|
2 | 2 |
|
3 | 3 | ## Resolved |
4 | 4 |
|
5 | | -- [x] Installer fork defaults and complete AST/watcher/macro/COUNT integration |
6 | | - landed via PR #12; Windows hook/macOS deadline repairs landed via PR #13 |
7 | | - (`bbabc073`). Historical Jenkins evidence remains revision-specific. |
8 | | - |
9 | 5 | - [x] Release-tooling implements deferred npm/PyPI/public publication, |
10 | 6 | removes Claude automation, enables CodeRabbit, adds Linux-only prerelease |
11 | 7 | selection. Package migration is deferred until exact fork assets and checksums |
|
27 | 23 | `cbm_operation_session_state_free()` before freeing each session. Valgrind |
28 | 24 | identified the omission through `test_daemon_application_free_releases_live_watch_once` |
29 | 25 | as 80 direct + 107 indirect lost bytes. |
| 26 | +- [x] Validated local `release-tooling` through Jenkins job #34 (full gate |
| 27 | + passed at `37dd63ba`). PR #12 merged with installer fix `c9841610` |
| 28 | + (install.sh and install.ps1 default to `ngallodev/codebase-memory-cli`). |
| 29 | + Resolved thread `PRRT_kwDORjrg5c6n5KE9` 2026-10-01. |
| 30 | +- [x] Validated complete AST slice `77ab4bea` / `7707c9d7` in shared |
| 31 | + release-tooling Jenkins job #34 (full gate passed at `37dd63ba`); all |
| 32 | + definition and auxiliary stacks ported to main. |
| 33 | +- [x] Added opt-in, single-run index metrics for performance diagnosis via |
| 34 | + `index --metrics-out` flag (`0f5a84bc`). Coverage in src/operations/index.c |
| 35 | + and tests/test_index_resilience.c. Merged via PR #12. |
| 36 | +- [x] Fixed the PR #12 Windows/macOS CI failures in `bbabc073` (PR #13). |
| 37 | + The Windows hook-lifecycle test now uses the fork's installed binary name |
| 38 | + (`tests/test_cli.c`), and the macOS connection-cap test gets a 30 s server |
| 39 | + hello deadline (`tests/test_daemon_runtime.c`). Jenkins #35 and full GitHub |
| 40 | + CI passed. The pre-commit hook also clang-formatted both files. |
30 | 41 |
|
31 | 42 | ## Open |
32 | 43 |
|
33 | | -- [ ] Reconcile interactive upstream assessment and independent completeness |
34 | | - audit (`CBM-UPSTREAM-ASSESS-20261002`, `CBM-UPSTREAM-VERIFY-20261002`). |
35 | | - Both workers running in pre-indexed isolated trees; frozen MCP `96c3f41c`, |
36 | | - CLI `9cb5cc21`, 19 new / 367 ancestry-differing nonmerge commits. Canonical |
37 | | - applicability/vertical classification and verification remain pending. |
38 | | - |
39 | | -- [x] Closed daemon/cache interactive workers after verifying committed clean |
40 | | - candidates `028a4bfb` / `fa5a180b`; sealed partial run receipts, preserved |
41 | | - branches/bundles/evidence and removed only their two worktrees. |
42 | | -- [ ] Review/integrate those candidates and run parent-scheduled Jenkins; |
43 | | - daemon AC-005 additionally needs a direct CLI diagnostic assertion. |
44 | | - |
45 | | -- [x] Specified all three selected verticals and prepared external interactive |
46 | | - GPT-6-Luna Agent Runs at frozen base `51aa7a86`, each in an independent |
47 | | - worktree with full permissions/approval never. See the latest 308 dispatch record. |
48 | | -- [x] Launched all three interactive GPT-6-Luna workers in Herdr's right-hand |
49 | | - column (`wC:p5`, `wC:p6`, `wC:p7`), bound to durable external Agent Runs. |
50 | | - Full permissions/approval never verified; all three acknowledged steering |
51 | | - and are observed working. Implementation/Jenkins/review remain pending. |
52 | | - |
53 | | -- [ ] Next upstream slice 1: adapt `ed76cf8e` daemon cache-ownership safety |
54 | | - across eager shutdown and quiescence callback; preserve foreign/unknown |
55 | | - sessions and existing exclusive mutation authority. Jenkins only. |
56 | | -- [ ] Next upstream slice 2: adapt `8093d10b` + `ea6fa49f` as a complete |
57 | | - binary `.res` / ReScript admission, traversal, scanner and checksum slice. |
58 | | -- [ ] Next upstream slice 3: adapt `46015320` to preserve `_config.db` and |
59 | | - `_cross_repo.db` across project-index list/count/reset; share exact-name |
60 | | - classification with cross-repo enumeration. |
61 | | -- [ ] Complete full 348-commit semantic disposition and fresh exact-worktree |
62 | | - graph indexes for both repos; this top-three selection is a bounded shortlist. |
63 | | -- [ ] Finish `manage-adr` alias/help and explicit outline parity: existing ADR |
64 | | - candidate failed Jenkins lint before tests and is not integrated. |
| 44 | +- [ ] Complete semantic reconciliation of all frozen upstream merges and |
| 45 | + constituent commits. Sol R2 inventories all 923 merges / 2,429 non-merges; |
| 46 | + 311 merges / 367 non-merges differ by ancestry. It reconciles the older 87 |
| 47 | + canonical rows, but 356 non-merge whole-commit remainders and all 311 merge |
| 48 | + dispositions remain unresolved. Census, patch ID and titles are not coverage. |
| 49 | + See [the canonical source audit](docs/UPSTREAM_308_CANDIDATE_AUDIT_20260912.md). |
| 50 | +- [ ] Finish independent reconciliation of the Luna team evidence. All R3 |
| 51 | + handoffs were partial; all 40 declared checksums and exact 111/170/86 assignments |
| 52 | + were verified and archived. R4 successors continue interactively in Herdr |
| 53 | + `wC:pB` / `wC:pC` / `wC:pD`. The 2026-10-02T23:53Z checkpoint contains 54 |
| 54 | + non-unresolved worker claims and 313 unresolved claims; no whole-commit coverage |
| 55 | + or completion is inferred. Parent retains 356 whole non-merge / 311 merge |
| 56 | + remainders pending exact source/test review. Full hashes are in the canonical |
| 57 | + ledger; the [consolidated report](docs/UPSTREAM_RECONCILIATION_REPORT_20261002.md) |
| 58 | + records provenance, source corrections, scope and validation limits. |
| 59 | + Parallel split followup: `CBM-RECON-ADAPTER-LOW-20261003` ran Luna low in |
| 60 | + Herdr `wC:pE` on 16 explicitly released, previously unreviewed runtime hashes; |
| 61 | + all 16 were source-reviewed, 8 closing as adapter-only and 8 retaining |
| 62 | + shared-semantic remainders. The consolidation auditor verified the run's |
| 63 | + 67-file manifest, confirmed every changed path is `src/mcp/**` or |
| 64 | + `tests/test_mcp.c` (that subtree exists upstream and is absent from |
| 65 | + production and both QA trees), and confirmed the 16 are upstream ancestors |
| 66 | + absent from those three CLI trees. |
| 67 | + After the Phase1 transfer, original runtime retains 39 hashes; all 367 hashes still have one |
| 68 | + effective primary owner. The transfer, acknowledgement, index and launch |
| 69 | + receipts are in the new run's evidence. Completed work was not reassigned. |
| 70 | + Phase1 successor plan now transfers 115 still-unreviewed runtime hashes to |
| 71 | + three prepared DeepSeek Flash assignments (39/26/50 disjoint groups), preserving the |
| 72 | + exact releases and completed work. Unstarted low2 was retired; corrected low1 |
| 73 | + finished partial with detailed handoff and its pane closed. See the canonical |
| 74 | + Phase1 plan. Native assignment delivery failed: all three returned predecessor |
| 75 | + reports, including after persisted steering and direct follow-ups. Sessions |
| 76 | + were stopped; zero assigned comparisons or correlated worker acknowledgements |
| 77 | + were established by that failed attempt. Interactive recovery now runs Codex |
| 78 | + `--model deepseek-flash` in Herdr `wC:pF` / `wC:pG` / `wC:pH`, with correct |
| 79 | + worker-issued run/cwd/count launch acknowledgements and actual process Jev-token |
| 80 | + inheritance verified. The same 115 hashes remain the comparison scope; see the |
| 81 | + canonical interactive recovery section. No completion/acceptance is inferred. |
| 82 | +- [ ] Review additional bounded missing behaviors: malformed confidence parsing |
| 83 | + `62b44519c68b403ea44d9a85ba8ee74cddee66e6` (first additional localized candidate), |
| 84 | + route handler position `592894a4387a50e1d128ff6a2695ff48f8cc32a5`, late annotation |
| 85 | + arguments `c36b4fbc446f9085704a6073b81ec743fd4180fc`, and client-registry selection |
| 86 | + `1e25d962d69999438229b0562d44ea60e97f5456`. Group route followups with slice 5; |
| 87 | + review full tests/dependencies before broadening implementation scope. |
| 88 | +- [ ] Resolve daemon memory policy before selecting active-job division |
| 89 | + `21591d51168d26f931f64d62edc36013cd1f6249`. Fork uses fixed worker-capacity |
| 90 | + division; upstream active divisor is not represented. Review aggregate |
| 91 | + reservation invariants, separate Windows job-memory capability hunks from |
| 92 | + MCP adapter policy in merge `2b8bd0c065c1d2728994fddc1683a66f70745748`. |
| 93 | +- [x] Daemon/cache/ReScript candidates and ADR alias/outline are integrated |
| 94 | + in local QA `559214af28ce2d6f5d6ec86662f6871316d28b45`, per acknowledged |
| 95 | + parent steering and source comparison. Candidate implementation is separate |
| 96 | + from runtime validation, acceptance and production-main integration. |
| 97 | +- [ ] Finish parent review and shared Jenkins validation for that QA revision |
| 98 | + and its later repairs. No current Jenkins success or acceptance is established |
| 99 | + by Sol R2. The pre-commit candidate retains formatting-only behavior; candidate |
| 100 | + ancestry does not establish production Git-child environment isolation. |
| 101 | +- [ ] NEW upstream slice 1: Cypher OPTIONAL relationship fallback growth, |
| 102 | + `2fabaa2b877986bdededb4f4f78e9a31014c6fdf`; existing append helper avoids |
| 103 | + the per-binding overflow. Jenkins `cypher`. |
| 104 | +- [ ] NEW upstream slice 2: SQLite zero-cell trailing interior-page packing, |
| 105 | + `e7324bb6904c6d0a295b956e83c8585219198358`. |
| 106 | + Jenkins `sqlite_writer,graph_buffer,pipeline`. |
| 107 | +- [ ] NEW upstream slice 3: Swift scanner shift widths, regression and integrity, |
| 108 | + `3f831936b5f3012d13175e7c97326165efecefad` -> |
| 109 | + `a2ea711c29560511eb8467bea7ad309ebf500d6e` -> |
| 110 | + `e8204092333ee1c38cd681a81abb36f125288894`. |
| 111 | + Jenkins `language,extraction,grammar_regression` and vendored integrity; |
| 112 | + preserve QA's ReScript manifest/checksum row. |
| 113 | +- [ ] NEW upstream slice 4: ancestor ignore chain with correct anchoring, |
| 114 | + `c8184ed94887a6b5e130726885b44a07c63a47b7` -> |
| 115 | + `282fbfcdabfec79c19c36f5bec5f75c22eb36df5`. |
| 116 | + Jenkins `discover,gitignore,git_context,pipeline`. |
| 117 | +- [ ] NEW upstream slice 5: deterministic route selection and >64 proto services, |
| 118 | + `6b48097471e8d1a3e5cf16447304899a21e12d7a`; adapt route hunks without |
| 119 | + importing upstream memory core or unrelated tests. Jenkins `pipeline,parallel`. |
| 120 | + Ranking is bounded by inspected source; unresolved work can change priorities. |
| 121 | +- [ ] Evaluate GitHub ancestry reconciliation after semantic review. Adapted |
| 122 | + commits have different IDs; documentation cannot alter GitHub's behind count. |
| 123 | + An upstream ancestry merge requires explicit review of missing/excluded changes |
| 124 | + and merge-resolution behavior. Do not use an `ours` merge to conceal unresolved |
| 125 | + work, rewrite published history, or declare the fork globally caught up. |
| 126 | +- [ ] Clarify CLI coverage array syntax and missing-path results. At dev CLI / |
| 127 | + project `tmp-cbm-upstream-sol-20261002`, generation `2026-10-02T17:30:32Z`, |
| 128 | + JSON/comma `--paths` values became literal nonexistent paths with |
| 129 | + `no_recorded_issue` / `freshness=missing`. Repeated `--paths` / `--scopes` |
| 130 | + flags work. Document this spelling and make missing paths prominent. |
| 131 | +- [ ] Align trace truncation with documented pagination. `extract_func_def` |
| 132 | + both-direction trace at limit 100 returned 127 total callees and truncation |
| 133 | + without a next cursor; limit 300 returned all 127 callees / four callers. |
| 134 | + Emit a cursor or document the limit restriction. Exact receipts are in Sol R2 |
| 135 | + evidence; these are observed CLI issues, with no source fix in this audit. |
65 | 136 | - [ ] Retain validation gaps: live watcher scheduler integration and AST |
66 | 137 | allocation-failure injection are not established by completed helper tests. |
67 | 138 | - [ ] PR #14 at `1ce4e164` has failing GitHub diagnostic, macOS LSan, |
|
83 | 154 | CLI-only process model before porting. |
84 | 155 | - `e5c17de6` and `b04f5450` coverage correctness/performance; manually |
85 | 156 | reconcile against the fork's changed coverage implementation. |
86 | | - - MCP-only output/search/pagination/compact-output commits are intentionally |
87 | | - excluded from this backlog item. |
| 157 | + - Exclude direct MCP-only adapter imports at the source seam; shared neutral |
| 158 | + capability parity still requires source verification. Historical exclusions |
| 159 | + are superseded by the canonical audit where mixed hunks were found. |
| 160 | + |
| 161 | +- [ ] Assess a lightweight path for store-only test execution. Run just the |
| 162 | + store test binaries after shared prerequisites are built, while keeping the |
| 163 | + existing incremental suite selection and full-suite gates intact. First |
| 164 | + inspect the runner's configuration and targets to determine whether this is |
| 165 | + already supported. The `cbm-upstream-count-20260929` focused run passed 96 |
| 166 | + store tests, but a cold worktree spent several minutes compiling shared |
| 167 | + prerequisites including `lsp_all.c`. |
88 | 168 |
|
89 | 169 | - [ ] Classify residual Heaptrack allocations in the neutral daemon test path. |
90 | 170 | - Baseline evidence: Heaptrack reported 91 leaked allocations across the |
|
111 | 191 | - [ ] Reconcile the GPT-6.1-Sol full upstream merge/constituent audit |
112 | 192 | `CBM-UPSTREAM-SOL-20261002-R2`, including ungrouped verticals and the next |
113 | 193 | five recommended slices. Do not count running audit work as coverage. |
| 194 | + |
| 195 | +- [ ] Complete DeepSeek Phase1 independent source review. All115 assignments and |
| 196 | + 342 evidence-manifest entries pass integrity checks; six bounded semantic |
| 197 | + questions were reviewed in three batched Jev requests. Documentary result is |
| 198 | + changes requested: narrow runtime `9724d903046626df0640ced218259cb58070e4e8` |
| 199 | + committed-counter/budget claim, preserve operations |
| 200 | + `b04f54506ab42eddcbd89232a0d47d8355edd596` mechanism remainder and |
| 201 | + `f6afd6138e605365a76d065d04521ec2037bed3d` title-policy question, and classify |
| 202 | + integration `00411ce7bc41995783bfb16aada673515585446d` changed Linux alias |
| 203 | + behavior as missing. Other109 hashes remain independently unreviewed; integration |
| 204 | +29 partial rows plus1 unresolved row require function-level work. Exact hashes, |
| 205 | +requests, full distributions and source corrections are in the canonical308 review |
| 206 | +section and consolidated report; machine crosswalk stays handoff evidence. No |
| 207 | +completion/acceptance or semantic-closure counter advanced. |
| 208 | + |
| 209 | +- [x] Gather Luna low pilot and successor source evidence; independent review is |
| 210 | + recorded in canonical308. Remaining16 full-hash whole-commit reviews are tracked |
| 211 | + below; former launch-only20-unverified snapshot is superseded. |
| 212 | + |
| 213 | +- [ ] Complete Luna medium runtime20 whole-commit source review: documentary |
| 214 | + changes requested; manifest31 passes, two bounded source dispositions verified, |
| 215 | +18 exact full-hash remainders in parent evidence/medium20-independent-review/review.json. |
| 216 | + Group262ab012 telemetry with3507d955 Windows OS hard cap; no runtime/acceptance credit. |
| 217 | + Canonical308 finished-batch review owns corrections. |
| 218 | +- [ ] Diagnose generic full-index pipeline failure for clean frozen worktree |
| 219 | + `/tmp/cbm-recon-luna-medium20-20261003`; logs preserved in medium20 staging. |
| 220 | + Worker safely uses verified existing runtime graph; no daemon changes made. |
| 221 | + |
| 222 | +- [x] Review resumed DeepSeek pilot20 submission; four bounded source dispositions |
| 223 | + verified and16 retained unresolved (see following task). Worker completion is not acceptance. |
| 224 | + |
| 225 | +- [ ] Resolve prior DeepSeek pilot20 independent review changes:16 exact full-hash |
| 226 | + whole-commit remainders in parent evidence/pilot20-independent-review/review.json. |
| 227 | + Correct false absent-alias/YAML/fd-guard/Codex claims, map missing CLI regressions, |
| 228 | + and preserve stale worker checksum failure; four source reviews credited, no |
| 229 | + runtime/acceptance claims. Canonical308 verification section owns corrections. |
| 230 | +- [ ] Complete remaining17 DeepSeek ops20 whole-commit source reviews, exact hashes |
| 231 | + in parent evidence/ops20-followup-v4-review/review.json. Three source dispositions |
| 232 | + verified, including missing Scala route composition d4931241. Preserve56516eff |
| 233 | + recursive-grep fallback correction,26650255/f339aa62 dependencies, separate |
| 234 | + e5c17de6 unchanged cap context and canonical SQL/stale refusal policy obligations. |
| 235 | +- [x] Review followup-v2/v3 evidence and record their semantic limits; v3 hunk |
| 236 | + inventory is discovery only. Followup-v4 corrects wrong-function/test exclusion |
| 237 | + for d4931241; other17 full-hash semantic mappings remain pending above. |
| 238 | +- [x] Independently verify d4931241adaef732066afb5ebf1cb225e2474403 source comparison: |
| 239 | + all3 hunks missing across production and both frozen QA refs. Applicable shared |
| 240 | + Scala regression/harness verified; prerequisite041eb99c11692b93d43ea72c884cc50c51285db5 |
| 241 | + present. Future implementation/Jenkins/acceptance remain unperformed. |
0 commit comments