早期的牛津lemmatize

# lemmaDict = IndexBuilder(r'E:\mdx-test\牛津高阶英语学习词典第八版.mdx')  # 108749, 貌似不错，返回 ['@@@LINK=fence\r\n'))))))
# def lemmatize(word):
#     lowercasedWord = word.lower()
#     dictResult = mdict.mdx_lookup(lowercasedWord)  # 对于牛津词典，Resisting直接查找所没有的，所以要转小写;
#     # print(dictResult)
#     # print(len(dictResult))
#     # print('dictResult------------------', word, mdict._mdx_file, repr(dictResult))
#     if dictResult == [] or len(dictResult) == 2:  # dictResult是[]:错误单词intuitiv6e、confpiracy或teaching posts；==2比如wanted的情况，可以成为一个独立单词：受通缉得的，也可以是want的过去式，目前选择前者；# integrated之类的词会有重定向和释义（在后）这两项
#         lemma = None  # 因为大多数单词都是小写形式,所以转成小写形式返回
#     elif '@@@LINK=' in dictResult[0]:  # positioned、fenced
#         # print('-----hunspell------>', hunSpell.stem(newWord.lower()))
#         lemma = dictResult[0][8:-1].rstrip()  # 有的词典的末尾是\n(牛津高阶英语学习词典第八版.mdx)，有的是\r\n(CollinsCOBUILDOverhaul+V+2-00.mdx),所以统一strip
#         # print(word, '-----ox------>', stem)
#     else:  # 返回释义的情况；如good
#         lemma = None
#     return lemma