PL/Python - Python 过程语言自定义函数实现指定内容名称提取

 函数定义

--药物名称提取
CREATE OR REPLACE FUNCTION "pygetdrugname"("text" text)
  RETURNS "text" AS $BODY$
	global text
	from sqlalchemy import create_engine
	import pandas as pd
	name='root'#mysql用户名
	pw=''#mysql密码
	host='127.0.0.1'#mysql连接地址
	port='3306'
	database='dev'#数据库

	# 用sqlalchemy构建数据库链接engine
	connect_info = 'mysql+pymysql://{}:{}@{}:{}/{}?charset=utf8'.format(name,pw,host,port,database)
	database_info = create_engine(connect_info)	
	data=pd.read_sql('SELECT DISTINCT me_name1 FROM be_bianma WHERE LENGTH(me_name1)>1 ORDER BY LENGTH(me_name1) DESC;',database_info)
	

	drug=[]
	for i in data['me_name1'].tolist():
			if i in text:
					drug.append(str(i))


	ret='/'.join(drug)
	if len(ret)==0:
		ret=None
	return ret

$BODY$
  LANGUAGE plpython3u;

函数使用

	-- 药品名称提取	
SELECT
	me_title,
	pygetdrugname(me_title)
FROM
	be_linchuangshiyan 
WHERE
	laiyuan = '中国临床试验注册中心'
AND
	pygetdrugname(me_title) IS NOT NULL;

 处理结果案例