将Flask缓存与生成器函数一起使用

Question

我正在尝试使用Flask-Cache来缓存生成器函数。Flask-Cache需要酸洗，因为生成器不能酸洗，这是一个问题。在Flask-Cache中使用生成器有什么变通方法吗？

Answer 1

如果您想缓存结果，只需在生成器上调用list并存储结果即可。如果结果太大而不能安全地存储在内存中(即使是临时的)，那么将结果包装在一个可以restore itself when unpickled的类实例中

# Warning: This is terribly tightly bound together
# There are better ways to do this - this is just by way of example

def some_generator(identifier):
    """This could be anything
        * A database wrapper
        * A permutation generator
        * Something that streams (and transforms) data from disk
        * The possibilities are endless"""
    if identifier == "result_set_1":
        results = [(x, y) for x in range(10) for y in range(x)]
    elif identifier == "result_set_2":
        results = [(x, 0) for x in range(20)]
    else:
        results = [(1, 2), (3, 4)]

    for row in results:
        yield row


class GeneratorWrapper(object):
    """
    Use::

        @cache
        some_function(identifier):
            return GeneratorWrapper(identifier)
    """

    def __init__(self, identifier=None):
        self.identifier = identifier
        if identifier is not None:
            self.data = some_generator(identifier)

    def __getstate__(self):
        """When pickled return the key to rebuilding the generator"""
        return self.identifier

    def __setstate__(self, state):
        """When unpickled, restore the generator using the pickled key"""
        self.identifier = state
        if state is not None:
            self.data = some_generator(state)

    def __iter__(self):
        return self.data