estel/dify
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix customer spliter character (#1915)

Browse Source 
Co-authored-by: jyong <jyong@dify.ai>
This commit is contained in:
Jyong
2024-01-04 16:21:48 +08:00
committed by GitHub
parent a798dcfae9
commit 4a3d15b6de
2 changed files with 12 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
api/core/data_loader/file_extractor.py
View File

@@ -65,7 +65,8 @@ class FileExtractor:
elif file_extension == '.pdf':
loader = PdfLoader(file_path, upload_file=upload_file)
elif file_extension in ['.md', '.markdown']:
loader = UnstructuredMarkdownLoader(file_path, unstructured_api_url)
loader = UnstructuredMarkdownLoader(file_path, unstructured_api_url) if is_automatic \
else MarkdownLoader(file_path, autodetect_encoding=True)
elif file_extension in ['.htm', '.html']:
loader = HTMLLoader(file_path)
elif file_extension == '.docx':
@@ -84,7 +85,8 @@ class FileExtractor:
loader = UnstructuredXmlLoader(file_path, unstructured_api_url)
else:
# txt
loader = UnstructuredTextLoader(file_path, unstructured_api_url)
loader = UnstructuredTextLoader(file_path, unstructured_api_url) if is_automatic \
else TextLoader(file_path, autodetect_encoding=True)
else:
if file_extension == '.xlsx':
loader = ExcelLoader(file_path)