Repository navigation
Typed retirement path for test_migration_debt (model + guard union + first DeleteRedundant) - #6261
Merged
Merged
Conversation
…demonstrator
Add test.retirement.model (RetirementDisposition = DeleteRedundant{covered_by}
| DeleteLowValue{reason}; TestModuleRetirement). Delete-guard gains a second
stem source: <stem>_retired.dag declarations union into the covered set feeding
both the debt roster and the delete-guard. Retire map_lookup_dual_dispatch_test.rs
(DeleteRedundant, covered by map_lookup_dual_dispatch_witness_test.dag).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
…xt import) Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
gunbai-bot
Bot
changed the title
Address claude-opus-4-7 review: require the constructor 'TestModuleRetirement {'
so a file merely importing the type can't silence the guard. Fix rustfmt line-wrap
that failed cargo fmt --all --check.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Contributor
Author
|
Addressed the REQUEST_CHANGES in 93508c3 — you were right that |
This was referenced Jul 5, 2026
briansrls
added a commit
that referenced
this pull request
Jul 5, 2026
…tirement path (delete-mass triage) (#6272) * pipeline.rs self_gen8 cluster: retire 28 of 31 fns (typed dispositions; 3 kept for def-unification lane) Delete-mass triage of the self_gen8 cluster per the test_migration_debt drain (#6261 retirement path). Ground truth measured 2026-07-05 via cargo test -p v1-compiler-tests self_gen8 -- --include-ignored: 22 passed, 9 failed - 8 of the 17 #[ignore]d fns carried stale red-on-main reasons and pass on main today. - 22 green fns -> DeleteRedundant: seed-corpus pattern families (homonymous type names, variant-specific imports, kernel-ambient types, direct type imports) are byte-pinned by RegenVerifyGate (regen_stage0 --verify, #5873) plus the emitted-seed rustc/clippy gates; beyond-corpus synthetic shapes (parametric-alias RHS, opaque parametric decls, proxy chains, wildcard imports) are executed by the v2 self-host fresh-emit lane over dag/std's real instances with deficits enumerated in the 1667-error receipt (#6253/#6258). An #[ignore]d pin runs in no gate - zero delivered signal. - 6 red fns -> DeleteLowValue: #[ignore] since #5427, never run in any gate, desired-but-absent seed-emitter behavior on synthetic fixtures; deficit ownership is the fresh-emit lane. - 3 red fns KEPT (stays_unemitted alias-to-opaque trio): dag/gunbc/plans/dag_v2_defork_audit.dag claims them for the def-unification lane (node://adhoc-9d2bb9c3-e7b) - premise flips, not deletes. Typed receipt: dag/test/retirement/pipeline_self_gen8_retired.dag (two TestModuleRetirement rows). Receipt-only w.r.t. the delete-guard - the contributed stem pipeline_self_gen8 matches no v1 test file and pipeline.rs itself survives, so its eventual file deletion still requires its own coverage. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * Repoint kept trio's #[ignore] reasons to their owning lane (def-unification, node://adhoc-9d2bb9c3-e7b) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Brian Searls <briansrls@gunb.ai> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
gunbai-bot Bot
pushed a commit
that referenced
this pull request
Jul 5, 2026
…ment wave-1, cutover pre-reqs, CI step-budget posture) Authority refresh after 2026-07-05's merges, ROADMAP.md regenerated via main_wet: spine marks #6255/#6262/#6243 landed and NOW = post-#6243 re-measure -> cutover #1 (pre-reqs #6243 + #6270 backfill met, defensive sweep pending); regen-cutover carrier gets the root-cause update (1482/1667 = one field_is_boxed predicate bug, receipt must be re-earned post-retraction); test-migration carrier gets #6261 + wave-1 (912->881 fns); ground truth: HAND roster 7->9 (dag_collect pair, #6262), GENERATED 92->90, CI INTERIM restated for #6273's loud step budgets. ci.yml: fold the 1-byte trailing-newline regen delta so committed == generated. All roadmap_authority_test pins verified against the regenerated projection (ALL PINS OK). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
briansrls
added a commit
that referenced
this pull request
Jul 5, 2026
…tirement wave-1, CI step-budget posture (#6287) * WIP: continue work on v1 burn down * WIP: continue work on v1 burn down * WIP: continue work on v1 burn down * WIP: continue work on v1 burn down * Drop get-form revert per operator collision resolution (#6255 owns the get arm); regen ROADMAP projection Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * WIP: continue work on v1 burn down * WIP: continue work on v1 burn down * Fix roadmap_authority_test pins: track the 2026-07-05 ground-truth refresh (cursor RC) + repoint pre-existing-stale charter-adjacency pin witness_b_ordered_interleaving pinned the pre-refresh prose (Ground truth 2026-07-01 date + old walls sentence); updated to the refreshed authority. Also fixed a pin this PR did NOT stale: the charter->core-design-rule adjacency has been red on main since [host-converge inventory] was appended after [charter] — repointed to the real paragraph tail. Discriminator date kept in lockstep. All pins verified against the regenerated ROADMAP.md (byte-identical projection). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * WIP: continue work on v1 burn down * §1 roadmap receipts: today's landings (emitter mass fix #6243, retirement wave-1, cutover pre-reqs, CI step-budget posture) Authority refresh after 2026-07-05's merges, ROADMAP.md regenerated via main_wet: spine marks #6255/#6262/#6243 landed and NOW = post-#6243 re-measure -> cutover #1 (pre-reqs #6243 + #6270 backfill met, defensive sweep pending); regen-cutover carrier gets the root-cause update (1482/1667 = one field_is_boxed predicate bug, receipt must be re-earned post-retraction); test-migration carrier gets #6261 + wave-1 (912->881 fns); ground truth: HAND roster 7->9 (dag_collect pair, #6262), GENERATED 92->90, CI INTERIM restated for #6273's loud step budgets. ci.yml: fold the 1-byte trailing-newline regen delta so committed == generated. All roadmap_authority_test pins verified against the regenerated projection (ALL PINS OK). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * WIP: continue work on v1 burn down * Fix cursor RC on #6287: 5-dissolve-patches carrier now matches ground truth (9 HAND files / ~22.6k; dag_collect pair slotted into the drain order with its content_hash dissolution) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> * WIP: continue work on v1 burn down * Fix cursor RC round 2 on #6287: roster receipts re-grounded against the live registry authority (7 HAND / 92 GENERATED) The 9-HAND/90-GENERATED claims described the state between #6262 and #6239 — #6239 returned the dag_collect pair to GENERATED (verified: regen_stage0.rs GENERATED includes the pair, HAND count 7, stage0_emit_model witness pins 92). Ground truth now records the full history (#6262 HAND-only -> #6239 GENERATED again), the drain carrier drops the pair as a HAND target (its content_hash grounding stays tracked by the .dag dissolve-on rows), and terminal collapse is back to 92. Consistency proven by assertion over the regenerated projection: no 90/9-file/22.6k string survives, exactly one 7-file count, pins intact. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> --------- Co-authored-by: Brian Searls <briansearls1@gmail.com> Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
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
1 participant
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.
Typed retirement path for
v2.lens.test_migration_debtFirst deliverable of the test-triage lane (operator 2026-07-05, scrutinize-first): a typed retirement path so a v1
#[test]module can leave the debt roster three ways — migrate (existing floor-witness path), delete-as-redundant, or delete-as-low-value — instead of migrate-only.The gap
The delete-guard (
test_migration_delete_guard_holds, cli_run.rs) fail-closes: a#[test]-bearing v1 module can only be deleted once an exact-stem floor*_test.dagwitness exists. That is the migrate path. It has no path for a module whose behavior is already covered by a differently-named witness (redundant) or is transient v1-Rust scaffolding (low-value) — both would trip the guard forever.The mechanism (reuses the existing stem-glob — no new mechanism)
.dagauthoritytest.retirement.model(dag/test/retirement/model.dag):RetirementDisposition = DeleteRedundant { covered_by: String } | DeleteLowValue { reason: String },TestModuleRetirement { module, disposition }. Migrate is not a retirement row — the floor witness self-evidences it (no parallel ledger, §2).<stem>_retired.dagdeclaration the delete-guard globs, symmetric with floor*_test.dagstems. Coverage =floor_stems ∪ retired_stems; the same union feeds both the debt roster (retired modules drop out → shrink-only ratchet stays valid) and the delete-guard (retired modules are deletable).covered_byis fail-closed by construction (for theDeleteRedundantarm): a retirement imports the covering witness's real decl and stores its value, so a dangling ref is a compile-clean RED.DeleteLowValue { reason }is prose (unstructurable residue, DESIGN §5/§6). TheDeleteRedundantarm is defined but not constructed in this PR (likeRatchetForeverinconstruction_justification, an unconstructed variant is safe); itscovered_bydemonstrator lands in triage tranche 1 on a genuinely fully-covered module.Guard delta is minimal (#6046)
The cli_run.rs change is exactly a second stem source unioned into the existing covered set —
test_migration_retired_stems()(glob*_retired.dag, require the file declaresTestModuleRetirementso an empty stub can't silence the guard) +test_migration_covered_stems()= floor ∪ retired — and two call-sites (build_test_migration_debt_report,test_migration_delete_guard_uncovered_deletes_inner) repointed at the union. No other net-new logic.Demonstrator:
map_lookup_dual_dispatch_test.rs→ DeleteLowValueA self-declared RESIDUAL of the 5-test-migration lane: 4 of its 6 tests already migrated to
dag/test/claim/map_lookup_dual_dispatch_witness_test.dag. The 2 remaining are not redundant-by-coverage (so notDeleteRedundant, per the review below): (1) an active v1-interpreter source-scan the module itself documents as "dissolving with the v1 interpreter" — a transient v1-Rust scaffold complementary to the migrated witnesses, dropped as low-value at v1 sunset; (2) an#[ignore]d inert test. Retired viadag/test/retirement/map_lookup_dual_dispatch_retired.dagwith an explicitDeleteLowValue { reason }. Debt: 73 → 72 modules.Receipts (LOCAL — CI floor is dark, 10-min timeout; operator merges on local receipts)
gunbc compile --source-root dag --source-root src/v2 --target dag— freshly-built gunbc, 0 errors; the newdag/test/retirement/*.dagresolve/typecheck. (The stale installed/usr/local/bin/gunbcreports 2 unrelatedinert_lens_*builtin-scope errors — a binary-version skew, not this change; a current-build compile is clean.)cargo test -p v1-compiler --lib test_migration_debt_tests→ 6 passed / 0 failed):retired_stem_is_covered_but_not_a_floor_stem(discriminating — the stem is covered only via the retirement, not any floor witness),retired_module_is_not_debt, anddelete_guard_holds_with_no_v1_test_deletions_in_diffnow green because the deletion ofmap_lookup_dual_dispatch_test.rsinorigin/main...HEADis covered by the retirement.delete_guard_holdsreturned false (uncovered delete) — the guard reds without the retirement, greens with it.covered_byis bound to an imported decl, so a dangling covering import is a compile-clean resolve error (same mechanism as the two unrelatedinert_lens_*not found in scopeerrors from the stale installed binary).Review addressed
DeleteRedundant, but the module's active source-scan test is complementary to the migrated witnesses, not covered —covered_byoverstated coverage (§3/§5 semantic dilution). Fixed by reclassifying toDeleteLowValuewith an explicit reason (the reviewer's first recommended option); the scan is transient v1-Rust scaffolding the module itself documents as dissolving with the v1 interpreter.Follow-on (noted, not in this PR)
dag/test/retirement/(the receipt).covered_byto an un-fakeable resolving reference for every row (a lens over retirement rows) is reflection-gated (gunbc#5364); today it is fail-closed at the demonstrator level + convention + review..daglens, the per-file_retired.dagstems collapse into a roster module.