github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-08-22 10:59:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(anthropic): clean up null file_id fields in citations during message formatting (#32592)

Browse Source 
When citations are returned from streaming, they include a `file_id:
null` field in their `content_block_location` structure.

When these citations are passed back to the API in subsequent messages,
the API rejects them with "Extra inputs are not permitted" for the
`file_id` field.
This commit is contained in:
Mason Daugherty 2025-08-18 13:01:52 -04:00 committed by GitHub
parent fb74265175
commit 8042b04da6
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 17 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
libs/partners/anthropic/langchain_anthropic/chat_models.py
View File

@ -397,14 +397,23 @@ def _format_messages(
# accepted.
# https://github.com/anthropics/anthropic-sdk-python/issues/461
if text.strip():
content.append(
{
k: v
for k, v in block.items()
if k
in ("type", "text", "cache_control", "citations")
},
)
formatted_block = {
k: v
for k, v in block.items()
if k in ("type", "text", "cache_control", "citations")
}
# Clean up citations to remove null file_id fields
if formatted_block.get("citations"):
cleaned_citations = []
for citation in formatted_block["citations"]:
cleaned_citation = {
k: v
for k, v in citation.items()
if not (k == "file_id" and v is None)
}
cleaned_citations.append(cleaned_citation)
formatted_block["citations"] = cleaned_citations
content.append(formatted_block)
elif block["type"] == "thinking":
content.append(
{