): + x['hospital'] = None + if np.isnan(x['property_fee']): + x['property_fee'] = None + return x + + def save2_clickhouse(): curr_month_start = output_date + '01' # result_df['avg_price'].fillna(-99.0, inplace=True) @@ -56,29 +93,28 @@ def save2_clickhouse(): # result_df['bank'].fillna(-99.0, inplace=True) # result_df['hospital'].fillna(-99.0, inplace=True) # result_df['property_fee'].fillna(-99.0, inplace=True) + result_df.fillna('null', inplace=True) result_df['plot_ratio'] = result_df['plot_ratio'].astype(str) result_df['landscaping_ratio'] = result_df['landscaping_ratio'].astype(str) result_df['property_life'] = result_df['property_life'].astype(str) result_df['finish_building_date'] = result_df['finish_building_date'].astype(str) result_df['house_type'] = result_df['house_type'].astype(str) - - #result_df.fillna('', inplace=True) - result_df['avg_price'].astype(int) - result_df['rent_price'].astype(int) - result_df['lng'].astype(int) - result_df['lat'].astype(int) - result_df['car_site_count'].astype(int) - result_df['households'].astype(int) - result_df['all_area'].astype(int) - result_df['build_area'].astype(int) - result_df['house_area'].astype(int) - result_df['deal_num'].astype(int) - result_df['school'].astype(int) - result_df['market'].astype(int) - result_df['subway'].astype(int) - result_df['bank'].astype(int) - result_df['hospital'].astype(int) - result_df['property_fee'].astype(int) + result_df['avg_price'] = result_df['avg_price'].astype(int, errors='ignore') + result_df['rent_price'] = result_df['rent_price'].astype(int, errors='ignore') + result_df['lng'] = result_df['lng'].astype(float, errors='ignore') + result_df['lat'] = result_df['lat'].astype(float, errors='ignore') + result_df['car_site_count'] = result_df['car_site_count'].astype(int, errors='ignore') + result_df['households'] = result_df['households'].astype(int, errors='ignore') + result_df['all_area'] = result_df['all_area'].astype(int, errors='ignore') + result_df['build_area'] = result_df['build_area'].astype(int, errors='ignore') + result_df['house_area'] = result_df['house_area'].astype(int, errors='ignore') + result_df['deal_num'] = result_df['deal_num'].astype(int, errors='ignore') + result_df['school'] = result_df['school'].astype(int, errors='ignore') + result_df['market'] = result_df['market'].astype(int, errors='ignore') + result_df['subway'] = result_df['subway'].astype(int, errors='ignore') + result_df['bank'] = result_df['bank'].astype(int, errors='ignore') + result_df['hospital'] = result_df['hospital'].astype(int, errors='ignore') + result_df['property_fee'] = result_df['property_fee'].astype(float, errors='ignore') for city in cities: temp_df = result_df[result_df['city'] == city] @@ -102,6 +138,6 @@ def up_to_nextcloud(): nxc.upload_file(aim_file_name, aim_file_path) -save2_clickhouse() up_to_nextcloud() +save2_clickhouse() # https://oapi.dingtalk.com/robot/send?access_token=e022515642e544942f86bbe234448edc3b8bddea8184e63b7edf4129b31c8e79 diff --git a/dags/mt/supply/deal.py b/dags/mt/supply/deal.py index 77b558a..09cfbf9 100644 --- a/dags/mt/supply/deal.py +++ b/dags/mt/supply/deal.py @@ -28,7 +28,7 @@ logging.info("日期信息:" + month_start + ' ' + curr_month_start) month_start_date = dateutils.parse_date(month_start, '%Y%m%d') curr_month_start_date = dateutils.parse_date(curr_month_start, '%Y%m%d') -result_df = pd.read_csv(path + '/FJW_COMM_' + dateutils.format_date(datetime.datetime.now(), '%Y%m') + '.csv', sep='@', encoding='utf-8') +result_df = pd.read_csv(path + '/FJW_COMM_' + output_date + '.csv', sep='@', encoding='utf-8') if 'deal_num' in result_df.columns: result_df = result_df.drop(columns=['deal_num']) diff --git a/dags/mt/supply/facilitys.py b/dags/mt/supply/facilitys.py index 7a708d6..6231864 100644 --- a/dags/mt/supply/facilitys.py +++ b/dags/mt/supply/facilitys.py @@ -73,7 +73,7 @@ district_facilitys_df = pd.merge(district_facilitys_df, district_facilitys_hospi district_facilitys_new_df = district_facilitys_df[ ['city', 'region', 'district_name', 'school', 'bank', 'hospital', 'supermarket', 'subway']] district_facilitys_new_df = district_facilitys_new_df.rename(columns={'district_name': 'name', 'supermarket': 'market'}) -district_facilitys_new_df = district_facilitys_new_df.drop_duplicates(subsets=['city', 'region', 'name'], keep='first') +district_facilitys_new_df = district_facilitys_new_df.drop_duplicates(subset=['city', 'region', 'name'], keep='first') result_df = pd.read_csv(path + '/FJW_COMM_' + output_date + '.csv', sep='@', encoding='utf-8') result_df = pd.merge(result_df, district_facilitys_new_df, how='left', on=['city', 'region', 'name']) @@ -83,25 +83,33 @@ result_df.loc[:, 'district_facilitys'] = None # 剔除特别大的 def format_facilitys(x): if type(x['school']) is not int: - x.school = None + x.school = 0 + else: + if x.school > 10: + x.school = 10 + if type(x['bank']) is not int: - x.bank = None + x.bank = 0 + else: + if x.bank > 30: + x.bank = 30 + if type(x['hospital']) is not int: - x.hospital = None + x.hospital = 0 + else: + if x.hospital > 10: + x.hospital = 10 if type(x['market']) is not int: - x.market = None + x.market = 0 + else: + if x.market > 15: + x.market = 15 + if type(x['subway']) is not int: - x.subway = None - if x.school > 10: - x.school = 10 - if x.market > 15: - x.market = 15 - if x.subway > 5: - x.subway = 5 - if x.bank > 30: - x.bank = 30 - if x.hospital > 10: - x.hospital = 10 + x.subway = 0 + else: + if x.subway > 5: + x.subway = 5 # 对于存在周边配套的输出字段 info = list()