[ex]递归实现简易tokenizer

问题

用 Python 实现函数 count_words()，该函数输入字符串 s 和数字 n，返回 s 中 n 个出现频率最高的单词。返回值是一个元组列表，包含出现次数最高的 n 个单词及其次数,即 [(<单词1>, <次数1>), (<单词2>, <次数2>), ... ]，按出现次数降序排列。

 1 """Count words."""
def get_pos(k_new, v_new, p_new, ksorted,vsorted,plist):
    if vsorted:
        i = 0
        while i < len(vsorted):
            if v_new > vsorted[i]:
                return get_pos(k_new, v_new, plist[i], ksorted[:i-1], vsorted[:i-1], plist[:i-1])
            elif v_new < vsorted[i]:
                return get_pos(k_new, v_new, plist[i]+1, ksorted[i+1:], vsorted[i+1:], plist[i+1:])
            elif k_new < ksorted[i]:
                return get_pos(k_new, v_new, plist[i], ksorted[:i-1], vsorted[:i-1], plist[:i-1])
            else:
                return get_pos(k_new, v_new, plist[i]+1, ksorted[i+1:], vsorted[i+1:], plist[i+1:])
    else:
        return p_new      
        
def count_words(s, n):
    """Return the n most frequently occuring words in s."""
    
    # Count the number of occurences of each word in s
    wordlist = s.split(' ')
    wdict = {}
    for word in wordlist:
        if word in wdict.keys():
            wdict[word] += 1
        else:
            wdict[word] = 1
    # Sort the occurences in descending order (alphabetically in case of ties)
    vs = []
    ks = []
    ps = []
    w_sorted = []
    for key, value in wdict.items():
        p_new = get_pos(key,value,0,ks,vs,ps)
        w_sorted.insert(p_new, (key,value))
        ks.insert(p_new, key)
        vs.insert(p_new, value)
        ps = range(len(vs))
    
    # Return the top n most frequent words.
    top_n = w_sorted[:n]
    return top_n


def test_run():
    """Test count_words() with some inputs."""
    print(count_words("cat bat mat cat bat cat", 3))
    print(count_words("betty bought a bit of butter but the butter was bitter", 3))


if __name__ == '__main__':
    test_run()