mirror of
https://github.com/danswer-ai/danswer.git
synced 2025-07-23 11:26:03 +02:00
Fix semantic id for web pdfs (#1823)
This commit is contained in:
@@ -264,7 +264,7 @@ class WebConnector(LoadConnector):
|
|||||||
id=current_url,
|
id=current_url,
|
||||||
sections=[Section(link=current_url, text=page_text)],
|
sections=[Section(link=current_url, text=page_text)],
|
||||||
source=DocumentSource.WEB,
|
source=DocumentSource.WEB,
|
||||||
semantic_identifier=current_url.split(".")[-1],
|
semantic_identifier=current_url.split("/")[-1],
|
||||||
metadata={},
|
metadata={},
|
||||||
)
|
)
|
||||||
)
|
)
|
||||||
|
Reference in New Issue
Block a user