Repository navigation
Conversation
Member
|
No need for the test. Add comment along the lines of "previous versions accepted trailing data" to help anyone that relied on it. |
dxbjavid
force-pushed
the
goldilocks-point-trailing-data
branch
from
October 5, 2026 19:17
dfebe5d to
317278c
Compare
Contributor
Author
|
makes sense, dropped the test and added a short comment at the length check noting that previous versions accepted trailing data. |
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
This comment was marked as resolved.
This comment was marked as resolved.
Sorry, something went wrong.
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
goldilocks.FromBytes, and Point.UnmarshalBinary which delegates to it, only required the input to be at least fp.Size+1 bytes, so any trailing bytes after a 57-byte point encoding were quietly ignored and a given point could then be decoded from more than one byte string. MarshalBinary only ever emits exactly fp.Size+1 bytes, and the same trailing-data laxity has already been tightened for the group elements, the bls12381 G1/G2 points, dleq proofs and ml-dsa signatures, so it seemed worth making the length check here exact as well rather than a lower bound. I've added a small regression test next to the existing canonical-encoding checks.