Repository navigation
fix(retain): fail a chunk whose reply is not JSON, even 'no facts' prose (#5565) - #5615
Merged
Merged
Conversation
…ose (#5565) Reverts the prose 'no facts' acceptance from #5516. Its keyword check also matched failure replies ('no facts could be extracted due to an error', 'Rate limited, nothing to extract right now.'), so a broken chunk reported success with 0 facts. A reply that is not JSON now always fails the chunk; only a JSON reply can say there are zero facts.
This was referenced Oct 16, 2026
Merged
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Closes #5565.
#5516 (for #5443) counted a plain-text "no facts" reply from the model as zero facts. The keyword check behind it also matched failure replies, for example
no facts could be extracted due to an errororRate limited, nothing to extract right now., so a broken chunk reported success with 0 facts.This removes that rule. A reply that is not JSON now always fails the chunk, as it did before #5516. Only a JSON reply can say there are zero facts.
_is_no_facts_proseand theJSONDecodeErrorhandler in_extract_facts_from_chunk.Trade-off: models that answer "no facts" in plain text (#5443) fail the retain again.