github/owl
1
0
Fork 0
mirror of https://github.com/camel-ai/owl.git synced 2026-03-22 14:07:17 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update document_toolkit.py

Browse Source
This commit is contained in:
Tao Sun
2025-04-22 12:19:24 +08:00
committed by GitHub
parent fbff7265a8
commit 07b811019e
1 changed files with 13 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
owl/utils/document_toolkit.py
View File

@@ -117,8 +117,19 @@ class DocumentProcessingToolkit(BaseToolkit):
return True, content
if self._is_webpage(document_path):
extracted_text = self._extract_webpage_content(document_path)
return True, extracted_text
try:
extracted_text = self._extract_webpage_content(document_path)
return True, extracted_text
except Exception:
try:
elements = self.uio.parse_file_or_url(document_path)
if elements is None:
logger.error(f"Failed to parse the document: {document_path}.")
return False, f"Failed to parse the document: {document_path}."
else:
return True, elements
except Exception:
return False, "Failed to extract content from the webpage."
else:
try: