Flask

参考

  1. http://flask.pocoo.org/docs/1.0/advanced_foreword/#thread-locals-in-flask
  2. https://zhuanlan.zhihu.com/p/33732859
  3. https://www.zhihu.com/question/25033592/answer/34449852
  4. https://www.zhihu.com/question/269905592/answer/364928400
  5. http://flask.pocoo.org/docs/1.0/appcontext/#purpose-of-the-context
  6. http://flask.pocoo.org/docs/1.0/reqcontext/
  7. http://flask.pocoo.org/docs/1.0/appcontext/#storing-data
    其实看5和6,7就够了,因为是权威的文档

Flask的上下文管理很重要,留坑

1-5为预备知识,从6正式开始

现象:

Flask 从客户端收到请求时,要让视图函数能访问一些对象,这样才能处理请求。请求对象就是一个很好的例子,它封装了客户端发送的HTTP 请求。要想让视图函数能够访问请求对象,一个显而易见的方式是将其作为参数传入视图函数,不过这会导致程序中的每个视图函数都增加一个参数。除了访问请求对象,如果视图函数在处理请求时还要访问其他对象,情况会变得更糟。为了避免大量可有可无的参数把视图函数弄得一团糟,Flask 使用上下文临时把某些对象变为全局可访问

PS: 访问其他对象,例如请求对象

django/tornado是通过传参数形式,Flask是通过上下文管理。

问题:flask的request和session设置方式比较新颖,如果没有这种方式,那么就只能通过参数的传递。flask是如何做的呢?

虽然threading.local与这个实现没关系,不过要先了解threading.local对象才能理解Flask的上下文管

1. threading.local对象,用于为每个线程开辟一块空间来保存它独有的值。

Flask的上下文管理借助了这个思想。Thread-local data is data whose values are thread specific
本地线程,保证即使是多个线程,自己的值也是互相隔离。

import threading


# class Foo(object):
#     def __init__(self):
#         self.name = 0
#
# local_values = Foo()

local_values = threading.local()


def func(num):
    local_values.name = num
    import time
    time.sleep(1)
    print(local_values.name, threading.current_thread().name)


for i in range(20):
    th = threading.Thread(target=func, args=(i,), name='线程%s' % i)
    th.start()

如果不用threading.local(),那么

因为修改的值始终保存在同一个对象里面,sleep一秒后,全部线程执行完毕,值变成了最后的值19。

2. 单进程单线程多协程中,threading.local会出问题,因为协程数据都存在同一个线程里。

解决方法为:
单进程单线程多协程中,程序不再支持协程,就可以使用threading.local对象。
单进程单线程多协程中,程序想支持协程,那么自定义类似threading.local对象。如下所示


> 自定义类似threading.local对象,支持协程(Python中greenlet就是协程),原理是一个标识identification对应一个字典(存储数据的地方)
"""
{
   identification:{k:v},一个标识对应一个字典,threading.local工作原理类似这样,看源码可知,
    class _localimpl:
    """A class managing thread-local dicts"""
}



"""
import threading
try:
    #获取标识identification
    from greenlet import getcurrent as get_ident # 协程
except ImportError:
    try:
        from thread import get_ident
    except ImportError:
        from _thread import get_ident # 线程

# 模拟threading.Local()且支持协程
class Local(object):
    def __init__(self):
        self.storage = {}
        self.get_ident = get_ident

    def set(self,k,v):
        ident = self.get_ident()
        origin = self.storage.get(ident)
        if not origin:
            origin = {k:v}
        else:
            origin[k] = v
        self.storage[ident] = origin

    def get(self,k):
        ident = self.get_ident()
        origin = self.storage.get(ident)
        if not origin:
            return None
        return origin.get(k,None)

local_values = Local()


def task(num):
    local_values.set('name',num)
    import time
    time.sleep(1)
    print(local_values.get('name'), threading.current_thread().name)


for i in range(20):
    th = threading.Thread(target=task, args=(i,),name='线程%s' % i)
    th.start()

3. 注意__setattr__的坑,为往后的点作铺垫.

http://www.cnblogs.com/allen2333/p/9019660.html
改正后,可以用__setattr__,getattr

import threading
try:
    from greenlet import getcurrent as get_ident # 协程
except ImportError:
    try:
        from thread import get_ident
    except ImportError:
        from _thread import get_ident # 线程


class Local(object):

    def __init__(self):
        object.__setattr__(self, '__storage__', {})
        object.__setattr__(self, '__ident_func__', get_ident)


    def __getattr__(self, name):
        try:
            return self.__storage__[self.__ident_func__()][name]
        except KeyError:
            raise AttributeError(name)

    def __setattr__(self, name, value):
        ident = self.__ident_func__()
        storage = self.__storage__
        try:
            storage[ident][name] = value
        except KeyError:
            storage[ident] = {name: value}

    def __delattr__(self, name):
        try:
            del self.__storage__[self.__ident_func__()][name]
        except KeyError:
            raise AttributeError(name)


local_values = Local()


def task(num):
    local_values.name = num
    import time
    time.sleep(1)
    print(local_values.name, threading.current_thread().name)


for i in range(20):
    th = threading.Thread(target=task, args=(i,),name='线程%s' % i)
    th.start()

4. 上下文原理

#!/usr/bin/env python
# -*- coding:utf-8 -*-
from functools import partial
from flask.globals import LocalStack, LocalProxy
 
ls = LocalStack()
 
 
class RequestContext(object):
    def __init__(self, environ):
        self.request = environ
 
 
def _lookup_req_object(name):
    top = ls.top
    if top is None:
        raise RuntimeError(ls)
    return getattr(top, name)
 
 
session = LocalProxy(partial(_lookup_req_object, 'request'))
 
ls.push(RequestContext('c1')) # 当请求进来时,放入
print(session) # 视图函数使用
print(session) # 视图函数使用
ls.pop() # 请求结束pop
 
 
ls.push(RequestContext('c2'))
print(session)
 
ls.push(RequestContext('c3'))
print(session)

5. Flask内部实现

#!/usr/bin/env python
# -*- coding:utf-8 -*-
 
from greenlet import getcurrent as get_ident
 
 
def release_local(local):
    local.__release_local__()
 
 
class Local(object):
    __slots__ = ('__storage__', '__ident_func__')
 
    def __init__(self):
        # self.__storage__ = {}
        # self.__ident_func__ = get_ident
        object.__setattr__(self, '__storage__', {})
        object.__setattr__(self, '__ident_func__', get_ident)
 
    def __release_local__(self):
        self.__storage__.pop(self.__ident_func__(), None)
 
    def __getattr__(self, name):
        try:
            return self.__storage__[self.__ident_func__()][name]
        except KeyError:
            raise AttributeError(name)
 
    def __setattr__(self, name, value):
        ident = self.__ident_func__()
        storage = self.__storage__
        try:
            storage[ident][name] = value
        except KeyError:
            storage[ident] = {name: value}
 
    def __delattr__(self, name):
        try:
            del self.__storage__[self.__ident_func__()][name]
        except KeyError:
            raise AttributeError(name)
 
 
class LocalStack(object):
    def __init__(self):
        self._local = Local()
 
    def __release_local__(self):
        self._local.__release_local__()
 
    def push(self, obj):
        """Pushes a new item to the stack"""
        rv = getattr(self._local, 'stack', None)
        if rv is None:
            self._local.stack = rv = []
        rv.append(obj)
        return rv
 
    def pop(self):
        """Removes the topmost item from the stack, will return the
        old value or `None` if the stack was already empty.
        """
        stack = getattr(self._local, 'stack', None)
        if stack is None:
            return None
        elif len(stack) == 1:
            release_local(self._local)
            return stack[-1]
        else:
            return stack.pop()
 
    @property
    def top(self):
        """The topmost item on the stack.  If the stack is empty,
        `None` is returned.
        """
        try:
            return self._local.stack[-1]
        except (AttributeError, IndexError):
            return None
 
 
stc = LocalStack()
 
stc.push(123)
v = stc.pop()
 
print(v)

6. 大概流程(重要,去看源码理解)

6.1 借鉴threading.Local(线程临时存对象的地方),Flask自定义Local对象(Local, LocalStack, LocalProxy)。Local类似threading.Local,是临时存对象的地方,比后者多出协程支持。

上下文管理:
类似threading.local ,Flask自己实现 Local类(临时存对象的地方),其中创建了一个字典,{用greenlet协程做唯一标识:存数据} 保证数据隔离
请求进来时:

  • 请求相关所有数据封装到了RequestContext中。ctx = 封装RequestContext(request,session)
  • 再将ctx = RequestContext对象添加到Local中(通过LocalStack将对象添加到Local对象中)
    执行view function时,调用request:
    -调用此类方法: request.method、print(request)、request + xxx。
from flask import request中查看request源码
_request_ctx_stack = LocalStack()
request = LocalProxy(partial(_lookup_req_object, 'request'))
  • request.method会执行LocalProxy中对应的方法(getattr) --> 对应的方法执行_get_current_object --> 偏函数(._lookup_req_object(), request)--> 通过LocalStack从Local中 --> 获取top = RequestContext --> ctx.request --> getattr --> ctx.request.method
    请求终止时:
  • ctx.auto_pop()
  • 通过LocalStack的pop方法,ctx从Local对象中移除。

session

  • 过程一样,只是最后是ctx.session,相比较于ctx.request

7. 应用上下文和请求上下文什么关系?

from flask import Flask, request, g, globals
app = Flask('_name_'), app._call_, ctx.push()
在这两段代码进去看源码

globals.py

_request_ctx_stack = LocalStack()
_app_ctx_stack = LocalStack()

在app.__call__进去,ctx = self.request_context(environ),再从ctx.push()进去

top = _request_ctx_stack.top

...
        top = _request_ctx_stack.top
        if top is not None and top.preserved:
            top.pop(top._preserved_exc)

        # Before we push the request context we have to ensure that there
        # is an application context.
        app_ctx = _app_ctx_stack.top
        if app_ctx is None or app_ctx.app != self.app:
            app_ctx = self.app.app_context()
            app_ctx.push()
            self._implicit_app_ctx_stack.append(app_ctx)
        else:
            self._implicit_app_ctx_stack.append(None)

        if hasattr(sys, 'exc_clear'):
            sys.exc_clear()

        _request_ctx_stack.push(self)

所以应用上下文在每个请求上下文被push到__request_ctx_stack之前,自己被push到_app_ctx_stack中

8. request和g的区别

g,每个请求周期都会创建一个用于在请求周期中传递值的一个容器。
request(在请求上下文中,LocalStack -- > 存储在Local中)和g都是临时容器,用来存储stuff。但是在程序中为了不修改request(封装了请求的参数),所以用g(在应用上下文中初始化,LocalStack -- >存储在Local中)来代替request存储数据、对象。
_request_ctx_stack = LocalStack(),_app_ctx_stack = LocalStack(),两个是不同的LocalStack --> 不同的Local。PS:Local是临时存储stuff的地方。

from flask import Flask, request, g, globals
app = Flask('name'), app.call, ctx.push()
在这两段代码进去看源码

_request_ctx_stack.Local() = {
    唯一标识ident: {
        "stack":[request_ctx, ]
    }
}

_app_ctx_stack.Local() = {
    唯一标识ident: {
        "stack":[app_ctx, ]
    }
}

request_ctx = RequestContext(),app_ctx = AppContext()

9. from flask import request,session,g,current_app

参考6,print(request,session,g,current_app),都会执行相应LocalProxy对象的 str,也就是都是从LocalProxy出发
唯一不同的是传递的偏函数的参数不一样


request = LocalProxy(partial(_lookup_req_object, 'request'))
session = LocalProxy(partial(_lookup_req_object, 'session'))
current_app = LocalProxy(_find_app)
g = LocalProxy(partial(_lookup_app_object, 'g'))

10. 一个程序多线程有多少个Local? Web访问多app应用时,上下文管理是如何实现?

还是原来的流程,wsgi --> push到Local。都是只有两个Local,通过唯一标识存储context.

_request_ctx_stack.Local() = {
    唯一标识ident: {
        "stack":[request_ctx, ]
    }

    唯一标识ident: {
        "stack":[request_ctx, ]
    }

....
   
}

_app_ctx_stack.Local() = {
    唯一标识ident: {
        "stack":[app_ctx, ]
    }

    唯一标识ident: {
        "stack":[app_ctx, ]
    }
}

11. Flask的Local中保存数据时,使用列表创建出来的栈。为什么用栈?

"stack":[app_ctx, ]

如果写web程序,web运行环境;栈中永远保存1条数据(可以不用栈)。
写脚本获取app信息时,可能存在app上下文嵌套关系,栈可能有多条数据。
意味着栈是备用的!
例如写测试脚本时,获取app1, app2的信息,测试一下数据库是否正确连接
总结来说,为了部落!(多应用)

from flask import Flask, current_app, globals, _app_ctx_stack

app1 = Flask('app01')
app1.debug = False  # 用户/密码/邮箱
# app_ctx = AppContext(self):
# app_ctx.app
# app_ctx.g

#RuntimeError: Working outside of application context.
#print(current_app.config['DEBUG'])

app2 = Flask('app02')
app2.debug = True  # 用户/密码/邮箱
# app_ctx = AppContext(self):
# app_ctx.app
# app_ctx.g

# 写脚本的时候有上下文嵌套的写法。但是写网站的时候没可能有两个或以上app嵌套,堆栈里的stack永远放的是一个。"stack":[app_ctx, ]
with app1.app_context():  # __enter__方法 -> push -> app_ctx添加到_app_ctx_stack.local
    # {<greenlet.greenlet object at 0x00000000036E2340>: {'stack': [<flask.ctx.AppContext object at 0x00000000037CA438>]}}
    print(_app_ctx_stack._local.__storage__)
    print(current_app)
    print(current_app.config['DEBUG'])

    # 写在里面,Local:{'stack' : [ctx1, ctx2]}有两个了!stack.top --> stack[-1] --> 还是拿栈顶的!
    with app2.app_context():
        # {<greenlet.greenlet object at 0x00000000036E2340>: {'stack': [<flask.ctx.AppContext object at 0x00000000037CA438> ]}}
        print(_app_ctx_stack._local.__storage__)
        print(current_app)
        print(current_app.config['DEBUG'])

"""    
with app2.app_context():
    # {<greenlet.greenlet object at 0x00000000036E2340>: {'stack': [<flask.ctx.AppContext object at 0x00000000037CA438> ]}}
    print(_app_ctx_stack._local.__storage__)
    print(current_app)
    print(current_app.config['DEBUG'])
"""



原文地址:https://www.cnblogs.com/allen2333/p/9019367.html