新增企业库处理流程

This commit is contained in:
沈健
2025-06-09 10:50:30 +08:00
parent 02e1aab542
commit 43f35fa63e
4 changed files with 229 additions and 3 deletions
+24 -1
View File
@@ -11,7 +11,30 @@ flag = False
for province in areautils.get_all_province():
print(f"开始处理{province}数据")
# 百度经纬度转高德经纬度
df = DIDATA_DEV_ELASTIC_CLIENT.scan_by_sql_to_data_frame(f"select location from didata_enterprise where province='{province}'")
df = DIDATA_DEV_ELASTIC_CLIENT.scan_to_data_frame(index='didata_enterprise', query={
"query": {
"bool": {
"must": [
{
"term": {
"province": {
"value": "" + province + ""
}
}
}
],
"must_not": [
{
"exists": {
"field": "gaode_location"
}
}
]
}
}
})
if len(df) == 0:
continue
def deal_location(item):
if pd.isna(item['location']):
return item
+66 -1
View File
@@ -1,3 +1,68 @@
from goojia_utils import DIDATA_DEV_ELASTIC_CLIENT, threads, geohashutils, areautils
import pandas as pd
from elasticsearch import helpers
import numpy as np
import re
import logging
log = logging.getLogger(__name__)
"""
通过开发区围栏更新企业开发区 TODO
"""
"""
def sync(province):
log.info(f"开始同步{province}企业数据的开发区数据")
# 先临时禁用索引刷新
DIDATA_DEV_ELASTIC_CLIENT.elastic_client.indices.put_settings(
index="didata_company",
body={"index": {"refresh_interval": "-1"}}
)
shape_df = DIDATA_DEV_ELASTIC_CLIENT.scan_by_sql_to_data_frame(f"select name, shape from development_zone where province='{province}'")
actions = list()
count = 0
for index, item in shape_df.iterrows():
if pd.isna(item['shape']):
continue
ret = DIDATA_DEV_ELASTIC_CLIENT.fetch_all_geo_points('didata_company', item['shape']['coordinates'][0],
field='location')
for info in ret:
doc_info = {'dev': item['name']}
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()
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()
log.info(f"完成同步{province}企业数据的开发区数据")
# 恢复索引刷新
DIDATA_DEV_ELASTIC_CLIENT.elastic_client.indices.put_settings(
index="didata_company",
body={"index": {"refresh_interval": "5s"}}
)
+130
View File
@@ -0,0 +1,130 @@
from goojia_utils import DIDATA_DEV_ELASTIC_CLIENT, threads, geohashutils, areautils
import pandas as pd
from elasticsearch import helpers
import numpy as np
import re
import logging
log = logging.getLogger(__name__)
"""
对于没有围栏结果的,通过match_phrase匹配出来的开发区一定正确,如果有围栏出来的开发区,则企业经纬度不正确
"""
dev_alias = {'福建永定工业园区': '永定区永定工业园', '福建漳州古雷港经济开发区': '古雷港经济开发区', '福建莆田湄洲湾北岸经济开发区': '湄洲湾北岸经济开发区',
'福建南平工业园区': '南平工业园区', '泉州台商投资区': '泉州市台商投资区', '福建洛江经济开发区': '洛江园区', '福建武平工业园区': '武平工业园',
'福建晋江经济开发区': '晋江市经济开发区', '漳州台商投资区': '龙池开发区', '莆田高新技术产业开发区': '莆田市高新技术',
'莆田高新技术产业开发区': '莆田市涵江区高新', '绥安工业开发区': '绥安工业园', '绥安工业开发区': '黄仓开发区', '绥安工业开发区': '黄仓工业园',
'绥安工业开发区': '旧镇工业园', '绥安工业开发区': '大南坂开发区', '绥安工业开发区': '大南坂工业开发区', '福建龙州工业园区': '龙州工业园',
'长汀经济开发区':'腾飞开发区', '福建福州金山工业园区': '金山投资区', '厦门象屿综合保税区': '厦门现代物流园', '福清融侨经济技术开发区': '融侨经济',
'福建洛江经济开发区': '万安开发区'}
dev_df = DIDATA_DEV_ELASTIC_CLIENT.scan_by_sql_to_data_frame("select province, city, district, name, alias from development_zone where province='福建省'")
# 不要去重,有可能存在别名, 如福建云霄常山经济开发区
#dev_df = dev_df[dev_df['name'] == '福建政和经济开发区']
actions = list()
for dev in dev_df.itertuples():
if dev.name in ['中国(福建)自由贸易试验区厦门片区', '中国(福建)自由贸易试验区福州片区', '中国(福建)自由贸易试验区平潭片区']:
# 中国(福建)自由贸易试验区 已包含,防止反复更新
continue
if dev.name in ['政和经济开发区']:
continue
count = 0
names = list()
names.append(dev.name)
if str(dev.alias) != 'nan':
names.extend(dev.alias)
#东侨经济技术开发区 福建省宁德市东侨经济开发区塔山村塔兴路25号
if '经济技术开发区' in dev.name:
names.append(re.sub('经济技术开发区', '经济开发区', dev.name))
# 福州经济技术开发区 福州开发区君竹路83号科技发展中心大楼第四层R472室(自贸试验区内)
names.append(re.sub('经济技术开发区', '开发区', dev.name))
# 长汀经济开发区 长汀县经济开发区工业新区河梁大山体
if '经济开发区' in dev.name:
if pd.notna(dev.city) and pd.notna(dev.district):
district_short = areautils.get_district_short(dev.city, dev.district)
if pd.notna(district_short):
names.append(re.sub(district_short, dev.district, dev.name))
# 漳州高新技术产业开发区 福建省漳州高新区九湖镇木棉村后壁山401-1号
if '高新技术产业开发区' in dev.name:
names.append(re.sub('高新技术产业开发区', '高新区', dev.name))
# 厦门火炬高技术产业开发区 厦门火炬高新区软件园三期溪西山尾路63号1303室-3单元之三
if '高技术产业开发区' in dev.name:
names.append(re.sub('高技术产业开发区', '高新区', dev.name))
#areautils.get_district_display()
# 福建诏安工业园区 福建省漳州市诏安工业园区丹诏大道15号
if pd.notna(dev.city):
province_short = areautils.get_province(dev.city)
new_name = re.sub(province_short, '', dev.name)
if pd.notna(dev.city) and pd.notna(dev.district):
city_short = areautils.get_city_short(dev.city)
district_short = areautils.get_district_short(dev.city, dev.district)
if pd.notna(city_short) and city_short in new_name:
names.append(new_name)
if pd.notna(district_short) and district_short in new_name:
names.append(new_name)
# 福建省漳州市长泰县经济开发区林墩办事处帝君路 福建长泰经济开发区
new_new_name = re.sub(district_short, district_short + '县', new_name)
names.append(new_new_name)
new_new_name = re.sub(district_short, district_short + '区', new_name)
names.append(new_new_name)
for key in dev_alias:
if dev.name == key:
names.append(dev_alias[key])
for name in names:
query = {
"query": {
"bool": {
"must_not": [
{
"term": {
"dev": {
"value": dev.name
}
}
},
{
"term": {
"dev_status": {
"value": 0
}
}
}
],
"must": [
{
"term": {
"province": {
"value": dev.province
}
}
},
{
"match_phrase": {
"registered_address": name
}
}
]
}
}
}
count_data = DIDATA_DEV_ELASTIC_CLIENT.elastic_client.count(index='didata_company',body=query)
if count_data['count'] > 0:
for doc in DIDATA_DEV_ELASTIC_CLIENT.scan(index='didata_company',query=query):
action = {
"_op_type": "update",
"_index": "didata_company",
"_id": doc['_id'],
"doc": {'dev': dev.name, 'dev_status': 0}
}
actions.append(action)
if len(actions) == 5000:
helpers.bulk(DIDATA_DEV_ELASTIC_CLIENT.elastic_client, actions)
count += len(actions)
print(f"已更新 {dev.name} 开发区" + str(count) + "记录")
actions.clear()
if len(actions) > 0:
helpers.bulk(DIDATA_DEV_ELASTIC_CLIENT.elastic_client, actions)
count += len(actions)
print(f"已更新 {dev.name} 开发区" + str(count) + "记录")
actions.clear()
+9 -1
View File
@@ -15,4 +15,12 @@ with DAG(
task_id='fill_commpany_comm',
bash_command='python /opt/airflow/dags/repo/dags/company/company/fill_company_comm.py',
)
t1
t2 = BashOperator(
task_id='fill_commpany_dev',
bash_command='python /opt/airflow/dags/repo/dags/company/company/fill_commpany_dev.py',
)
t3 = BashOperator(
task_id='mark_right_dev',
bash_command='python /opt/airflow/dags/repo/dags/company/company/mark_right_dev.py',
)
t1 >> t2 >> t3