Skip to content

ABI Layer 4: prove the FFI result-code seam is injective + faithful#45

Merged
hyperpolymath merged 4 commits into
mainfrom
claude/new-session-znxgm7
Jun 28, 2026
Merged

ABI Layer 4: prove the FFI result-code seam is injective + faithful#45
hyperpolymath merged 4 commits into
mainfrom
claude/new-session-znxgm7

Conversation

@hyperpolymath

Copy link
Copy Markdown
Owner

Summary

Layer 4 (seal the ABI↔FFI seam): proves the FFI result-code encoding (resultToInt) is a sound, lossless wire encoding — decoder intToResult with resultRoundTrip, from which resultToIntInjective is derived. Distinct ABI outcomes never collide on the C wire. Complements the structural abi-ffi-gate.py.

New module *.ABI.FfiSeam (imports Types). Positive + non-vacuity controls. (Note: multToInt in Foreign is export-only so it doesn't reduce across modules — left to a future Foreign-visibility change rather than a proof cheat.)

Testing

Idris2 0.7.0 --build → exit 0, zero warnings. Adversarial rejection confirmed. build/ removed. No believe_me/postulate/sorry.

🤖 Generated with Claude Code

https://claude.ai/code/session_01A6PSzJWpRxtzGDjUCEh7Mx


Generated by Claude Code

claude and others added 4 commits June 27, 2026 22:02
Flagship semantic proof: an Alloy binary relation satisfies the `one`
field multiplicity iff every live source maps to exactly one target
(target-count = 1) — the OpenAPI "required, single-valued" property.
Sound+complete Dec, certifier into the ABI Result codes proven sound,
positive control + negative control (a one-source/two-target instance,
the Alloy counterexample, is provably rejected). Verified with idris2
0.7.0 (build clean, zero warnings) + adversarial false-proof rejection.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A6PSzJWpRxtzGDjUCEh7Mx
Adds Alloyiser.ABI.Invariants over the existing Semantics model: proves the
multiplicity-subsumption theorem oneImpliesLone (OneTargetFor s r ->
LoneTargetFor s r) plus its field-level lift, a sound+complete Dec for
LoneTargetFor, a sound ABI-Result certifier, and certifyOneEntailsLone tying
the two verdicts. Controls: zero-target source is lone (positive), two-target
source is not lone (negative), and lone does NOT entail one (strictness /
non-vacuity). Distinct from and deeper than the Layer-2 `one` theorem.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A6PSzJWpRxtzGDjUCEh7Mx
Prove the alloyiser FFI result-code encoding is sound: distinct ABI
outcomes never collide on the C wire and the integer faithfully
round-trips back to the ABI value.

- intToResult decoder (boolean Bits32 == so round-trip Refls reduce)
- resultRoundTrip : intToResult (resultToInt r) = Just r
- resultToIntInjective derived from the round-trip via justInj+cong
- positive controls (decode 0/7/8) and a machine-checked non-vacuity
  control (resultToInt Ok /= resultToInt Error)

Genuine proof only: no believe_me/postulate/assert_total/%hint.
multToInt is export-only (non-reducing across modules) so it is out of
scope here without widening Foreign's visibility.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01A6PSzJWpRxtzGDjUCEh7Mx
Signed-off-by: Jonathan D.A. Jewell <6759885+hyperpolymath@users.noreply.github.com>
@hyperpolymath hyperpolymath marked this pull request as ready for review June 28, 2026 06:48
@hyperpolymath hyperpolymath merged commit 764a7f8 into main Jun 28, 2026
6 checks passed
@hyperpolymath hyperpolymath deleted the claude/new-session-znxgm7 branch June 28, 2026 06:48
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