andre/gpt_academic
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

换一种方式判断

Browse Source
This commit is contained in:
w_xiaolizu
2023-06-06 12:00:18 +08:00
parent 91621af63f
commit 86c8184d20
1 changed files with 10 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
toolbox.py
View File

@ -260,14 +260,16 @@ def text_divide_paragraph(input_str):
"""
将文本按照段落分隔符分割开，生成带有段落标签的HTML代码。
"""
code_blocks = re.findall('```.*?```', str(input_str), re.DOTALL)
for block in code_blocks:
input_str = input_str.replace(block, f'{{{{{{{{{{{code_blocks.index(block)}}}}}}}}}}}')
# 将除了三个反引号之间的文本块以外的 "\n" 替换为 "\n\n"
input_str = re.sub(r'(?!```)(?<!\n)\n(?!(\n|^)( {0,3}[\*\+\-]|[0-9]+\.))', '\n\n', str(input_str))
# 还原未处理的文本块
for i, block in enumerate(code_blocks):
input_str = input_str.replace(f'{{{{{{{{{{{i}}}}}}}}}}}', block)
if input_str:
code_blocks = re.findall('```.*?```', input_str, re.DOTALL)
for block in code_blocks:
input_str = input_str.replace(block, f'{{{{{{{{{{{code_blocks.index(block)}}}}}}}}}}}')
# 将除了三个反引号之间的文本块以外的 "\n" 替换为 "\n\n"
input_str = re.sub(r'(?!```)(?<!\n)\n(?!(\n|^)( {0,3}[\*\+\-]|[0-9]+\.))', '\n\n', input_str)
# 还原未处理的文本块
for i, block in enumerate(code_blocks):
input_str = input_str.replace(f'{{{{{{{{{{{i}}}}}}}}}}}', block)
return input_str