Kevin Hu
commited on
Commit
·
7fcb06c
1
Parent(s):
17b116d
fix bug of regx (#1703)
Browse files### What problem does this PR solve?
#1689
### Type of change
- [x] Bug Fix (non-breaking change which fixes an issue)
- rag/nlp/__init__.py +1 -1
rag/nlp/__init__.py
CHANGED
|
@@ -344,7 +344,7 @@ def make_colon_as_title(sections):
|
|
| 344 |
if txt[-1] not in ":：":
|
| 345 |
continue
|
| 346 |
txt = txt[::-1]
|
| 347 |
-
arr = re.split(r"([。？！!?;；]|
|
| 348 |
if len(arr) < 2 or len(arr[1]) < 32:
|
| 349 |
continue
|
| 350 |
sections.insert(i - 1, (arr[0][::-1], "title"))
|
|
|
|
| 344 |
if txt[-1] not in ":：":
|
| 345 |
continue
|
| 346 |
txt = txt[::-1]
|
| 347 |
+
arr = re.split(r"([。？！!?;；]| \.)", txt)
|
| 348 |
if len(arr) < 2 or len(arr[1]) < 32:
|
| 349 |
continue
|
| 350 |
sections.insert(i - 1, (arr[0][::-1], "title"))
|