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)
以下是修改后的版本:
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)
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)
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))
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))
下面是一个简单的函数,它在第五句的末尾添加了一个换行符
以下是修改后的版本:
试试这个:
使用正则表达式。在“[not.]后跟“.”的5个匹配项之后添加\n
更高级的正则表达式句子匹配器,通过匹配结尾标点来处理缩写和其他标点。
参考资料:https://mikedombrowski.com/2017/04/regex-sentence-splitter/
注意:仍有一些边缘情况无法使用此选项,例如T.V.后面跟着Mr.需要双空格来表示单独的句子。带有句子的引文将被拆分。等等
任何比这更复杂的东西,您都可能需要查看语言处理工具包
相关问题 更多 >
编程相关推荐