修復sys报错

This commit is contained in:
沈健
2025-06-05 16:59:52 +08:00
parent ba1d065337
commit c4368c1421
4 changed files with 101 additions and 0 deletions
View File
+3
View File
@@ -0,0 +1,3 @@
"""
第一步: 将百度经纬度转为高德经纬度(只需跑批一次即可)
"""
+80
View File
@@ -0,0 +1,80 @@
"""
通过居委围栏更新隶属关系
"""
from goojia_utils import DIDATA_DEV_ELASTIC_CLIENT, areautils
import pandas as pd
from elasticsearch import helpers
import logging
log = logging.getLogger(__name__)
def sync(province):
log.info(f"开始同步{province}企业数据的居委村委数据")
ids = DIDATA_DEV_ELASTIC_CLIENT.scan_by_sql_to_data_frame(f"select company_name from didata_company where province='{province}'")
company_ids = ids['_id'].to_list()
shape_df = DIDATA_DEV_ELASTIC_CLIENT.scan_by_sql_to_data_frame(f"select city_name, area_name, street_name, name, shape from community where province_name='{province}' ")
if len(shape_df) == 0:
log.info(f"{province}无围栏数据")
actions = list()
count = 0
errors = list()
for index, item in shape_df[shape_df['shape'].notna()].iterrows():
ret = DIDATA_DEV_ELASTIC_CLIENT.fetch_all_geo_points('didata_enterprise', item['shape']['coordinates'][0], field='gaode_location')
for info in ret:
source = info['_source']
if 'district' in source and (source['district'] != item['area_name']):
errors.append(source)
continue
if ('registered_address' not in source) or ('company_name' not in source):
continue
doc_info = {'province': source['province'], 'city': source['city'], 'district': source['district'],
'company_name': source['company_name'], 'registered_address': source['registered_address'],
'location': source['gaode_location'], 'township': item['street_name'],
'committee': item['name']}
if info['_id'] not in company_ids:
action = {
"_op_type": "index",
"_index": "didata_company",
"_id": info['_id'],
"_source": doc_info,
"_retry_on_conflict": 5
}
else:
action = {
"_op_type": "update",
"_index": "didata_company",
"_id": info['_id'],
"doc": doc_info,
"_retry_on_conflict": 5
}
actions.append(action)
if len(actions) == 10000:
for success, info in helpers.parallel_bulk(
DIDATA_DEV_ELASTIC_CLIENT.elastic_client,
actions,
thread_count=5, # 并行线程数
chunk_size=1000 # 批量大小
):
if not success:
print(f"索引失败: {info}")
count += len(actions)
print("更新:" + str(count))
actions.clear()
#pd.DataFrame(errors)[['registered_address', 'company_name', 'province', 'city', 'district', 'gaode_location']].to_csv(f'{province}经纬度错误地址.csv', index=False)
if len(actions) > 0:
for success, info in helpers.parallel_bulk(
DIDATA_DEV_ELASTIC_CLIENT.elastic_client,
actions,
thread_count=5, # 并行线程数
chunk_size=1000 # 批量大小
):
if not success:
print(f"索引失败: {info}")
count += len(actions)
print("更新:" + str(count))
actions.clear()
#pd.DataFrame(errors)[['registered_address', 'company_name', 'province', 'city', 'district', 'gaode_location']].to_csv(f'{province}经纬度错误地址.csv', index=False)
log.info(f"完成同步{province}企业数据的居委村委数据")
for province in areautils.get_all_province():
sync(province)
+18
View File
@@ -0,0 +1,18 @@
from datetime import timedelta
from airflow.operators.bash import BashOperator
import pendulum
from airflow import DAG
with DAG(
dag_id="didata_company",
schedule_interval=None,
start_date=pendulum.datetime(2025, 6, 15, tz="Asia/Shanghai"),
catchup=True,
description="企业涉农处理"
) as dag:
t1 = BashOperator(
task_id='comm',
bash_command='python /opt/airflow/dags/repo/dags/company/company/comm.py',
)
t1