Skip to content

FV CI integration#2

Merged
mitschabaude merged 21 commits into
zksec/clean-integrationfrom
fv-ci-integration
Apr 17, 2026
Merged

FV CI integration#2
mitschabaude merged 21 commits into
zksec/clean-integrationfrom
fv-ci-integration

Conversation

@gio54321
Copy link
Copy Markdown
Collaborator

No description provided.

gio54321 and others added 21 commits March 17, 2026 12:20
Use fully-qualified paths for trait methods that aren't in scope, and
Self:: for the struct's own field. Silences -D warnings in the docs
job.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tells contributors how to regenerate the Lean files when drift is
detected, so CI failures are self-explanatory.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
check_all compares Ragu.lean against generated_ragu_root, but export_all
never wrote it. When a target was added or removed, running export alone
could not make check pass, which made the existing remediation hint
misleading.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Re-adds point_negate.rs to lean_extraction (removed in 513effa) and
introduces the hand-written Instances/Point/Negate.lean wrapper that
ties the autogenerated operations/output to the Circuits.Point.Negate
formal circuit via FormalCircuit.isGeneralFormalCircuit.

Ragu.lean is regenerated to import the new Instances.Point.Negate.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Matches the Lean module name (AddIncomplete) and aligns with the
point_double/Double and point_negate/Negate conventions.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replaces an assert_eq! with a proper Result return, matching the error
handling style of the rest of the function.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The dedicated variant matches what the error actually is (a fixed-length
vector mismatch) and includes expected/actual in structured form.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The doc claimed the variant held a first_idx, but there is none. Indices
are implicit in the op-sequence position, matching Lean's
Operation.witness representation.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@mitschabaude mitschabaude merged commit f8134d6 into zksec/clean-integration Apr 17, 2026
14 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants