asyncio剖析—认识awaitable(一)

Asyncio 我的理解

1. __await__ 协议是啥？ 为啥Awaitable，Coroutine，Future，Task都有 __await__
2. asyncio 调用栈都有哪些,具体包含那几部

Awaitable源码

class Awaitable(metaclass=ABCMeta):

    __slots__ = ()

    @abstractmethod
    def __await__(self):
        yield

    @classmethod
    def __subclasshook__(cls, C):
        if cls is Awaitable:
            return _check_methods(C, "__await__")
        return NotImplemented

Coroutine源码

class Coroutine(Awaitable):

    __slots__ = ()

    @abstractmethod
    def send(self, value):
        """Send a value into the coroutine.
        Return next yielded value or raise StopIteration.
        """
        raise StopIteration

    @abstractmethod
    def throw(self, typ, val=None, tb=None):
        """Raise an exception in the coroutine.
        Return next yielded value or raise StopIteration.
        """
        if val is None:
            if tb is None:
                raise typ
            val = typ()
        if tb is not None:
            val = val.with_traceback(tb)
        raise val

    def close(self):
        """Raise GeneratorExit inside coroutine.
        """
        try:
            self.throw(GeneratorExit)
        except (GeneratorExit, StopIteration):
            pass
        else:
            raise RuntimeError("coroutine ignored GeneratorExit")

    @classmethod
    def __subclasshook__(cls, C):
        if cls is Coroutine:
            return _check_methods(C, '__await__', 'send', 'throw', 'close')
        return NotImplemented


Coroutine.register(coroutine)

Future源码

class Future:
    """This class is *almost* compatible with concurrent.futures.Future.

    Differences:

    - This class is not thread-safe.

    - result() and exception() do not take a timeout argument and
      raise an exception when the future isn't done yet.

    - Callbacks registered with add_done_callback() are always called
      via the event loop's call_soon().

    - This class is not compatible with the wait() and as_completed()
      methods in the concurrent.futures package.

    (In Python 3.4 or later we may be able to unify the implementations.)
    """

    # Class variables serving as defaults for instance variables.
    _state = _PENDING
    _result = None
    _exception = None
    _loop = None
    _source_traceback = None

    # This field is used for a dual purpose:
    # - Its presence is a marker to declare that a class implements
    #   the Future protocol (i.e. is intended to be duck-type compatible).
    #   The value must also be not-None, to enable a subclass to declare
    #   that it is not compatible by setting this to None.
    # - It is set by __iter__() below so that Task._step() can tell
    #   the difference between `yield from Future()` (correct) vs.
    #   `yield Future()` (incorrect).
    _asyncio_future_blocking = False

    _log_traceback = False

    def __init__(self, *, loop=None):
        """Initialize the future.

        The optional event_loop argument allows explicitly setting the event
        loop object used by the future. If it's not provided, the future uses
        the default event loop.
        """
        if loop is None:
            self._loop = events.get_event_loop()
        else:
            self._loop = loop
        self._callbacks = []
        if self._loop.get_debug():
            self._source_traceback = events.extract_stack(sys._getframe(1))

    _repr_info = base_futures._future_repr_info

    def __repr__(self):
        return '<%s %s>' % (self.__class__.__name__, ' '.join(self._repr_info()))

    # On Python 3.3 and older, objects with a destructor part of a reference
    # cycle are never destroyed. It's not more the case on Python 3.4 thanks
    # to the PEP 442.
    if compat.PY34:
        def __del__(self):
            if not self._log_traceback:
                # set_exception() was not called, or result() or exception()
                # has consumed the exception
                return
            exc = self._exception
            context = {
                'message': ('%s exception was never retrieved'
                            % self.__class__.__name__),
                'exception': exc,
                'future': self,
            }
            if self._source_traceback:
                context['source_traceback'] = self._source_traceback
            self._loop.call_exception_handler(context)

    def cancel(self):
        """Cancel the future and schedule callbacks.

        If the future is already done or cancelled, return False.  Otherwise,
        change the future's state to cancelled, schedule the callbacks and
        return True.
        """
        self._log_traceback = False
        if self._state != _PENDING:
            return False
        self._state = _CANCELLED
        self._schedule_callbacks()
        return True

    def _schedule_callbacks(self):
        """Internal: Ask the event loop to call all callbacks.

        The callbacks are scheduled to be called as soon as possible. Also
        clears the callback list.
        """
        callbacks = self._callbacks[:]
        if not callbacks:
            return

        self._callbacks[:] = []
        for callback in callbacks:
            self._loop.call_soon(callback, self)

    def cancelled(self):
        """Return True if the future was cancelled."""
        return self._state == _CANCELLED

    # Don't implement running(); see http://bugs.python.org/issue18699

    def done(self):
        """Return True if the future is done.

        Done means either that a result / exception are available, or that the
        future was cancelled.
        """
        return self._state != _PENDING

    def result(self):
        """Return the result this future represents.

        If the future has been cancelled, raises CancelledError.  If the
        future's result isn't yet available, raises InvalidStateError.  If
        the future is done and has an exception set, this exception is raised.
        """
        if self._state == _CANCELLED:
            raise CancelledError
        if self._state != _FINISHED:
            raise InvalidStateError('Result is not ready.')
        self._log_traceback = False
        if self._exception is not None:
            raise self._exception
        return self._result

    def exception(self):
        """Return the exception that was set on this future.

        The exception (or None if no exception was set) is returned only if
        the future is done.  If the future has been cancelled, raises
        CancelledError.  If the future isn't done yet, raises
        InvalidStateError.
        """
        if self._state == _CANCELLED:
            raise CancelledError
        if self._state != _FINISHED:
            raise InvalidStateError('Exception is not set.')
        self._log_traceback = False
        return self._exception

    def add_done_callback(self, fn):
        """Add a callback to be run when the future becomes done.

        The callback is called with a single argument - the future object. If
        the future is already done when this is called, the callback is
        scheduled with call_soon.
        """
        if self._state != _PENDING:
            self._loop.call_soon(fn, self)
        else:
            self._callbacks.append(fn)

    # New method not in PEP 3148.

    def remove_done_callback(self, fn):
        """Remove all instances of a callback from the "call when done" list.

        Returns the number of callbacks removed.
        """
        filtered_callbacks = [f for f in self._callbacks if f != fn]
        removed_count = len(self._callbacks) - len(filtered_callbacks)
        if removed_count:
            self._callbacks[:] = filtered_callbacks
        return removed_count

    # So-called internal methods (note: no set_running_or_notify_cancel()).

    def set_result(self, result):
        """Mark the future done and set its result.

        If the future is already done when this method is called, raises
        InvalidStateError.
        """
        if self._state != _PENDING:
            raise InvalidStateError('{}: {!r}'.format(self._state, self))
        self._result = result
        self._state = _FINISHED
        self._schedule_callbacks()

    def set_exception(self, exception):
        """Mark the future done and set an exception.

        If the future is already done when this method is called, raises
        InvalidStateError.
        """
        if self._state != _PENDING:
            raise InvalidStateError('{}: {!r}'.format(self._state, self))
        if isinstance(exception, type):
            exception = exception()
        if type(exception) is StopIteration:
            raise TypeError("StopIteration interacts badly with generators "
                            "and cannot be raised into a Future")
        self._exception = exception
        self._state = _FINISHED
        self._schedule_callbacks()
        if compat.PY34:
            self._log_traceback = True
        else:
            self._tb_logger = _TracebackLogger(self, exception)
            # Arrange for the logger to be activated after all callbacks
            # have had a chance to call result() or exception().
            self._loop.call_soon(self._tb_logger.activate)

    def __iter__(self):
        if not self.done():
            self._asyncio_future_blocking = True
            yield self  # This tells Task to wait for completion.
        assert self.done(), "yield from wasn't used with future"
        return self.result()  # May raise too.

    if compat.PY35:
        __await__ = __iter__ # make compatible with 'await' expression

Task源码

class Task(futures.Future):
    """A coroutine wrapped in a Future."""

    # An important invariant maintained while a Task not done:
    #
    # - Either _fut_waiter is None, and _step() is scheduled;
    # - or _fut_waiter is some Future, and _step() is *not* scheduled.
    #
    # The only transition from the latter to the former is through
    # _wakeup().  When _fut_waiter is not None, one of its callbacks
    # must be _wakeup().

    # Weak set containing all tasks alive.
    _all_tasks = weakref.WeakSet()

    # Dictionary containing tasks that are currently active in
    # all running event loops.  {EventLoop: Task}
    _current_tasks = {}

    # If False, don't log a message if the task is destroyed whereas its
    # status is still pending
    _log_destroy_pending = True

    @classmethod
    def current_task(cls, loop=None):
        """Return the currently running task in an event loop or None.

        By default the current task for the current event loop is returned.

        None is returned when called not in the context of a Task.
        """
        if loop is None:
            loop = events.get_event_loop()
        return cls._current_tasks.get(loop)

    @classmethod
    def all_tasks(cls, loop=None):
        """Return a set of all tasks for an event loop.

        By default all tasks for the current event loop are returned.
        """
        if loop is None:
            loop = events.get_event_loop()
        return {t for t in cls._all_tasks if t._loop is loop}

    def __init__(self, coro, *, loop=None):
        assert coroutines.iscoroutine(coro), repr(coro)
        super().__init__(loop=loop)
        if self._source_traceback:
            del self._source_traceback[-1]
        self._coro = coro
        self._fut_waiter = None
        self._must_cancel = False
        self._loop.call_soon(self._step)
        self.__class__._all_tasks.add(self)

    # On Python 3.3 or older, objects with a destructor that are part of a
    # reference cycle are never destroyed. That's not the case any more on
    # Python 3.4 thanks to the PEP 442.
    if compat.PY34:
        def __del__(self):
            if self._state == futures._PENDING and self._log_destroy_pending:
                context = {
                    'task': self,
                    'message': 'Task was destroyed but it is pending!',
                }
                if self._source_traceback:
                    context['source_traceback'] = self._source_traceback
                self._loop.call_exception_handler(context)
            futures.Future.__del__(self)

    def _repr_info(self):
        return base_tasks._task_repr_info(self)

    def get_stack(self, *, limit=None):
        """Return the list of stack frames for this task's coroutine.

        If the coroutine is not done, this returns the stack where it is
        suspended.  If the coroutine has completed successfully or was
        cancelled, this returns an empty list.  If the coroutine was
        terminated by an exception, this returns the list of traceback
        frames.

        The frames are always ordered from oldest to newest.

        The optional limit gives the maximum number of frames to
        return; by default all available frames are returned.  Its
        meaning differs depending on whether a stack or a traceback is
        returned: the newest frames of a stack are returned, but the
        oldest frames of a traceback are returned.  (This matches the
        behavior of the traceback module.)

        For reasons beyond our control, only one stack frame is
        returned for a suspended coroutine.
        """
        return base_tasks._task_get_stack(self, limit)

    def print_stack(self, *, limit=None, file=None):
        """Print the stack or traceback for this task's coroutine.

        This produces output similar to that of the traceback module,
        for the frames retrieved by get_stack().  The limit argument
        is passed to get_stack().  The file argument is an I/O stream
        to which the output is written; by default output is written
        to sys.stderr.
        """
        return base_tasks._task_print_stack(self, limit, file)

    def cancel(self):
        """Request that this task cancel itself.

        This arranges for a CancelledError to be thrown into the
        wrapped coroutine on the next cycle through the event loop.
        The coroutine then has a chance to clean up or even deny
        the request using try/except/finally.

        Unlike Future.cancel, this does not guarantee that the
        task will be cancelled: the exception might be caught and
        acted upon, delaying cancellation of the task or preventing
        cancellation completely.  The task may also return a value or
        raise a different exception.

        Immediately after this method is called, Task.cancelled() will
        not return True (unless the task was already cancelled).  A
        task will be marked as cancelled when the wrapped coroutine
        terminates with a CancelledError exception (even if cancel()
        was not called).
        """
        self._log_traceback = False
        if self.done():
            return False
        if self._fut_waiter is not None:
            if self._fut_waiter.cancel():
                # Leave self._fut_waiter; it may be a Task that
                # catches and ignores the cancellation so we may have
                # to cancel it again later.
                return True
        # It must be the case that self._step is already scheduled.
        self._must_cancel = True
        return True

    def _step(self, exc=None):
        assert not self.done(), \
            '_step(): already done: {!r}, {!r}'.format(self, exc)
        if self._must_cancel:
            if not isinstance(exc, futures.CancelledError):
                exc = futures.CancelledError()
            self._must_cancel = False
        coro = self._coro
        self._fut_waiter = None

        self.__class__._current_tasks[self._loop] = self
        # Call either coro.throw(exc) or coro.send(None).
        try:
            if exc is None:
                # We use the `send` method directly, because coroutines
                # don't have `__iter__` and `__next__` methods.
                result = coro.send(None)
            else:
                result = coro.throw(exc)
        except StopIteration as exc:
            if self._must_cancel:
                # Task is cancelled right before coro stops.
                self._must_cancel = False
                self.set_exception(futures.CancelledError())
            else:
                self.set_result(exc.value)
        except futures.CancelledError:
            super().cancel()  # I.e., Future.cancel(self).
        except Exception as exc:
            self.set_exception(exc)
        except BaseException as exc:
            self.set_exception(exc)
            raise
        else:
            blocking = getattr(result, '_asyncio_future_blocking', None)
            if blocking is not None:
                # Yielded Future must come from Future.__iter__().
                if result._loop is not self._loop:
                    self._loop.call_soon(
                        self._step,
                        RuntimeError(
                            'Task {!r} got Future {!r} attached to a '
                            'different loop'.format(self, result)))
                elif blocking:
                    if result is self:
                        self._loop.call_soon(
                            self._step,
                            RuntimeError(
                                'Task cannot await on itself: {!r}'.format(
                                    self)))
                    else:
                        result._asyncio_future_blocking = False
                        result.add_done_callback(self._wakeup)
                        self._fut_waiter = result
                        if self._must_cancel:
                            if self._fut_waiter.cancel():
                                self._must_cancel = False
                else:
                    self._loop.call_soon(
                        self._step,
                        RuntimeError(
                            'yield was used instead of yield from '
                            'in task {!r} with {!r}'.format(self, result)))
            elif result is None:
                # Bare yield relinquishes control for one event loop iteration.
                self._loop.call_soon(self._step)
            elif inspect.isgenerator(result):
                # Yielding a generator is just wrong.
                self._loop.call_soon(
                    self._step,
                    RuntimeError(
                        'yield was used instead of yield from for '
                        'generator in task {!r} with {!r}'.format(
                            self, result)))
            else:
                # Yielding something else is an error.
                self._loop.call_soon(
                    self._step,
                    RuntimeError(
                        'Task got bad yield: {!r}'.format(result)))
        finally:
            self.__class__._current_tasks.pop(self._loop)
            self = None  # Needed to break cycles when an exception occurs.

    def _wakeup(self, future):
        try:
            future.result()
        except Exception as exc:
            # This may also be a cancellation.
            self._step(exc)
        else:
            # Don't pass the value of `future.result()` explicitly,
            # as `Future.__iter__` and `Future.__await__` don't need it.
            # If we call `_step(value, None)` instead of `_step()`,
            # Python eval loop would use `.send(value)` method call,
            # instead of `__next__()`, which is slower for futures
            # that return non-generator iterators from their `__iter__`.
            self._step()
        self = None  # Needed to break cycles when an exception occurs.

问题1 Awaitable obj

Coroutine 继承Awaitable
Future 实现__await__ 协议
        if compat.PY35:
        __await__ = __iter__ # make compatible with 'await' expression
Task 继承Future

我的理解 所有 Awaitable都可以变为coroutine

tasks.py 中如下：
@coroutine
def _wrap_awaitable(awaitable):
    """Helper for asyncio.ensure_future().

    Wraps awaitable (an object with __await__) into a coroutine
    that will later be wrapped in a Task by ensure_future().
    """
    return (yield from awaitable.__await__())

问题二 asyncio 调用栈

asyncio 调用栈

asyncio.get_event_loop() #根据操作系统生成select轮询 命名loop
                |
                |
                |
                v
asyncio.ensure_future(start_server) 
                | 函数作用：
                | 1.future直接返回
                | 2.iscoroutine封装成task
                | 3.__await__对象 调用_wrap_awaitable封装成coroutine
                | 
                v
        task = tasks.Task(coro, loop=self) 
                |
                |
                | self=loop,callback是coro包装成的TaskStepMethWrapper
                |
                v
        handle = events.Handle(callback, args, self) #将task封装成event 放入loop 循环中
        yeild future 挂起future自身等待轮询
                |
                | 重点（类似代码）：
                |    self.coro.send(future.result) #初始化task，启动协程
                |                  |
                |                  v
                |    yield from connect(sock, ('example.com', 80)) #委托生成器到connect
                |                  |
                |                  v
                |    selector.register(sock.fileno(), EVENT_WRITE,on_connected) #注册event 到loop
                |    yield from f #委托生成器到future 
                |                  |
                |                  v
                |        def __iter__(self):
                |            yield self  #挂起future自己，直到loop轮询过来进入下一步
                |            return self.result
                |
                |
                v
        return task
                |
                |
                |
                |
                v
loop.run_until_complete() 启动loop

• 国外大神剖析 http://aosabook.org/en/500L/a-web-crawler-with-asyncio-coroutines.html
• 学习博客 https://hatboy.github.io/archives/page/2/
• github async demo：https://github.com/langgithub/async-aiotutorial