pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

use minio to store uploaded files; build dialog server; (#16)

Browse Source 
* format code

* use minio to store uploaded files; build dialog server;
This commit is contained in:
KevinHuSh
2023-12-25 19:05:59 +08:00
committed by GitHub
parent d4fd138954
commit 3245107dc7
13 changed files with 520 additions and 134 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
python/nlp/huchunk.py
View File

@ -372,7 +372,7 @@ class PptChunker(HuChunker):
def __call__(self, fnm):
from pptx import Presentation
ppt = Presentation(fnm)
ppt = Presentation(fnm) if isinstance(fnm, str) else Presentation(BytesIO(fnm))
flds = self.Fields()
flds.text_chunks = []
for slide in ppt.slides:
@ -396,7 +396,9 @@ class TextChunker(HuChunker):
@staticmethod
def is_binary_file(file_path):
mime = magic.Magic(mime=True)
file_type = mime.from_file(file_path)
if isinstance(file_path, str):
file_type = mime.from_file(file_path)
else:file_type = mime.from_buffer(file_path)
if 'text' in file_type:
return False
else: