Fix semantic id for web pdfs (#1823)

This commit is contained in:
Moshe Zada 2024-07-14 21:38:11 +03:00 committed by GitHub
parent 017af052be
commit 6ee1292757
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -264,7 +264,7 @@ class WebConnector(LoadConnector):
id=current_url,
sections=[Section(link=current_url, text=page_text)],
source=DocumentSource.WEB,
semantic_identifier=current_url.split(".")[-1],
semantic_identifier=current_url.split("/")[-1],
metadata={},
)
)