Flask源码分析

软件发布|下载排行|最新软件

当前位置:首页IT学院IT技术

Flask源码分析

功夫小熊猫   2020-03-11 我要评论
首先,推荐一个便宜的云主机,感觉比阿里便宜,新用户一年68元,滴滴云 附上活动链接:https://i.didiyun.com/2d7Jy4Nzle8 Flask的最小应用程序如下: ```python from flask import Flask app = Flask(__name__) @app.route('/') def index(): return 'Hello World' if __name__ == '__main__': app.run() ``` #### app.run() Flask的类从flask包中引入后,直接制造app实例,使用app实例执行run方法。 所以先来看看Flask类中的__init___()和run()方法 ```python def __init__(  # 这里用的比较多的是,static_folder,static_url_path静态目录的指定,实例中都是默认参数 self, import_name, static_url_path=None, static_folder="static", static_host=None, host_matching=False, subdomain_matching=False,    template_folder="templates",    instance_path=None,    instance_relative_config=False,    root_path=None, ``` ```python def run(self, host=None, port=None, debug=None, load_dotenv=True, **options): try: run_simple(host, port, self, **options) finally: self._got_first_request = False ``` run方法中省略调试相关的代码,关键执行代码是run_simple方法,带入默认的host和port以及对象,host为127.0.0.1,port为5000。那么就来分析run_simple() #### run_simple() run_simple方法来自werkzeug中的serving.py。 ```python def run_simple( hostname, port, application ): def inner(): try: fd = int(os.environ["WERKZEUG_SERVER_FD"]) except (LookupError, ValueError): fd = None srv = make_server( hostname, port, application, threaded, processes, request_handler, passthrough_errors, ssl_context, fd=fd, ) if fd is None: log_startup(srv.socket) srv.serve_forever() inner() ``` run_simple()主要是执行了inner(),inner中初始化srv实例后执行srv.serve_forever()方法。先看make_server()方法,根据进程和线程使用不同的类进行初始化,且这些类都继承了BaseWSGIServer。 ```python def make_server(    host=None,  port=None,  app=None,  threaded=False, processes=1,  request_handler=None, passthrough_errors=False, ssl_context=None,    fd=None,):  if threaded and processes > 1: raise ValueError("cannot have a multithreaded and multi process server.") elif threaded:        return ThreadedWSGIServer(省略 )    elif processes > 1: return ForkingWSGIServer(省略 ) else:  return BaseWSGIServer(省略 ) ``` 这里以BaseWSGIServer为例进行分析。BaseWSGIServer继承HTTPServer,这里带入的参数除了host,port还有handler,sWSGIRequestHandler,WSGIRequestHandler继承BaseHTTPRequestHandler。主要是为了实现http的数据请求和发送。 ```python class BaseWSGIServer(HTTPServer, object):    """Simple single-threaded, single-process WSGI server."""    multithread = False    multiprocess = False def __init__(    self,  host, port, app, handler=None,  passthrough_errors=False, ssl_context=None,  fd=None,): if handler is None:  handler = WSGIRequestHandler ``` 先看一下执行函数,serve_forever,执行的是 HTTPServer.serve_forever() ```python def serve_forever(self):  self.shutdown_signal = False try:        HTTPServer.serve_forever(self) except KeyboardInterrupt:  pass finally: self.server_close() ``` 为了更好的说明,把WSGIRequestHandler和BaseWSGIServer继承过来,单独运行获得http的响应。 ```python from werkzeug.serving import WSGIRequestHandler,BaseWSGIServer class New_Handler(WSGIRequestHandler): page = '''\

hello

''' def do_GET(self): current_path = self.path # self.send_error(415,'your url is %s' % current_path) self.send_response(200) self.send_header("Content-Type", "text/html") self.send_header("Content-Length", 100) self.end_headers() self.wfile.write(self.page.encode('utf-8')) tcp_server = BaseWSGIServer(host="127.0.0.1",port=5000,app=None,handler=New_Handler) if __name__ == "__main__": tcp_server.serve_forever() ``` ![](https://img2020.cnblogs.com/blog/1639143/202003/1639143-20200308151037279-988954675.jpg) 既然http的请求响应入口已经找到了,那么就来看看请求路由的处理,所以要研究装饰器@app.route('/'),来看看Flask中路由处理思路。 装饰器如下: ```python def route(self, rule, **options): def decorator(f): endpoint = options.pop('endpoint', None) self.add_url_rule(rule, endpoint, f, **options) return f return decorator # 路由装饰器的使用 @app.route('/') def hello_world(): return "hello_world" ``` 以简单根目录请求为例,通过装饰器可以看出,首先执行self.add_url_rule(rule, endpoint, f, **options),然后再执行实例方法hello_world()。

Copyright 2022 版权所有 软件发布 访问手机版

声明:所有软件和文章来自软件开发商或者作者 如有异议 请与本站联系 联系我们