highperfocused/danswer
1
0
Fork 0
mirror of https://github.com/danswer-ai/danswer.git synced 2025-06-06 13:09:39 +02:00
Code Issues Packages Projects Releases Wiki Activity

Confluence fixes

Browse Source
This commit is contained in:
Weves 2025-03-06 11:42:18 -08:00
parent eaf054ef06
commit 34fc84298d
2 changed files with 13 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
backend/onyx/connectors/confluence/connector.py
View File

@ -240,7 +240,7 @@ class ConfluenceConnector(
# Extract basic page information # Extract basic page information
page_id = page["id"] page_id = page["id"]
page_title = page["title"] page_title = page["title"]
page_url = f"{self.wiki_base}/wiki{page['_links']['webui']}" page_url = f"{self.wiki_base}{page['_links']['webui']}"
# Get the page content # Get the page content
page_content = extract_text_from_confluence_html( page_content = extract_text_from_confluence_html(

15
backend/onyx/indexing/indexing_pipeline.py
View File

@ -464,12 +464,21 @@ def index_doc_batch(
), ),
) )
successful_doc_ids = {record.document_id for record in insertion_records} all_returned_doc_ids = {
if successful_doc_ids != set(updatable_ids): record.document_id for record in insertion_records
}.union(
{
record.failed_document.document_id
for record in vector_db_write_failures
if record.failed_document
}
)
if all_returned_doc_ids != set(updatable_ids):
raise RuntimeError( raise RuntimeError(
f"Some documents were not successfully indexed. " f"Some documents were not successfully indexed. "
f"Updatable IDs: {updatable_ids}, " f"Updatable IDs: {updatable_ids}, "
f"Successful IDs: {successful_doc_ids}" f"Returned IDs: {all_returned_doc_ids}"
"This should never happen."
) )
last_modified_ids = [] last_modified_ids = []