zhuchaowe/maxkb
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: 上传的文档中 未智能处理空白段落

Browse Source
This commit is contained in:
shaohuzhang1 2024-02-19 11:29:24 +08:00
parent 948d0c4f3c
commit 7eb18fbf30
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
apps/common/util/split_model.py
View File

@ -332,6 +332,8 @@ class SplitModel:
""" """
result_tree = self.parse_to_tree(text.replace('\r', '\n'), 0) result_tree = self.parse_to_tree(text.replace('\r', '\n'), 0)
result = result_tree_to_paragraph(result_tree, [], []) result = result_tree_to_paragraph(result_tree, [], [])
# 过滤段落内容不为空字符串的数据
result = [item for item in result if 'content' in item and len(item.get('content').strip()) > 0]
return [{**item, 'title': item.get('title').replace("#", '') if 'title' in item else ''} for item in result] return [{**item, 'title': item.get('title').replace("#", '') if 'title' in item else ''} for item in result]