Repository navigation
chore(deps): bump ordeal from 0.16.1 to 0.18.0 - #997
Conversation
Bumps [ordeal](https://github.com/pulseengine/ordeal) from 0.16.1 to 0.18.0. - [Release notes](https://github.com/pulseengine/ordeal/releases) - [Changelog](https://github.com/pulseengine/ordeal/blob/main/CHANGELOG.md) - [Commits](pulseengine/ordeal@v0.16.1...v0.18.0) --- updated-dependencies: - dependency-name: ordeal dependency-version: 0.18.0 dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com>
|
🔒 0.x MINOR bump — held for manual review. |
ordeal 0.17.0 marked BvTerm/BoolTerm #[non_exhaustive] (ordeal#104, motivated by synth's builds breaking on 0.x bumps twice). All six #997 CI failures (Test, Clippy, Z3, Kani, fact-spec #494, VCR-MEM-004) were this ONE mechanical E0004 break: 6 match sites in term.rs + 2 in the z3-solver-gated oracle translation (solver.rs) + Kani via synth-backend's dev-dep edge. No variants were added 0.16.1→0.18.0. Per ordeal's documented consumer policy ("conservative error, never a guess") and this repo's expand-or-loud-reject convention: - bv_rank/bool_rank/canonicalize_bv/canonicalize_bool and the z3 bv_to_z3/bool_to_z3 get PANICKING catch-alls — dead code today, and a panic can never mint a false "verified"; a silent clone in canonicalize would re-open the commuted-Mul CDCL cliff (#849 class). - The two Display impls get non-panicking <unmodeled-...> markers (diagnostics must not panic mid-report). Also rewrites the stale pin comment (still said "PINNED at 0.9.1" over an =0.16.1 pin) to state the real bump protocol. Evidence (this lane, macOS arm64): workspace 2830 passed/0 failed (197s cold); z3-solver,arm 270/270; synth-verify 268/268 with pure test-exec 35.0s on BOTH 0.16.1 and 0.18.0 (z3 suite 35s vs 38s) — identical verdicts, no 0.12-style hang; riscv + verify feature sets build and their oracle cargo gates pass; claim_check 49/49. Closes the compile break on #997. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01YJK5LZZEkV5smCY1jKn18L
|
Dependency-upgrade lane result (0.16.1 → 0.18.0, two 0.x-minor = two breaking steps by repo rule). Root cause of all six red checks: one mechanical API break. ordeal 0.17.0 sealed No behavioral or performance change measured (macOS arm64, warm builds,
What 0.17/0.18 actually are: 0.17 = sealed enums + opt-in Recommendation: upgrade. The pin existed for the 0.12 perf-hang class; 0.18.0 shows no trace of it, and 0.17's sealing + changelog-callout policy makes the next variant addition a loud, documented event instead of a surprise. Auto-merge stays OFF — merge on green CI per the lane coordinator. |
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
…le-and-assert, fail-closed tested classifier (#1022) The hold was documented but not enforced: it skipped its own enable step, posted a comment, and left an already-enabled auto-merge armed (#938 sat three days with auto-merge ENABLED and two hold comments); its label silently failed behind '|| true' (the label did not exist in the repo); and its inline shell classifier failed OPEN on an empty fetch-metadata previous-version ('case "" in 0.*)' matches nothing -> hold=false). Auto-merge was hand-disabled four times (#938, ordeal 0.9->0.12, #997, #998). Now: - scripts/dependabot_minor_hold.py is the single classifier source — cargo's compatibility rule (leftmost nonzero component), FAIL-CLOSED: 0.x-minor, 0.0.x-any, major, prerelease, downgrade, no-change, unparseable/empty metadata, and any held member of a grouped update all HOLD; only compatible patch / >=1.0-minor upgrades yield the exact 'decision=automerge' the enable step matches on (a crashed classifier leaves the output empty -> hold path). - The hold path actively runs 'gh pr merge --disable-auto' and ASSERTS the resulting state is off — the success criterion is 'auto-merge is off', never 'a comment was posted'. Label added without '|| true' (the major-bump-hold label now exists in the repo); a labelling failure is a red job. A hold is a delay: merge by hand once the FULL suite is green, incl. the separate --features z3-solver path ('Z3 Verification', already a required context). - scripts/test_dependabot_minor_hold.py: 34-case table + grouped/CLI/ output-injection tests in the required Claim Check job; 5/5 seeded classifier mutations red the suite. Step-body rehearsal (mock gh, extracted verbatim from the YAML) demonstrated the assert and the label step CAN fail — and caught a real defect at authoring (the comment body escaping the YAML block scalar). - claims.yaml SYNTH-DEPENDABOT-MINORHOLD-ENFORCED-965 pins the mechanics (disable-auto present, assert present, exact-match enable, no silent label, classifier invoked, test wired); 3/3 negative-control weakenings red the pin. Refs #965 Claude-Session: https://claude.ai/code/session_01YJK5LZZEkV5smCY1jKn18L Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
1 participant
Bumps ordeal from 0.16.1 to 0.18.0.
Release notes
Sourced from ordeal's releases.
Changelog
Sourced from ordeal's changelog.
... (truncated)
Commits
4695e85feat(release): v0.18.0 — TR-030 rivet ingestion demonstrated; cosign-signed c...e1f8c69chore(deps): Bump sha2 from 0.10.9 to 0.11.0 (#115)1a5a6eechore(rivet): TR-009 resolved with the caching rejection measured; VER-028 ve...3f3aebcchore(deps): Bump criterion from 0.5.1 to 0.8.2 (#114)3af2bdafeat: v0.18.0 replan + TR-031 executed — BMC practical-k envelope, consumer p...22c4e34chore(release): v0.17.0 — certificate as evidence + the CaDiCaL accelerator (...d37829bfeat: optional CaDiCaL accelerator backend with checker-gated LRAT (TR-004 / ...621b784chore(rivet): VER-010 executed — the benchmark numbers land in the trace (VE-...79cd181feat(v0.17.0): ordeal-cert/v1 bundle serialization (TR-025 ordeal half, #91) ...95ceb46feat: criterion latency benchmarks — the honest numbers (TR-009 item 3, #57) ...Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting
@dependabot rebase.Dependabot commands and options
You can trigger Dependabot actions by commenting on this PR:
@dependabot rebasewill rebase this PR@dependabot recreatewill recreate this PR, overwriting any edits that have been made to it@dependabot show <dependency name> ignore conditionswill show all of the ignore conditions of the specified dependency@dependabot ignore this major versionwill close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this minor versionwill close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)@dependabot ignore this dependencywill close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)