当前位置:主页 > 软件编程 > Python代码 >

Python中lru_cache的使用和实现详解

时间:2022-04-15 09:30:32 | 栏目:Python代码 | 点击:

在计算机软件领域,缓存(Cache)指的是将部分数据存储在内存中,以便下次能够更快地访问这些数据,这也是一个典型的用空间换时间的例子。一般用于缓存的内存空间是固定的,当有更多的数据需要缓存的时候,需要将已缓存的部分数据清除后再将新的缓存数据放进去。需要清除哪些数据,就涉及到了缓存置换的策略,LRU(Least Recently Used,最近最少使用)是很常见的一个,也是 Python 中提供的缓存置换策略。

下面我们通过一个简单的示例来看 Python 中的 lru_cache 是如何使用的。

def factorial(n):
  print(f"计算 {n} 的阶乘")
  return 1 if n <= 1 else n * factorial(n - 1)

a = factorial(5)
print(f'5! = {a}')
b = factorial(3)
print(f'3! = {b}')

上面的代码中定义了函数 factorial,通过递归的方式计算 n 的阶乘,并且在函数调用的时候打印出 n 的值。然后分别计算 5 和 3 的阶乘,并打印结果。运行上面的代码,输出如下

计算 5 的阶乘
计算 4 的阶乘
计算 3 的阶乘
计算 2 的阶乘
计算 1 的阶乘
5! = 120
计算 3 的阶乘
计算 2 的阶乘
计算 1 的阶乘
3! = 6

可以看到, factorial(3) 的结果在计算 factorial(5) 的时候已经被计算过了,但是后面又被重复计算了。为了避免这种重复计算,我们可以在定义函数 factorial 的时候加上 lru_cache 装饰器,如下所示

import functools
# 注意 lru_cache 后的一对括号,证明这是带参数的装饰器
@functools.lru_cache()
def factorial(n):
  print(f"计算 {n} 的阶乘")
  return 1 if n <= 1 else n * factorial(n - 1)

重新运行代码,输入如下

计算 5 的阶乘
计算 4 的阶乘
计算 3 的阶乘
计算 2 的阶乘
计算 1 的阶乘
5! = 120
3! = 6

可以看到,这次在调用 factorial(3) 的时候没有打印相应的输出,也就是说 factorial(3) 是直接从缓存读取的结果,证明缓存生效了。

被 lru_cache 修饰的函数在被相同参数调用的时候,后续的调用都是直接从缓存读结果,而不用真正执行函数。下面我们深入源码,看看 Python 内部是怎么实现 lru_cache 的。写作时 Python 最新发行版是 3.9,所以这里使用的是Python 3.9 的源码 ,并且保留了源码中的注释。

def lru_cache(maxsize=128, typed=False):
  """Least-recently-used cache decorator.
  If *maxsize* is set to None, the LRU features are disabled and the cache
  can grow without bound.
  If *typed* is True, arguments of different types will be cached separately.
  For example, f(3.0) and f(3) will be treated as distinct calls with
  distinct results.
  Arguments to the cached function must be hashable.
  View the cache statistics named tuple (hits, misses, maxsize, currsize)
  with f.cache_info(). Clear the cache and statistics with f.cache_clear().
  Access the underlying function with f.__wrapped__.
  See: http://en.wikipedia.org/wiki/Cache_replacement_policies#Least_recently_used_(LRU)
  """

  # Users should only access the lru_cache through its public API:
  #    cache_info, cache_clear, and f.__wrapped__
  # The internals of the lru_cache are encapsulated for thread safety and
  # to allow the implementation to change (including a possible C version).
  
  if isinstance(maxsize, int):
    # Negative maxsize is treated as 0
    if maxsize < 0:
      maxsize = 0
  elif callable(maxsize) and isinstance(typed, bool):
    # The user_function was passed in directly via the maxsize argument
    user_function, maxsize = maxsize, 128
    wrapper = _lru_cache_wrapper(user_function, maxsize, typed, _CacheInfo)
    wrapper.cache_parameters = lambda : {'maxsize': maxsize, 'typed': typed}
    return update_wrapper(wrapper, user_function)
  elif maxsize is not None:
    raise TypeError(
      'Expected first argument to be an integer, a callable, or None')
  
  def decorating_function(user_function):
    wrapper = _lru_cache_wrapper(user_function, maxsize, typed, _CacheInfo)
    wrapper.cache_parameters = lambda : {'maxsize': maxsize, 'typed': typed}
    return update_wrapper(wrapper, user_function)
  
  return decorating_function

这段代码中有如下几个关键点

关键字参数

maxsize 表示缓存容量,如果为 None 表示容量不设限, typed 表示是否区分参数类型,注释中也给出了解释,如果 typed == True ,那么 f(3) 和 f(3.0) 会被认为是不同的函数调用。

第 24 行的条件分支

如果 lru_cache 的第一个参数是可调用的,直接返回 wrapper,也就是把 lru_cache 当做不带参数的装饰器,这是 Python 3.8 才有的特性,也就是说在 Python 3.8 及之后的版本中我们可以用下面的方式使用 lru_cache,可能是为了防止程序员在使用 lru_cache 的时候忘记加括号。

import functools
# 注意 lru_cache 后面没有括号,
# 证明这是将其当做不带参数的装饰器
@functools.lru_cache
def factorial(n):
  print(f"计算 {n} 的阶乘")
  return 1 if n <= 1 else n * factorial(n - 1)

注意,Python 3.8 之前的版本运行上面代码会报错:TypeError: Expected maxsize to be an integer or None。

lru_cache 的具体逻辑是在 _lru_cache_wrapper 函数中实现的,还是一样,列出源码,保留注释。

def _lru_cache_wrapper(user_function, maxsize, typed, _CacheInfo):
  # Constants shared by all lru cache instances:
  sentinel = object()     # unique object used to signal cache misses
  make_key = _make_key     # build a key from the function arguments
  PREV, NEXT, KEY, RESULT = 0, 1, 2, 3  # names for the link fields

  cache = {}
  hits = misses = 0
  full = False
  cache_get = cache.get  # bound method to lookup a key or return None
  cache_len = cache.__len__ # get cache size without calling len()
  lock = RLock()      # because linkedlist updates aren't threadsafe
  root = []        # root of the circular doubly linked list
  root[:] = [root, root, None, None]   # initialize by pointing to self

  if maxsize == 0:

    def wrapper(*args, **kwds):
      # No caching -- just a statistics update
      nonlocal misses
      misses += 1
      result = user_function(*args, **kwds)
      return result

  elif maxsize is None:

    def wrapper(*args, **kwds):
      # Simple caching without ordering or size limit
      nonlocal hits, misses
      key = make_key(args, kwds, typed)
      result = cache_get(key, sentinel)
      if result is not sentinel:
        hits += 1
        return result
      misses += 1
      result = user_function(*args, **kwds)
      cache[key] = result
      return result

  else:

    def wrapper(*args, **kwds):
      # Size limited caching that tracks accesses by recency
      nonlocal root, hits, misses, full
      key = make_key(args, kwds, typed)
      with lock:
        link = cache_get(key)
        if link is not None:
          # Move the link to the front of the circular queue
          link_prev, link_next, _key, result = link
          link_prev[NEXT] = link_next
          link_next[PREV] = link_prev
          last = root[PREV]
          last[NEXT] = root[PREV] = link
          link[PREV] = last
          link[NEXT] = root
          hits += 1
          return result
        misses += 1
      result = user_function(*args, **kwds)
      with lock:
        if key in cache:
          # Getting here means that this same key was added to the
          # cache while the lock was released. Since the link
          # update is already done, we need only return the
          # computed result and update the count of misses.
          pass
        elif full:
          # Use the old root to store the new key and result.
          oldroot = root
          oldroot[KEY] = key
          oldroot[RESULT] = result
          # Empty the oldest link and make it the new root.
          # Keep a reference to the old key and old result to
          # prevent their ref counts from going to zero during the
          # update. That will prevent potentially arbitrary object
          # clean-up code (i.e. __del__) from running while we're
          # still adjusting the links.
          root = oldroot[NEXT]
          oldkey = root[KEY]
          oldresult = root[RESULT]
          root[KEY] = root[RESULT] = None
          # Now update the cache dictionary.
          del cache[oldkey]
          # Save the potentially reentrant cache[key] assignment
          # for last, after the root and links have been put in
          # a consistent state.
          cache[key] = oldroot
        else:
          # Put result in a new link at the front of the queue.
          last = root[PREV]
          link = [last, root, key, result]
          last[NEXT] = root[PREV] = cache[key] = link
          # Use the cache_len bound method instead of the len() function
          # which could potentially be wrapped in an lru_cache itself.
          full = (cache_len() >= maxsize)
      return result

  def cache_info():
    """Report cache statistics"""
    with lock:
      return _CacheInfo(hits, misses, maxsize, cache_len())

  def cache_clear():
    """Clear the cache and cache statistics"""
    nonlocal hits, misses, full
    with lock:
      cache.clear()
      root[:] = [root, root, None, None]
      hits = misses = 0
      full = False

  wrapper.cache_info = cache_info
  wrapper.cache_clear = cache_clear
  return wrapper

函数开始的地方 2~14 行定义了一些关键变量,

接下来根据 maxsize 不同,定义不同的 wrapper 。

最后给 wrapper 添加两个属性函数 cache_info 和 cache_clear , cache_info 显示当前缓存的命中情况的统计数据, cache_clear 用于清空缓存。对于上面阶乘相关的代码,如果在最后执行 factorial.cache_info() ,会输出

CacheInfo(hits=1, misses=5, maxsize=128, currsize=5)

第一次执行 factorial(5) 的时候都没命中,所以 misses = 5,第二次执行 factorial(3) 的时候,缓存命中,所以 hits = 1。

最后需要说明的是, 对于有多个关键字参数的函数,如果两次调用函数关键字参数传入的顺序不同,会被认为是不同的调用,不会命中缓存。另外,被 lru_cache 装饰的函数不能包含可变类型参数如 list,因为它们不支持 hash。

总结一下,这篇文章首先简介了一下缓存的概念,然后展示了在 Python 中 lru_cache 的使用方法,最后通过源码分析了 Python 中 lru_cache 的实现细节。

您可能感兴趣的文章:

相关文章