我的第一个 Python 小项目，开放完整代码

# encoding: utf-8
"""
@file: kwic_service.py
@desc: providing functions about KWIC presentation
@author: group3
@time: 5/9/2021
"""

import re
from typing import List

获取关键词sel_word的窗口，默认窗口长度为5

def get_keyword_window(sel_word: str, words_of_sentence: List, length=5) -> List[str]:
    """
    find the index of sel_word at sentence, then decide words of @length size
    by backward and forward of it.
    For example: I am very happy to this course of psd if sel_word is happy, then
    returning: [am, very, happy, to, this]

    if length is even, then returning [very, happy, to, this]

    remember: sel_word being word root
    """
    if length <= 0 or len(words_of_sentence) <= length:
        return words_of_sentence
    index = -1
    for iw, word in enumerate(words_of_sentence):
        word = word.lower()
        if len(re.findall(sel_word.lower(), word)) > 0:
            index = iw
            break

    if index == -1:
        # log.warning("warning: cannot find %s in sentence: %s" % (sel_word, words_of_sentence))
        return words_of_sentence
    # backward is not enough
    if index < length // 2:
        back_slice = words_of_sentence[:index]
        # forward is also not enough,
        # showing the sentence is too short compared to length parameter
        if (length - index) >= len(words_of_sentence):
            return words_of_sentence
        else:
            return back_slice + words_of_sentence[index: index + length - len(back_slice)]
    # forward is not enough
    if (index + length // 2) >= len(words_of_sentence):
        forward_slice = words_of_sentence[index:len(words_of_sentence)]
        # backward is also not enough,
        # showing the sentence is too short compared to length parameter
        if index - length <= 0:
            return words_of_sentence
        else:
            return words_of_sentence[index - (length - len(forward_slice)):index] + forward_slice

    return words_of_sentence[index - length // 2: index + length // 2 + 1] if length % 2 \
        else words_of_sentence[index - length // 2 + 1: index + length // 2 + 1]

KWIC显示逻辑，我放在另外一个方法中，鉴于代码长度，放在这里文章显示太长了，所以完整代码全部归档到这里：

 http://www.zglg.work/Python-20-topics/python-project1-kwic/

测试代码

# encoding: utf-8
"""
@file: test_kwic_show.py
@desc:
@author: group3
@time: 5/3/2021
"""
from src.feature.kwic import kwic_show

if __name__ == '__main__':
    words = ['I', 'am', 'very', 'happy', 'to', 'this', 'course', 'of', 'psd']

    print(kwic_show('English', words, 'I', window_size=1)[0])
    print(kwic_show('English', words, 'I', window_size=5)[0])
    
    print(kwic_show('English', words, 'very', token_space_param=5)[0])
    print(kwic_show('English', words, 'very', window_size=6, token_space_param=5)[0])
    print(kwic_show('English', words, 'very', window_size=1, token_space_param=5)[0])
    
    # test boundary
    print(kwic_show('English', words, 'stem', align_param=20)[0])
    print(kwic_show('English', words, 'stem', align_param=100)[0])
    print(kwic_show('English', words, 'II', window_size=1)[0])
    print(kwic_show('English', words, 'related', window_size=10000)[0])

打印结果

                                  I 
                                  I am very happy to
                        I am     very     happy to this course of psd
                        I am     very     happy to this
                                 very     
None
None
None
None