欧美bbbwbbbw肥妇,免费乱码人妻系列日韩,一级黄片

python Tornado事件循環(huán)示例源碼解析

 更新時(shí)間:2023年09月08日 09:32:13   作者:BruceChen7  
這篇文章主要為大家介紹了python Tornado事件循環(huán)示例源碼解析,有需要的朋友可以借鑒參考下,希望能夠有所幫助,祝大家多多進(jìn)步,早日升職加薪

hello world

#!/usr/bin/env python
import tornado.httpserver
import tornado.ioloop
import tornado.options
import tornado.web
from tornado.options import define, options
define("port", default=8888, help="run on the given port", type=int)
class MainHandler(tornado.web.RequestHandler):
    def get(self):
        self.write("Hello, world")
def main():
    tornado.options.parse_command_line()
    application = tornado.web.Application([
        (r"/", MainHandler),
    ])
    http_server = tornado.httpserver.HTTPServer(application)
    http_server.listen(options.port)
    tornado.ioloop.IOLoop.instance().start()
if __name__ == "__main__":
    main()

tornado提供了高效的異步機(jī)制,我們先不管Application實(shí)例化過程,以及http_server創(chuàng)建socket、bind、listen的過程,直接調(diào)IOLoop.instance().start進(jìn)行源碼分析。

IOLoop.instance()

    @classmethod
    def instance(cls):
        """Returns a global IOLoop instance.
        Most single-threaded applications have a single, global IOLoop.
        Use this method instead of passing around IOLoop instances
        throughout your code.
        A common pattern for classes that depend on IOLoops is to use
        a default argument to enable programs with multiple IOLoops
        but not require the argument for simpler applications:
            class MyClass(object):
                def __init__(self, io_loop=None):
                    self.io_loop = io_loop or IOLoop.instance()
        """
        if not hasattr(cls, "_instance"):
            cls._instance = cls()
        return cls._instance

顯然是一個(gè)單例模式,注意tornado中的注釋,大多數(shù)單線程只能有一個(gè)ioloop。

start()

這個(gè)函數(shù)將開始事件循環(huán)。

def start():
     """
     Starts the I/O loop.
     The loop will run until one of the I/O handlers calls stop(), which
      will make the loop stop after the current event iteration completes.
      """
        # 判斷是否設(shè)置了,如果是,將直接退出。
        if self._stopped:
            self._stopped = False
            return
        self._running = True
        while True:
            # Never use an infinite timeout here - it can stall epoll
            # 設(shè)置輪詢時(shí)間
            poll_timeout = 0.2
            # Prevent IO event starvation by delaying new callbacks
            # to the next iteration of the event loop.
            callbacks = self._callbacks
            self._callbacks = []
            # 及時(shí)調(diào)用回調(diào)函數(shù)
            for callback in callbacks:
                self._run_callback(callback)
            if self._callbacks:
                poll_timeout = 0.0
            # 如果設(shè)置了超時(shí)時(shí)間
            if self._timeouts:
                # 獲取當(dāng)前時(shí)間
                now = time.time()
                while self._timeouts and self._timeouts[0].deadline <= now:
                    timeout = self._timeouts.pop(0)
                    self._run_callback(timeout.callback)
                if self._timeouts:
                    milliseconds = self._timeouts[0].deadline - now
                    poll_timeout = min(milliseconds, poll_timeout)
            # 再一次檢查事件循環(huán)是否在運(yùn)行
            if not self._running:
                break
            # 目前不清楚作用
            if self._blocking_signal_threshold is not None:
                # clear alarm so it doesn't fire while poll is waiting for
                # events.
                signal.setitimer(signal.ITIMER_REAL, 0, 0)

            try:
                # 開始等待事件發(fā)生
                # _impl初始化和poll源代碼見下面
                event_pairs = self._impl.poll(poll_timeout)
            except Exception, e:
                # Depending on python version and IOLoop implementation,
                # different exception types may be thrown and there are
                # two ways EINTR might be signaled:
                # * e.errno == errno.EINTR
                # * e.args is like (errno.EINTR, 'Interrupted system call')
                if (getattr(e, 'errno', None) == errno.EINTR or
                    (isinstance(getattr(e, 'args', None), tuple) and
                     len(e.args) == 2 and e.args[0] == errno.EINTR)):
                    continue
                else:
                    raise
            if self._blocking_signal_threshold is not None:
                signal.setitimer(signal.ITIMER_REAL,
                                 self._blocking_signal_threshold, 0)
            # Pop one fd at a time from the set of pending fds and run
            # its handler. Since that handler may perform actions on
            # other file descriptors, there may be reentrant calls to
            # this IOLoop that update self._events
            self._events.update(event_pairs)
            while self._events:
                fd, events = self._events.popitem()
                try:
                    # 見下面的分析
                    self._handlers[fd](fd, events)
                except (KeyboardInterrupt, SystemExit):
                    raise
                except (OSError, IOError), e:
                    if e.args[0] == errno.EPIPE:
                        # Happens when the client closes the connection
                        pass
                    else:
                        logging.error("Exception in I/O handler for fd %d",
                                      fd, exc_info=True)
                except:
                    logging.error("Exception in I/O handler for fd %d",
                                  fd, exc_info=True)
        # reset the stopped flag so another start/stop pair can be issued
        self._stopped = False
        if self._blocking_signal_threshold is not None:
            signal.setitimer(signal.ITIMER_REAL, 0, 0)

我們看看輸入localhost:8888,event_pairs的值:

可以看出,event_pairs是一個(gè)元組列表,其中第一個(gè)成員4為accept套接字值,1表示為事件類型。我們看看事件類型為:

 _EPOLLIN = 0x001
 _EPOLLPRI = 0x002
 _EPOLLOUT = 0x004  
 _EPOLLERR = 0x008
 _EPOLLHUP = 0x010
 _EPOLLRDHUP = 0x2000
 _EPOLLONESHOT = (1 << 30)
 _EPOLLET = (1 << 31)
 NONE = 0
 READ = _EPOLLIN
 WRITE = _EPOLLOUT
 ERROR = _EPOLLERR | _EPOLLHUP | _EPOLLRDHUP

可見上述,是文件描述符4,可讀事件發(fā)生。

self._impl

我們跟蹤self._impl初始化過程,可以看到事件循環(huán)核心epoll是如何被使用的。在IOLoop實(shí)例化開始:

class IOLoop(object):
    def __init__(self, impl = None):
        self._impl = impl or _poll

# Choose a poll implementation. Use epoll if it is available, fall back to
# select() for non-Linux platforms
# hasattr(object, attrname)表示某個(gè)對(duì)象中是否包含屬性
if hasattr(select, "epoll"):
    # Python 2.6+ on Linux
    # 在linux上使用的是select.epoll
    _poll = select.epoll
elif hasattr(select, "kqueue"):
    # Python 2.6+ on BSD or Mac
    _poll = _KQueue
else:
    try:
        # Linux systems with our C module installed
        import epoll
        _poll = _EPoll
    except:
        # All other systems
        import sys
        if "linux" in sys.platform:
            logging.warning("epoll module not found; using select()")
        _poll = _Select

從上面的代碼中,可以看到,_poll是對(duì)于多個(gè)平臺(tái)下epoll、_kQueue的抽象??匆幌聅elect.epoll下的返回結(jié)果:其返回對(duì)象是一個(gè)邊沿觸發(fā)的polling對(duì)象,當(dāng)然也可以用作水平觸發(fā)。

返回的select.epoll對(duì)象的方法:

  • epoll.close() 關(guān)閉epoll fd文件描述符
  • epoll.fileno() 返回epoll fd文件描述符只
  • epoll.register(fd, eventmask) 注冊(cè)fd某個(gè)事件
  • epoll.poll([timeout = -1, maxevents = -1]) wait for events. timeout in seconds

self._handlers[fd](fd, events)

顯然,self._handlers[fd]是返回一個(gè)回調(diào)函數(shù),用來(lái)處理fd上的事件events,這里測(cè)試的fd為4,事件EPOLLIN。我們來(lái)跟蹤一下self._handlers變化過程??纯丛贗OLoop初始化的過程。

def __init__(self):
    self._handles = {}
    ....
    if os.name != 'nt':
        r, w = os.pipe()
        self._set_nonblocking(r)
        self._set_nonblocking(w)
        .....
        self._waker_reader = os.fdopen(r, "rb", 0)
        self._waker_writer = os.fdopen(w, "wb", 0)    
    # 顯然這是對(duì)讀管道文件描述符事件處理函數(shù)
    self.add_handler(r, self._read_waker, self.READ)

add_handler(self, fd, handler, events)

def add_handler(self, fd, handler, events):
    self._handlers[fd] = stack_context.wrap(handler)
    self._impl.register(fd, events| self.ERROR)

可見add_handler干了兩件事情:

  • 回調(diào)函數(shù)設(shè)置,當(dāng)然不僅僅是簡(jiǎn)單的將handler賦值,而是使用了stack_context.wrap包裹了該函數(shù),具體實(shí)現(xiàn),見下面。
  • epoll對(duì)象添加該事件,就是在代碼的第二行。所以self._handlers[fd](fd, args)實(shí)際上就是設(shè)置的回調(diào)函數(shù)。那么用stack_context.wrap()來(lái)包裹究竟是為了什么了?

update_handler(self, fd, events)

 def update_handler(self, fd, events):
        """Changes the events we listen for fd."""
        self._impl.modify(fd, events | self.ERROR)

該函數(shù)用來(lái)修改fd感興趣的事件

以上就是python Tornado事件循環(huán)示例源碼解析的詳細(xì)內(nèi)容,更多關(guān)于python Tornado事件循環(huán)的資料請(qǐng)關(guān)注腳本之家其它相關(guān)文章!

相關(guān)文章

最新評(píng)論