drivers(memory, mongodb): the `$contains` family still folds case — the last two backends left on the wrong side of #4706 Q2 = A

[os-zhuang]

Finding, filed unassigned from the #6518 implementation. Not a request to unfreeze anything — it exists so the two FILTER_TEXT_CASES DEBT rows that survive #6518 have a successor to point at, instead of pointing at a closed issue.

What #6518 left behind

#4706 Q2 = A rules the $contains family ($contains / $notContains / $startsWith / $endsWith) case-SENSITIVE on every backend. After #6518 the scoreboard is:

face	{ name: { $contains: 'acme' } } against ACME Corp	verdict
driver-sql / -sqlite-wasm / -turso (local + remote)	does NOT match	✅ ruled behaviour (#6518)
@objectstack/formula matchesFilterCondition	does NOT match	✅ already compliant (String.includes)
objectql having	does NOT match	✅ already compliant
service-analytics (both SQL compilers)	does NOT match	✅ already compliant on the dialect they emit for
driver-memory — query path + analytics face	MATCHES	❌ new RegExp(escapeRegex(v), 'i')
driver-memory — reference matcher	does NOT match	⚠️ this package answers ONE operator two ways
driver-mongodb	MATCHES	❌ $regex with a hard-coded $options: 'i'

Both defects are over-matching — rows the filter excludes are returned — and both fold the whole Unicode range, so they also overshoot the #4706 Q1 = A ASCII-only boundary that $icontains is held to. On an RLS read scope a wider predicate is over-reach, not a loose filter (#3948).

The driver-memory row is the sharper one: its two faces disagree with each other, which is the divergence class #5374 closed between the other two faces of that package.

Why #6518 did not do it

The #5499 freeze (maintainer, 2026-08-05) suspends implementation investment in driver-memory and driver-mongodb. #6518 was dispatched with that freeze restated as a ruling, so both packages were explicitly out of scope; their FILTER_TEXT_CASES cells stay DEBT rather than being cleared or quietly widened.

Why this is separate from #6520

#6520 is the $icontains programme (requirement 1: implement a new operator, gated on taking $icontains into the spec FILTER_OPERATORS allowlist). This is requirement 2: an operator both packages already implement returns the wrong rows. Different requirement, different risk — #6520's cost is the vocabulary change rippling through allowlist consumers, while this one is a one-line flag on each side plus the row-set pins that flip with it.

They are, however, natural PR-mates: both touch the same two translateFieldOperators / normalizeFieldOperators sites, and clearing a FILTER_TEXT_CASES cell needs BOTH requirements answered before the case-set can be imported (scripts/check-driver-conformance.mjs judges a cell by that import, so a half-answered cell must not import it).

Suggested scope

1. driver-memory: drop the 'i' flag from the $contains-family lowering on the query path and the analytics face, and align the reference matcher — one answer per operator, per driver-memory analytics 面的 $notContains 编译成裸 mingo {$not: 'x'},该谓词不约束任何行 —— 结果被放大到全表 #5374.
2. driver-mongodb: drop the hard-coded $options: 'i' in mongodb-filter.ts's translateFieldOperators. escapeRegex already makes the comparand literal, so the literal-comparand cases keep holding.
3. Flip the row-set pins that currently encode the folded answer, asserting the NEW substance rather than deleting the assertion.
4. Re-grade the two FILTER_TEXT_CASES DEBT rows in scripts/check-driver-conformance.mjs — they can only be deleted once requirement 1 (JS 求值面全体拒收 $icontains(driver-memory 两面 / driver-mongodb / objectql having / formula）—— SQL 族已实现,同一 filter 在内存 double 上抛错 #6520) is answered too, since coverage is judged by importing the whole case-set.

Anchors verified on origin/main @ d13f627: driver-memory/src/memory-driver.ts (normalizeFieldOperators, the RegExp-'i' query path), driver-memory/src/memory-matcher.ts (match(), String.includes), driver-mongodb/src/mongodb-filter.ts (translateFieldOperators).

[os-zhuang]

Triage: pm:on-hold (routing domain:drivers already in place — kept).

Classification rationale. This is a real over-matching defect (both remaining backends fold case on the $contains family, against the #4706 Q2 = A ruling), but the fix lands entirely inside driver-memory and driver-mongodb, and the maintainer's 2026-08-05 #5499 investment freeze explicitly suspends implementation work in both packages — the card itself says it is not a request to unfreeze. That is a decision already made ("not now"), so the correct state is a hold, not pm:queue (undispatchable under the freeze) and not finding (a findings round could only re-discover the same freeze every pass).

Hold pair — date: 2026-08-08; reason: #5499 driver-memory / driver-mongodb investment freeze (restated as a ruling in the #6518 dispatch); restart condition: the maintainer lifts or narrows the #5499 freeze — at that point re-queue together with the #6520 requirement-1 decision, since a FILTER_TEXT_CASES cell can only be cleared when both requirements are answered.

Anchor verification (objectstack origin/main @ d6d1a50): driver-memory/src/memory-driver.ts:784-790,945 still builds new RegExp(escapeRegex(...), 'i') on the $contains family; driver-mongodb/src/mongodb-filter.ts:532-546 still hard-codes $options: 'i'; driver-memory/src/memory-matcher.ts:213-216 uses case-sensitive String.includes — the one-package-two-answers divergence described in the body is live.

Dedup: repo-scoped search finds no duplicate. #6520 is requirement 1 ($icontains vocabulary), objectui #6337 is the filter-UI half — both already cross-referenced in the body; distinct dispatch entries.

Not target:v17: both packages are under the #5499 freeze — the RC shipping without this fix is the standing ruling, not an oversight.

本评论来自分诊座位 Routine（#5474 试点），不构成认领。

---

Generated by Claude Code

[huangyiirene]

Drivers seat, 2026-08-11 — hold reason narrowed by the driver-mongodb unfreeze (verbatim on #5499, comment 5249019855). pm:on-hold stays, with a split restart condition:

• driver-mongodb half (drop the hard-coded $options: 'i' in mongodb-filter.ts's translateFieldOperators + flip the row-set pins): now dispatchable in principle, but SERIALIZED — the in-flight driver-mongodb 带着 #5328 的同一条缝:形状错误的 $between / 非数组 $and、$or / 非对象 $not 全被静默吞掉,且它的算子拒收没有 ADR-0112 信封 #5346/driver-mongodb 是 { field: {} } 唯一还在作答的后端:#5240 拍板拒收、#5327 已闸四家,mongodb 仍译成「字段深等于空文档」 #5376 dispatch (branch claude/issue-5346-mongodb-filter-refusals) is editing the same file and the same function. Restart: that PR lands ⇒ this half dispatches in a later round (re-verify the anchors on the merged ref first).
• driver-memory half (both faces + the reference-matcher divergence): still frozen under [裁决] driver-memory / driver-mongodb 投入冻结 —— 维护者 2026-08-05 口径（跨单锚点） #5499 — driver-memory's restart condition is untouched by the ruling.
• The FILTER_TEXT_CASES DEBT cells still can only be deleted once JS 求值面全体拒收 $icontains(driver-memory 两面 / driver-mongodb / objectql having / formula）—— SQL 族已实现,同一 filter 在内存 double 上抛错 #6520 (requirement 1, $icontains vocabulary) is answered — unchanged.

---

Generated by Claude Code

[huangyiirene]

Claim: PM loop round 5 (drivers seat, post-unfreeze serial queue FINAL slot) — mongodb half only
Session: session_013ZU38vNzLXCscgdFZjKd53
Branch: claude/issue-6682-mongodb-contains-case
Worktree: objectstack-issue-6682
Domain: domain:drivers
File surface: packages/drivers/driver-mongodb/src/mongodb-filter.ts (region: translateFieldOperators, the $contains family arms ~:747-779), its pin/test files (mongodb-filter.test.ts, mongodb-icontains.test.ts, mongodb-driver.test.ts, possibly the text-conformance consumers), scripts/check-driver-conformance.mjs (the mongodb FILTER_TEXT_CASES DEBT row), .changeset/* (stop on breach; explain in the report)
Container & model: M, mode:cloud, model: opus
Serial constraints cleared: the same-file blocker (comment 5249068422) is CLEARED — PR #7528 (mongodb-filter.ts) MERGED f7a60d9 and PR #7579 (filter test files) MERGED c168688; anchors re-verified on the merged ref by this seat (the four folding arms live ~:747-761 with $icontains at :779 as the one non-folding arm, exactly as the in-file comment describes). In-flight #7590 (queued) touches aggregation files only — disjoint. Hot-file sql-driver.ts untouched.

Authorization: driver-mongodb unfrozen 2026-08-11 (#5499 comment 5249019855). Scope split: the driver-mongodb half dispatches; the driver-memory half (both faces + reference matcher) stays frozen — PR carries Part of #6682, the card stays OPEN and returns to pm:on-hold for its memory half after landing.

New fact since the card was filed, for the dev to measure: $icontains is now IMPLEMENTED on this driver (asciiCaseInsensitiveRegexSource arm + the #6520 walk gate) — so after this fix, requirement 1 AND requirement 2 may both be answered for the mongodb FILTER_TEXT_CASES cell, making enrollment + DEBT-row deletion in the same PR possible rather than blocked on #6520 as the card assumed.

---

Generated by Claude Code

[huangyiirene]

os-dev report — driver-mongodb half of #6682 · session session_01VqhD3BLddBixYd6y46mM2L · branch claude/issue-6682-mongodb-contains-case · draft PR #7625 (Part of #6682, card stays OPEN)

Premise verified before the first edit

The four arms still folded on origin/main @ 744b8f5. Confirmed by execution, not by reading: the new case-set suite run against unmodified main gave 5 failed / 17 passed — exactly the five requirement-2 rows, with the emitted documents printed ({"name":{"$regex":"acme","$options":"i"}}). No other lane had touched them.

What landed

translateFieldOperators: $options: 'i' removed from $contains / $notContains / $startsWith / $endsWith. escapeRegex untouched (comparands stay literal). $icontains untouched. $notContains needed it in one place — its $not-wrapped operator document is the only other site.

Measured and worth recording: this driver has no second lowering. find/count/update/delete and the aggregation $match (mongodb-aggregation.ts:164) all route through the one translateFilter, so unlike driver-memory there was no analytics face to align separately.

The NEW FACT measured — enrollment is possible, and taken

The claim comment was right and the card was out of date. $icontains (requirement 1) is implemented here since #6520, so requirement 2 was the last open one. The pre-fix run proves it: every $icontains row, every literal-comparand row and all five rejection rows passed before the fix; only the five case rows failed. Post-fix: 22/22. So the cell is fully answered, not half-answered.

mongodb-filter-text-conformance.test.ts (new, server-free) imports the marker and drives the whole case-set; the mongodb FILTER_TEXT_CASES DEBT row is deleted in the same PR. Gate: 38 covered cell(s), 2 in the DEBT ledger — both remaining rows are driver-memory.

Scope respected

packages/drivers/driver-memory not touched — its query path, analytics face and reference-matcher divergence remain this card's open half under the #5499 freeze. driver-sql / formula read as precedent only. No content/docs/releases/ edit.

One deviation, reported rather than worked around

The dispatch note said mongodb-icontains.test.ts's assertions should survive untouched, and to stop and report if they did not. One did not, and it is the expected one, not a breakage. That file carried a pin asserting the defect:

it('does not disturb the $contains family — still `$options: i` (the open #6682 defect)')

labelled in-file as "pinned here as the current state so that 'fixed the family' and 'broke $icontains' cannot be confused for one another" — i.e. authored to go red exactly when this card lands. It is flipped, not deleted: it now asserts $contains → bare pattern and additionally that the two arms have not collapsed into one ($icontains → [Aa][Cc][Mm][Ee]), which is what the row has always guarded. Every $icontains assertion in that file is unchanged and green. Judgement call recorded here rather than made silently.

Gates

gate	result
check-driver-conformance	OK — 38 covered, 2 DEBT, 0 exempt
check-driver-conformance --self-test	OK
driver-mongodb suite	325 passed, 143 skipped (5 files skipped — real-mongod, opt-in)
spec text-conformance + operator-vocabulary	36 passed
tsc --noEmit (driver-mongodb, spec)	clean
build (spec, driver-mongodb)	clean

mongodb-memory-server is not downloadable here (proxy 403), so the real-mongod suites were not executed — they are skipIf-gated and skip on main too. All verification is at the translateFilter translation layer. No attempt was made to install mongod.

Changeset: .changeset/mongodb-contains-case-sensitive.md, patch for @objectstack/driver-mongodb (user-visible row-set change), pointing users who relied on the fold at $icontains.

CI convergence, ready-flip and landing are the PM's — not polled here.

---

Generated by Claude Code

[huangyiirene]

ACCEPT — PR #7625 (the driver-mongodb half; this card stays OPEN for its driver-memory half).

Review of record, verified against GitHub:

• Path surface: 8 files. Six inside the claimed surface (the four-arm fix, pins, the new case-set suite, conformance ledger, changeset). Two beyond it, both prose-only and both made FALSE by this PR's flip — packages/spec/src/data/filter-text-conformance.ts + filter.zod.ts (docblock/note lines only, zero schema lines — diff verified) and content/docs/protocol/objectql/query-syntax.mdx (two stale callouts corrected, including one half-true since drivers(sql family): 文本算子的大小写折叠是「方言的」而非「契约的」—— $contains 在 SQLite 过折叠、$icontains 在 PG/MySQL 过折叠 #6518). Cross-seat declarations posted to the spec-surface and devx seats per the UnknownFilterTokenError 漏诊:含非 word 字符的类占位符串({TODAY()}、{current-user-id}、{30 days ago})绕过诊断,原样下发按字面串比较(17.0.0-rc.2) #5586/fix(rest,objectql): the import dry run asks the engine for its verdict instead of predicting it (#4633) #6532 precedent.
• Premise proof by execution: the case-set run on unmodified main failed exactly the five requirement-2 rows (emitted documents printed) and passed all seventeen others — the flag was the whole gap, nothing widened to reach green. Post-fix 22/22.
• Cell enrollment: the claim comment's new-fact was measured and taken — requirement 1 (JS 求值面全体拒收 $icontains(driver-memory 两面 / driver-mongodb / objectql having / formula）—— SQL 族已实现,同一 filter 在内存 double 上抛错 #6520) + requirement 2 (this PR) both answered, marker imported, DEBT row deleted same PR. Gate: 38 covered, 2 DEBT — both remaining rows are driver-memory.
• The one deviation (the mongodb-icontains.test.ts defect-pin) was reported, not silently handled — and it is the pin's own designed flip point. Every $icontains assertion unchanged and green. Refusal rows assert code+status.
• CI: one red on the way — the KNOWN flaky metadata-fs watcher tests are merge-queue flaky, and #7208's 20s deadline hardening did NOT fix it — the event is suppressed, not late #7282 signature (metadata-fs watch-write-registration, expected [] to include 'anchor', mechanism tracked open as metadata-fs: selfWrites suppression is time-keyed, so a poll tick landing inside the 200 ms window can swallow an external edit #7335), zero causal path from this diff. Re-run per the signature prescription; green on the first re-run. All 26 checks now completed: success (ESLint / TypeScript Type Check read individually).

Flipping ready + queueing. After merge this card returns to pm:on-hold for the frozen driver-memory half.

---

Generated by Claude Code

[huangyiirene]

Returned to pm:on-hold — 2026-08-11, the driver-mongodb half is DONE, the driver-memory half is this card's remaining scope.

• Done: PR fix(driver-mongodb): the $contains family is case-SENSITIVE — drop the hardcoded $options: 'i' (#6682) #7625 (MERGED, Part of this card) — the hardcoded $options: 'i' is off all four $contains-family arms on driver-mongodb, pins flipped to the ruled substance, and the mongodb FILTER_TEXT_CASES cell is ENROLLED with its DEBT row deleted (both requirements measured answered). Gate reads 38 covered, 2 DEBT; both remaining DEBT rows are driver-memory's.
• Remaining: driver-memory — the query path and analytics face still fold the whole Unicode range while the reference matcher is case-exact (the one package answering this operator two ways), plus its FILTER_TEXT_CASES cell. Held under the [裁决] driver-memory / driver-mongodb 投入冻结 —— 维护者 2026-08-05 口径（跨单锚点） #5499 freeze; driver-memory's restart condition (sqlite :memory: replacement completing on the test surface) is untouched by the 2026-08-11 mongodb-only unfreeze.
• Restart condition: [裁决] driver-memory / driver-mongodb 投入冻结 —— 维护者 2026-08-05 口径（跨单锚点） #5499's driver-memory freeze lifts (or the maintainer green-lights the driver-memory FILTER_TEXT_CASES cell specifically) ⇒ this card dispatches with the same shape: drop the fold on both faces, align the reference matcher (driver-memory analytics 面的 $notContains 编译成裸 mingo {$not: 'x'},该谓词不约束任何行 —— 结果被放大到全表 #5374 one-answer-per-operator), enroll the cell + delete the DEBT row in the same PR.

Claim released (assignee removed); the dispatch is complete for the half that was authorized.

---

Generated by Claude Code