如何为每第 n 个句子换行？答案

【问题标题】：How to put a newline for every n'th sentence?如何为每第 n 个句子换行？
【发布时间】：2020-07-13 16:30:21
【问题描述】：

苦苦思索如何为长文本字符串中的每 5 个句子添加一个新行。

输入示例

text = 'The puppy is cute. Summer is great. Happy Friday. Sentence4. Sentence5. Sentence6. Sentence7.

期望的输出：

The puppy is cute. Summer is great. Happy friday. Sentence4. Sentence5.
Sentence6. Sentence7.

有人可以帮忙吗？

【问题讨论】：

句子是如何定义的？您是否允许输入诸如先生或其他缩写？还是椭圆？等等你有没有尝试过？
句子以大写字母开头，以句号结尾。所以句号可以用来区分句子
您的示例包含第一个字母的小写。这是否意味着您希望它不使用大写字母来区分？
很高兴使用大写字母。我将用大写字母更新上面的示例。

标签： python string newline

【解决方案1】：

试试这个：

text = 'The puppy is cute. Summer is great. Happy friday. sentence4. sentence5. sentence6. sentence7.'
splittext = text.split(".")
for x in range(5, len(splittext), 5):
    splittext[x] = "\n"+splittext[x].lstrip()
text = ".".join(splittext)
print(text)

【讨论】：

感谢您的功能！

【解决方案2】：

使用正则表达式。在“[not .] 后跟 .”的 5 次匹配后添加 \n。

import re
text = 'The puppy is cute. Summer is great. Happy friday. sentence4. sentence5. sentence6. sentence7.'

print(re.sub(r'((?:[^.]+\.\s*){5})',r'\1\n',text))

一个更高级的正则表达式句子匹配器，通过匹配结束标点符号来处理缩写和其他标点符号。
参考：https://mikedombrowski.com/2017/04/regex-sentence-splitter/
注意：仍然存在失败的边缘情况，例如 T.V. 后跟 Mr. 需要双空格来表示单独的句子。带有句子的引用将被拆分。等等

import re
sentence_regex = r'((.*?([\.\?!][\'\"\u2018\u2019\u201c\u201d\)\]]*\s*(?<!\w\.\w.)(?<![A-Z][a-z][a-z]\.)(?<![A-Z][a-z]\.)(?<![A-Z]\.)\s+)){5})'
text = 'The puppy is cute. Watch T.V.  Mr. Summers is great. Say "my name."  My name is.  Or not... Happy friday? Sentence4. Sentence5. Sentence6. Sentence7.'
text += " " + text

print(re.sub(sentence_regex,r'\1\n',text))

任何比这更复杂的东西，您都可能想研究语言处理工具包。

【讨论】：

【解决方案3】：

这是一个在第 5 句末尾添加换行符的简单函数

def new_line(sentence: str):
    # characters that mark the end of a sentence
    end_of_sentence_markers = ['.', '!', '?', '...']
    # after n sentences insert new_line
    n = 5

    # keeps track 
    count = 0
    # final string as list for efficiency
    final_str = []
    # split at space
    sentence_split = sentence.split(' ')

    # traverse the sentence split
    for word in sentence_split:
        # if end of sentence is present then increase count
        if word[-1] in end_of_sentence_markers:
            count += 1
        # if count is equal to n then add newline otherwise add space
        if count == n:
            final_str.append(word + '\n')
            count = 0
        else:
            final_str.append(word + ' ')


    # return the string version of the list
    return ''.join(final_str)

这是修改后的版本：

def new_line_better(sentence: str, n: int):
    # final string as list for efficiency
    final_str = []
    # split at period and remove extra spaces
    sentence_split = list( map( lambda x : x.strip(),  sentence.split('.') ) )
    # pop off last space
    sentence_split.pop()
    
    # keeps track 
    count = 0
    # traverse the sentences
    for sentence in sentence_split:
        count += 1
        if count == n:
            count = 0
            final_str.append(sentence+'.\n')
        else:
            final_str.append(sentence+'. ')

    # return the string version of the list
    return ''.join(final_str)

【讨论】：

【解决方案4】：

另一种方法：

text = 'The puppy is cute. Summer is great. Happy friday. sentence4. sentence5. sentence6. sentence7.'

out = ''
for i, e in enumerate(text.split(".")):
    if (i > 0)  & (i % 5 == 0):
        out = out + '\n'
    out = out + e + '.'
out

结果：

'The puppy is cute. Summer is great. Happy friday. sentence4. sentence5.\n sentence6. sentence7..'

【讨论】：

【解决方案5】：

使用列表理解

text = 'The puppy is cute. Summer is great. Happy friday. sentence4. sentence5. sentence6. sentence7.'
lines = text.split(".")
result = ".".join([l if i % 5 else "\n"+l for (i, l) in enumerate(lines)]).lstrip()
print(result)

【讨论】：