将 Flask 请求/应用上下文复制到另一个进程

Question

简短版

我该如何序列化 Flask 应用程序或请求上下文，或者是该上下文的子集（即任何可以成功序列化的内容），以便我可以从另一个进程而不是线程中访问该上下文？

详细版

我有一些需要访问 Flask 请求上下文或应用上下文的函数，我想在后台运行这些函数。

Flask 内置了 @copy_current_request_context 装饰器，用于将函数包装在请求上下文的副本中，以便您可以在不同的线程中运行它：

from threading import Thread
from flask import Flask, request, copy_current_request_context

app = Flask(__name__)

@app.route('/')
def index():
    request.foo = 'bar'
    @copy_current_request_context
    def baz():
        print(request.foo)
    thr = Thread(target=baz)
    thr.start()
    return 'ok'

尽管 Flask 没有提供内置的装饰器来复制应用程序上下文，但它提供了相应的机制 - 在 Access flask.g inside greenlet 中描述了一种解决方案。

我尝试过的

首先，我希望使用 Pickle 来解决。这是由 concurrent.futures.ProcessPoolExecutor 使用的内容。不幸的是，由于应用程序上下文中存在线程锁对象，因此 Pickle 失败了:

>>> with app.test_request_context('/'):
...     pickle.dumps(appctx)
...
Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
TypeError: can't pickle _thread.lock objects

它也无法复制装饰/包装函数：

>>> with app.test_request_context('/'):
...     bar = copy_current_request_context(pow)
...     pickle.dumps(bar)
...
Traceback (most recent call last):
  File "<stdin>", line 3, in <module>
_pickle.PicklingError: Can't pickle <function pow at 0x110bee8c8>: it's not the same object as builtins.pow

接下来我尝试了dill。它通过了以上两个测试，但它无法序列化可能出现在上下文中的许多其他内容。特别是应用程序上下文容易出现扩展链接自身的情况：SQLAlchemy就是一个很好的例子。当你使用SQLAlchemy并尝试序列化你的应用程序上下文时，会发生以下情况：

>>> from flask_sqlalchemy import SQLAlchemy
>>> db = SQLAlchemy(app)
>>> with app.test_request_context('/'):
...     from flask.globals import _app_ctx_stack
...     appctx = _app_ctx_stack.top
...     dill.dumps(appctx)
...
Traceback (most recent call last):
  File "<stdin>", line 4, in <module>
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 294, in dumps
    dump(obj, file, protocol, byref, fmode, recurse)#, strictio)
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 287, in dump
    pik.dump(obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 437, in dump
    self.save(obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 549, in save
    self.save_reduce(obj=obj, *rv)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 662, in save_reduce
    save(state)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 902, in save_module_dict
    StockPickler.save_dict(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 856, in save_dict
    self._batch_setitems(obj.items())
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 882, in _batch_setitems
    save(v)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 549, in save
    self.save_reduce(obj=obj, *rv)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 662, in save_reduce
    save(state)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 902, in save_module_dict
    StockPickler.save_dict(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 856, in save_dict
    self._batch_setitems(obj.items())
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 882, in _batch_setitems
    save(v)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 816, in save_list
    self._batch_appends(obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 843, in _batch_appends
    save(tmp[0])
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 1386, in save_function
    obj.__dict__), obj=obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 638, in save_reduce
    save(args)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 786, in save_tuple
    save(element)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 771, in save_tuple
    save(element)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 1129, in save_cell
    pickler.save_reduce(_create_cell, (f,), obj=obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 638, in save_reduce
    save(args)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 771, in save_tuple
    save(element)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 549, in save
    self.save_reduce(obj=obj, *rv)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 662, in save_reduce
    save(state)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 902, in save_module_dict
    StockPickler.save_dict(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 856, in save_dict
    self._batch_setitems(obj.items())
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 882, in _batch_setitems
    save(v)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 549, in save
    self.save_reduce(obj=obj, *rv)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 662, in save_reduce
    save(state)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 902, in save_module_dict
    StockPickler.save_dict(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 856, in save_dict
    self._batch_setitems(obj.items())
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 882, in _batch_setitems
    save(v)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 549, in save
    self.save_reduce(obj=obj, *rv)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 662, in save_reduce
    save(state)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 902, in save_module_dict
    StockPickler.save_dict(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 856, in save_dict
    self._batch_setitems(obj.items())
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 882, in _batch_setitems
    save(v)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 504, in save
    f(self, obj) # Call unbound method with explicit self
  File "/Users/dchevell/Development/python3/sandbox/env/lib/python3.7/site-packages/dill/_dill.py", line 1330, in save_type
    StockPickler.save_global(pickler, obj)
  File "/usr/local/Cellar/python/3.7.3/Frameworks/Python.framework/Versions/3.7/lib/python3.7/pickle.py", line 957, in save_global
    (obj, module_name, name)) from None
_pickle.PicklingError: Can't pickle <class 'sqlalchemy.orm.session.SignallingSession'>: it's not found as sqlalchemy.orm.session.SignallingSession

所以：有没有什么方法可以解决这个问题？我想到的一个想法是在“即时”分叉新进程 - 即在我想要运行后台任务之前立即进行 - 可能意味着我根本不需要复制上下文，只要分叉发生在设置上下文之后，它应该已经拥有它了。然而，我不知道如何在ProcessPoolExecutor中使其工作，在那里进程是长期存在的池。我唯一的想法是在每个任务结束时强制工作进程关闭，但我相当确定这只会导致破损的池。

- daveruinseverything

---

这是一个非常巧妙的问题，大力点赞。你有所进展吗？我试图在 Flask 中使用带有 OpenCV 的 futures，但 futures 和 OpenCV 不相容，这促使我去尝试 loky，但又引发了当前的问题。 - jtlz2

Answer 1

不确定这对您是否合适，但有可能从其他进程中使用Flask应用程序上下文。这需要您使用应用程序工厂模式。

# your_project/app.py

def create_app(config_file):
    # Load the config file somehow
    # Create your app instance
    app = Flask(__name__)
    # Initialize your extensions
    db.init_app(app)
    # Mount routes, etc
    app.register_blueprint(...)
    # Return your app
    return app

# your_project/wsgi.py
from .app import create_app

app = create_app("./config/local.cfg")

你需要定义一个应用工厂，并在 wsgi.py 或其他方便的位置使用它来创建你的应用作为入口点，然后你可以从运行在另一个进程中的后台脚本中访问该应用实例。请注意，保留 HTML 标签。

# your_project/background_script.py
from .wsgi import app

with app.app_context():
    # Do some stuff

顺便说一句，这也是使用Flask与Celery的一种方式，例如在这个问题中描述的那样。 希望这可以帮助你。

- Icebreaker454