版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/u010339879/article/details/84890642
sqlalchemy 中使用 sql 查询数据库
- 背景
有时候 ，我们希望通过原生 sql 来查询数据库，这个时候 应该怎么办呢？
sqlalchemy 已经给我们提供了这样的接口， 通过 text 就可以轻松实现了。
- 例1
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@Time : 2018/12/8 10:03
@File : query_data5.py
@Author : [email protected]
"""
from sqlalchemy.orm import sessionmaker
from sqlalchemy import create_engine, text, Integer, Float
from secure import XINYONGFEI_BI_URL
def get_session():
engine = create_engine(XINYONGFEI_BI_URL, pool_size=10, pool_recycle=7200,
pool_pre_ping=True, encoding='utf-8')
session_factory = sessionmaker(bind=engine)
session = session_factory()
return session
if __name__ == '__main__':
session = get_session()
sql = """\
select `id`, `name`, `mobile`
from `TUser` limit 5 ;
"""
print(f'sql:{sql}')
try:
resultproxy = session.execute(
text(sql)
)
except Exception as e:
print(e)
results = []
else:
results = resultproxy.fetchall()
print(results)
sql: select `id`, `name`, `mobile`
from `TUser` limit 5 ;
[(1.0, 'frank0', '123492130'), (2.0, 'frank1', '123492131'), (3.0, 'frank2', '123492132'), (4.0, 'frank3', '123492133'), (5.0, 'frank4', '123492134')]
这里没有参数 直接用 text 包装一下， 就可以执行了。
- 例子2
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@Time : 2018/11/26 11:15
@File : test_use_sql.py
@Author : [email protected]
site-packages/sqlalchemy/orm/session.py
"""
from sqlalchemy.ext.automap import automap_base
from sqlalchemy.orm import Session, sessionmaker
from sqlalchemy import create_engine, text
from secure import XINYONGFEI_BI_URL
def to_dict(data):
my_column = ["id", "order_number", "loan_set_id", "account_id", "amount", "success_time", "fail_time", "memo",
"order_memo", "order_status", "status", "created_time", "updated_time"]
return dict(zip(my_column, data))
def get_session():
engine = create_engine(XINYONGFEI_BI_URL, pool_size=10, pool_recycle=7200,
pool_pre_ping=True, encoding='utf-8')
session_factory = sessionmaker(bind=engine)
session = session_factory()
return session
def get_table(table_name):
"""
通过表名，映射数据库里面的表
:param table_name: str
:return:
"""
Base = automap_base()
engine = create_engine(XINYONGFEI_BI_URL, pool_size=10, pool_recycle=7200,
pool_pre_ping=True, encoding='utf-8')
Base.prepare(engine, reflect=True)
classes = Base.classes
try:
table = getattr(classes, table_name)
return table
except AttributeError:
raise AttributeError(f'not exist table_name:{table_name}')
if __name__ == '__main__':
pass
session = get_session()
sql = """
select `id`, `order_number`, `loan_set_id`, `account_id`, `amount`, `success_time`, `fail_time`, `memo`,
`order_memo`, `order_status`, `status`, `created_time`, `updated_time` from BidOrder
where account_id = :account_id and amount = :amount ;
"""
# print(f'sql:{sql}')
try:
resultproxy = session.execute(
text(sql), {"account_id": 100000, "amount": 500}
)
except Exception as e:
print(e)
results = []
else:
results = resultproxy.fetchall()
for result in results:
print(to_dict(result))
"""
传入参数 :param
这里传入两个参数
:account_id ,
:amount
sql 要通过 text 包装一下， 这样就可以 通过 session 对象 的 execute 方法，向数据库里面查询，写入数据了。
注意参数是一个字典， key 就是 sql 中定义的参数
session.execute(
text(sql), {"account_id": 100000, "amount": 500}
)
"""
这里执行sql 如下：
SELECT `id AS `, `order_number AS `, `loan_set_id AS `, `account_id AS `, `amount AS `
, `success_time AS `, `fail_time AS `, `memo AS `, `order_memo AS `, `order_status AS `
, `status AS `, `created_time AS `, `updated_time AS `
FROM BidOrder
WHERE account_id = 100000
AND amount = 500;
总结
本文主要讲了如何使用sqlalchemy 执行原生的sql, 主要是通过text 这个 对象来实现的。 当然还可以直接绑定参数，定制column 更多可以参考官网给的例子，例子还是非常丰富的.
参考文档
text 的使用 https://docs.sqlalchemy.org/en/latest/orm/tutorial.html#using-textual-sql
sql Expressions https://docs.sqlalchemy.org/en/latest/core/sqlelement.html#sqlalchemy.sql.expression.text
session 的基本操作，查询代码，已经写得比较清楚了。
site-packages/sqlalchemy/orm/session.py