sqlalchemy 中 automap的使用

问题

1. 有时候 我们可能希望映射数据库中已经有的表，怎么办呢？

sqlalchemy 给我们提供了这样的一个方法，方便我们映射 数据库中的已经存在的表， 用法 也比较简单 ，automap_base
这个对象，就可以完成我们想映射表的需求了。

• 来看一个小例子

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@Time    : 2018/11/23 09:45
@File    : myautomap.py
@Author  : [email protected]
"""

from sqlalchemy.ext.automap import automap_base
from sqlalchemy.orm import Session, sessionmaker
from sqlalchemy import create_engine
from sqlalchemy.schema import Table
from datetime import datetime

from secure import XINYONGFEI_BI_URL


def to_dict(self):
    return {c.name: getattr(self, c.name, None) for c in self.__table__.columns}


Base = automap_base()

# engine, suppose it has two tables 'Bank' and 'BidOrder' set up
engine = create_engine(XINYONGFEI_BI_URL, pool_size=10, pool_recycle=7200,
                       pool_pre_ping=True, encoding='utf-8')

# reflect the tables
Base.prepare(engine, reflect=True)

# mapped classes are now created with names by default
# matching that of the table name.
# notice 注意这里 Bank 对应数据库中的表
Bank = Base.classes.Bank

session_factory = sessionmaker(bind=engine)
session = session_factory()

# 查询数据
query = session.query(Bank).filter_by(name='frank')
print(query)

m = query.first()

mydict = to_dict(m)

for k, v in mydict.items():
    print(k, v)

if __name__ == '__main__':
    data = {'id': 22, 'name': 'frank2', 'short_name': 'fk2', 'code': '10000', 'limit_daily': 3000,
            'limit_single': 120000, 'status': 'valid', 'preset_id': 0, 'created_user': 2,
            'created_time': datetime(2018, 9, 14, 15, 27, 57),
            'updated_time': datetime(2018, 11, 22, 15, 29, 32)
            }

    bank = Bank(**data)
    # 添加一条数据
    session.add(bank)

    session.commit()

XINYONGFEI_BI_URL 这个是接数据库的URL

具体可以参考官方给的链接：
database-urls

XINYONGFEI_BI_URL = ‘mysql+pymysql://xxxx:xxxxxxxx#@xxxxx:xxxxxxx/xinyongfei_db?charset=utf8’

结果如下：

SELECT `Bank`.id AS `Bank_id`, `Bank`.name AS `Bank_name`, `Bank`.short_name AS `Bank_short_name`, `Bank`.code AS `Bank_code`, `Bank`.limit_daily AS `Bank_limit_daily`, `Bank`.limit_single AS `Bank_limit_single`, `Bank`.status AS `Bank_status`, `Bank`.preset_id AS `Bank_preset_id`, `Bank`.created_user AS `Bank_created_user`, `Bank`.created_time AS `Bank_created_time`, `Bank`.updated_time AS `Bank_updated_time` 
FROM `Bank` 
WHERE `Bank`.name = %(name_1)s
id 10
name frank
short_name fd
code fsa
limit_daily 3000
limit_single 120000
status valid
preset_id 0
created_user 2
created_time 2018-09-14 15:27:57
updated_time 2018-11-22 15:29:32

这里会把 所有的表映射出来，engine 里面所有已经存在的表，都可以通过这种方式映射出来。

2.有的时候 我们可能只需要 映射部分表，也是有这个需求吧。
sqlalchemy 已经给我们提供了这样的接口

我们可以 自己生成一个 meta 对象 ，之后把这个 传给 auto_base() 就可以自定义 ，需要反射的表了。

• 来看下一个例子

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@Time    : 2018/11/23 19:45
@File    : myautomap.py
@Author  : [email protected]
"""

from sqlalchemy.ext.automap import automap_base
from sqlalchemy.orm import Session, sessionmaker
from sqlalchemy import create_engine, MetaData

from secure import XINYONGFEI_BI_URL


def to_dict(self):
    return {c.name: getattr(self, c.name, None) for c in self.__table__.columns}


metadata = MetaData()

# engine, suppose it has two tables 'Bank' and 'BidOrder' set up
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()

metadata.reflect(engine, only=['BidOrder', 'Book', 'TUser'])

Base = automap_base(metadata=metadata)

# calling prepare() just sets up mapped classes and relationships.
Base.prepare()

# mapped classes are now created with names by default
# matching that of the table name.
TUser = Base.classes.TUser

Book = Base.classes.Book

# 这里 没有Bank
# Bank = Base.classes.Bank
# print(Bank)

print(TUser)
print(Book)
print(type(TUser))
print(type(Book))

if __name__ == '__main__':
    pass

结果如下：

Connected to pydev debugger (build 182.4505.26)
<class 'sqlalchemy.ext.automap.TUser'>
<class 'sqlalchemy.ext.automap.Book'>
<class 'sqlalchemy.ext.declarative.api.DeclarativeMeta'>
<class 'sqlalchemy.ext.declarative.api.DeclarativeMeta'>

这个only 参数 就可以指定要映射的表
metadata.reflect(engine, only=[‘BidOrder’, ‘Book’, ‘TUser’])

如果 这个时候取bank 就会报错了。 Bank = Base.classes.Bank

参考文档

automap的使用