Flask和SQLAlchemy以及MetaData对象

问题描述

这是我第一次使用这个环境。

我愿意使用的SQLAlchemy部分只是一个允许使用autoload = True的Table对象来查询数据库的部分。我这样做是因为我的表已经存在于数据库（MySQL服务器），并没有通过定义瓶模型创建。

我已经通过了所有的文档，我不似乎找不到答案。这里有一些代码：

$ $ p $ $ $ $ $ $
$ b metadata = None

def connect_db（）：
engine = create_engine（app.config ['DATABASE_URI']）
全局元数据
metadata =元数据（绑定=引擎）
返回engine.connect（）


@ app.before_request
def before_request（）：
g.db = connect_db （）


@ app.teardown_request
def teardown_request（例外）：
g.db.close（）

现在您可能想知道为什么我使用全局变量名称元数据。还有一些代码：

$ p $ @ app.route（'/ test /< int：id>'）
def test（test_result_id）：

testTable = Table（'test_table'，metadata，autoload = True）


正如你可以看到我需要该对象是全球性的，以便从一个函数内访问它。

另外我声明相同的变量testTable中的每个函数都需要它。我觉得这不是正确的做法。

感谢所有！

解决方案

你见过 class。

it's the first time i am using this environment.

The part of SQLAlchemy i am willing to use is just the one that allows me to query the database using Table objects with autoload = True. I am doing this as my tables already exist in the DB (mysql server) and were not created by defining flask models.

I have gone through all the documentation and i don't seem to find an answer. Here is some code:

app = Flask(__name__)
app.config.from_object(__name__)

metadata = None

def connect_db():
    engine = create_engine(app.config['DATABASE_URI'])
    global metadata
    metadata = MetaData(bind=engine)
    return engine.connect()


@app.before_request
def before_request():
    g.db = connect_db()


@app.teardown_request
def teardown_request(exception):
    g.db.close()

Now you could be wondering why i use that global var named metadata. Ok some more code:

@app.route('/test/<int:id>')
def test(test_result_id):

    testTable = Table('test_table', metadata , autoload=True)

As you can see i need that object to be global in order to access it from within a function.

Also I am declaring the same var testTable in each function that needs it. I have the feeling this is not the right approach. I coudn't find any best practice advice for a case like mine.

Thanks all!

解决方案

Have you seen this snippet in the SQLAlchemy docs?

Maybe this would work:

# This is fine as a global global
metadata = MetaData()

@app.before_first_request
def autoload_tables():
    meta.reflect(bind=g.db.bind)

@app.route('/')
def index():
    users_table = meta.tables['users']

That way your tables are reflected only once per process which is probably what you want. Note that your engine should be a global too, so you needn't create a new engine in @app.before_request - app creation is a more appropriate place.

If your case is very special you might need one engine per request, in which case you should consider the ThreadLocalMetaData class.

这篇关于Flask和SQLAlchemy以及MetaData对象的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！