fix(miner): report the real post-failure schema version for a partially-applied migration

[jeffrey701]

Closes #6767

Problem

migrateStore (packages/loopover-miner/lib/migrate-cli.js) reports versionAfter: versionBefore in its catch branch — "nothing changed". But schema-version.js's applySchemaMigrations applies and stamps each migration in its own transaction, specifically so a failure part-way through leaves the file at the last fully-applied version. So when open(dbPath) throws while applying migration N of several pending ones, the migrations before N have already committed to disk and the real on-disk version has advanced past versionBefore — but the catch never re-reads it, and the failed result misreports what's actually on disk.

Fix

Re-read the real on-disk version in the catch branch, guarded by its own try: a genuinely corrupt/unopenable file (the same case that can leave versionBefore null) may fail this read too, in which case the pre-failure reading is all we can honestly report.

Tests

test/unit/miner-migrate-cli.test.ts adds a regression test that injects a store whose open runs a migrations array where the first migration commits (stamping BASELINE+1) and the second throws — asserting the reported versionAfter is BASELINE+1 (the real partial progress) and that it matches the file's actual on-disk PRAGMA user_version. It fails before the fix (expected 1 to be 2).

Both branches of the new guard are covered:

• re-read succeeds → the new regression test;
• re-read throws → the existing "corrupted file" case, whose unreadable file makes the re-read throw too, falling back to versionBefore.

Existing expectations are unchanged (the fake-store case still reports versionAfter: 0).

Validation

• npx vitest run test/unit/miner-migrate-cli.test.ts → 11/11 pass (10 existing + 1 new)
• npm run typecheck → clean (0 errors)
• Scope: 1 source file + 1 test file (+67/−1)

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.64%. Comparing base (615dd46) to head (03bef9e).
⚠️ Report is 5 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #6860   +/-   ##
=======================================
  Coverage   93.64%   93.64%           
=======================================
  Files         685      685           
  Lines       68238    68242    +4     
  Branches    18707    18707           
=======================================
+ Hits        63903    63907    +4     
  Misses       3350     3350           
  Partials      985      985           

Flag	Coverage Δ
shard-1	43.23% <0.00%> (-0.46%)	⬇️
shard-2	36.93% <0.00%> (+0.24%)	⬆️
shard-3	32.59% <0.00%> (-0.12%)	⬇️
shard-4	34.36% <0.00%> (-0.40%)	⬇️
shard-5	31.48% <100.00%> (+0.33%)	⬆️
shard-6	45.97% <0.00%> (+0.27%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
packages/loopover-miner/lib/migrate-cli.js	100.00% <100.00%> (ø)

[loopover-orb]

Tip

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-17 09:11:49 UTC}

2 files · 1 AI reviewer · no blockers · readiness 88/100 · CI green · clean

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
This fixes a real bug: migrateStore's catch branch hardcoded versionAfter to versionBefore, misreporting the schema version when applySchemaMigrations commits some migrations before a later one throws. The fix re-reads the on-disk version in the catch, guarded by its own try/catch to handle the case where the file itself is now unreadable, falling back to versionBefore. The regression test correctly exercises the real applySchemaMigrations per-migration-transaction behavior rather than fabricating an impossible state, and verifies against the actual on-disk PRAGMA user_version.

Nits — 2 non-blocking

• migrate-cli.js: the inner try/catch could just let peekSchemaVersion's exception fall through since versionAfter is already initialized to versionBefore, making the explicit catch block redundant (`let versionAfter = versionBefore; try { versionAfter = peekSchemaVersion(dbPath); } catch {}` would suffice without reassigning in the catch).
• migrate-cli.js:88-93: simplify the try/catch to an empty catch block since versionAfter is already defaulted to versionBefore before the try.

Decision drivers

• ✅ Code review — No blockers (1 reviewer)
• ✅ Gate result — Passing (No configured blocker found.)

Context & advisory signals — never blocks the verdict

Signal	Result	Evidence
Linked issue	✅ Linked	#6767
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Change scope	❌ 8/20	High review scope from cached public metadata (1 linked issue).
Validation posture	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 121 registered-repo PR(s), 59 merged, 30 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	jeffrey701; Gittensor profile; 121 PR(s), 30 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Linked issue satisfaction

Addressed
The diff adds a guarded re-read of the on-disk schema version in migrateStore's catch branch, falling back to versionBefore only if that read itself throws, exactly matching the issue's requirement, and includes a regression test with a multi-migration array where a later migration fails after an earlier one commits, asserting the reported versionAfter reflects real partial progress.

Review context

• Author: jeffrey701
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: TypeScript, Clojure, JavaScript, Rust
• Official Gittensor activity: 121 PR(s), 30 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Start here: Add a concise scope and risk note.
• Then work through the remaining 1 step in the Signals table above.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Change scope = cached public metadata such as size labels, draft state, and review-burden hints.
• Validation posture = whether the PR provides enough public validation/test evidence for maintainer review.
• Contributor workload = public contributor activity and cleanup pressure, not a repo-wide quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

🧪 Chat with LoopOver

Ask LoopOver a question about this PR directly in a comment — grounded only in the same cached, public-safe facts shown above, never a new claim.

• @loopover ask <question> answers contribution-quality Q&A with source citations and freshness.
• @loopover chat <question> answers in natural prose from cached decision-pack facts via local inference (maintainer/collaborator; read-only).
• A plain-language @loopover mention with a real question is routed to the closest matching read-only command automatically — no exact syntax required.

Full command reference: https://loopover.ai/docs/loopover-commands

🧪 Experimental — new and may change.

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by LoopOver, a quiet PR intelligence layer for OSS maintainers.

• Re-run LoopOver review

[loopover-orb]

LoopOver approves — the gate is satisfied and CI is green.