Skip to content

Finish 0.6.0 admission repairs and reconcile release readiness - #80

Closed
abogun-product wants to merge 4 commits into
drevendev:release/0.6.0from
abogun-product:zendreven/release-0.6.0-readiness
Closed

abogun-product wants to merge 4 commits into
drevendev:release/0.6.0from
abogun-product:zendreven/release-0.6.0-readiness

Conversation

@abogun-product

@abogun-product abogun-product commented Oct 21, 2026 •

Copy link
Copy Markdown

Superseded by PR #81, which includes every readiness repair here plus the 0.6 Git-history proof and executable assurance increment. The cumulative release PR #75 now has the same final source head. This closure is supersession, not a merge or an independent acceptance verdict. Historical evidence below remains available.

The 0.6 candidate could turn duplicate revisions into a stronger Integrity sample, overwrite carried failures through inadmissible timestamps, and treat contradictory/unknown CI attempts as complete history. Native source replay also accepted omission reasons without checking their predicates. This PR fixes those admission paths in the active release/0.6.0 branch and makes first-report/release documentation match current evidence.

Target: B3

Target: E4

Current/saved CI inventory now admits every row before proving branch departure, refuses duplicate revision identities, preserves an already-known commit instant, and retains carried failures through malformed generations. Attempt coverage checks exact observed identities without allocating by an untrusted maximum; UNKNOWN remains uncertain, foreign vocabulary is refused, and legitimate same-attempt transitions remain valid. Activity shares admission and exposes rejected inputs. Current revision metadata stays bound to the reconciled history.

These enforce existing accepted admission/history obligations. Rule ids, thresholds, windows, canonical contract identifiers and historical artifact bytes are preserved. No new external-check semantics or migration is introduced.

Native source omission admission now replays safe-path, policy, mode, language and ordered budget predicates. The operational proof_scope receipt explicitly marks omitted blob sizes and historical ancestry UNPROVEN: a tree authenticates the blob id, not an omitted blob's reported size. Canonical v1 members and frozen replay remain unchanged. The README first-report command was exercised through a clean 0.6.0 wheel, created the documented report, and verified it after 17 read-only requests.

Validation: complete Python 3.12/3.13/3.14 CI passed on exact head 22eb5cb90c9b72181f78ea47670ac69a3ada35d8: full unit/conformance suite, 148 CLI vectors, example verification and the replay/determinism checks declared by this release's workflow. The final suite contains 2,404 cases, including 31 new admission regressions. A complete local 2,403-case run passed before the final metadata correction; its focused regression and the final 2,404-case CI passed afterward. Clean wheel installation, core/work/source/packet examples and live first-report verification also passed. No independent acceptance is inferred from CI.

The common repair is propagated through all existing release heads without force-pushing. The cumulative #75 tree equals this readiness tree; #79's Dependabot correction is retained. Final-head evidence:

| PR | Verified head | Python 3.12/3.13/3.14 CI |

| --- | --- | --- |

| #57 | b849923 | passed |

| #59 | a41fac6 | passed |

| #64 | c98e7ed | passed |

| #69 | aa5e0e1 | passed |

| #75 | 3d0982b | passed |

| #80 | 22eb5cb | passed |

Independent exact-head review and maintainer release merges/tags remain required. B1/B7, source-history proof (#73), omitted-size proof, core GitLab/Instrument gates and operational Dependabot confirmation (#63) retain their boundaries. External deployment/scheduling stays owner-deferred in #71.

Refs #38, #57, #73, #75. Fixes #78.

Upstream PR workflow is currently action_required and still needs maintainer approval: https://github.com/drevendev/Devostasis/actions/runs/39591939816. The linked completed fork matrices are exact-head test evidence; they do not bypass that approval or independent review.

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.

1 participant