highperfocused/danswer
1
0
Fork 0
mirror of https://github.com/danswer-ai/danswer.git synced 2025-05-02 07:50:21 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix for citation update

Browse Source
This commit is contained in:
joachim-danswer 2025-02-03 15:07:10 -08:00 committed by Evan Lohn
parent 7ac6d3ed50
commit 3ce8923086
1 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
backend/onyx/agents/agent_search/shared_graph_utils/utils.py
View File

@ -346,7 +346,10 @@ def retrieve_search_docs(
def get_answer_citation_ids(answer_str: str) -> list[int]:
citation_ids = re.findall(r"\[\[D(\d+)\]\]", answer_str)
"""
Extract citation numbers of format [D<number>] from the answer string.
"""
citation_ids = re.findall(r"\[D(\d+)\]", answer_str)
return list(set([(int(id) - 1) for id in citation_ids]))
@ -422,10 +425,9 @@ def remove_document_citations(text: str) -> str:
Text with citations removed
"""
# Pattern explanation:
# \[\[D\d+\]\]\(\) matches:
# \[\[ - literal [[ characters
# D - literal D character
# \[(?:D|Q)?\d+\] matches:
# \[ - literal [ character
# (?:D|Q)? - optional D or Q character
# \d+ - one or more digits
# \]\] - literal ]] characters
# \(\) - literal () characters
return re.sub(r"\[\[(?:D|Q)?\d+\]\](?:\([^)]*\))?", "", text)
# \] - literal ] character
return re.sub(r"\[(?:D|Q)?\d+\]", "", text)