objectstack-ai
diff --git a/‎.changeset/21376-boolean-comparand-compilers.md‎
Lines changed: 17 additions & 0 deletions b/‎.changeset/21376-boolean-comparand-compilers.md‎
Lines changed: 17 additions & 0 deletions
diff --git a/‎content/docs/permissions/rls.mdx‎
Lines changed: 11 additions & 2 deletions b/‎content/docs/permissions/rls.mdx‎
Lines changed: 11 additions & 2 deletions
diff --git a/‎packages/plugins/plugin-security/src/rls-boolean-comparand-door.test.ts‎
Lines changed: 300 additions & 0 deletions b/‎packages/plugins/plugin-security/src/rls-boolean-comparand-door.test.ts‎
Lines changed: 300 additions & 0 deletions
@@ -0,0 +1,17 @@
+---
+'@objectstack/plugin-security': minor
+'@objectstack/service-analytics': minor
+---
+
+Row-level security policies and the analytics native-SQL path judge a comparand against a declared boolean field by the platform's boolean-comparand rule, the one the data engine's `where` already applies
+
+Clause-②: no (narrowing)
+
+<!-- adr-0087: not-required (no-migration-prescription) a refusal or narrowing of a filter comparand against a declared boolean column at two compilers outside the engine's where door, the same comparand that door already judges: no authorable key, spelling, export or stored shape moves. RowLevelSecurityPolicySchema, every permission set, every dataset, cube and analytics query parse and save as before, the predicate's and the filter's text are untouched, @objectstack/plugin-security and @objectstack/service-analytics export the same names with the same types, and no stored row is read or rewritten. Which boolean the author meant by a refused comparand is not something a ledger entry can decide, so there is nothing for objectstack migrate meta to rewrite. The other categories are closed on facts: both packages publish (not unpublished); no ADR-0087 id covers a filter comparand's type and this diff adds none (not registered / already-registered); and the change is runtime behaviour with no published interface or type changed (not runtime-interface-only / type-surface-only). -->
+
+**BREAKING**: this narrows what two compilers outside the engine's `where` door accept. The RLS compile seam now drops a row-level policy, and the analytics native-SQL face now refuses a query, when either compares a declared boolean field with a comparand outside the accepted set. It ships as `minor` under the launch-window convention for accept-set narrowings. No export, type or error code changes.
+
+- **Row-level security (`@objectstack/plugin-security`).** A compiled `using` / `check` predicate on a `boolean` or `toggle` column (or a `formula` returning `boolean`) is judged by `booleanComparandDoorVerdict` from `@objectstack/spec/data`, in the same pass as the number rule. `'true'` / `'false'`, `'1'` / `'0'` and `1` / `0` are read as the boolean each names. Anything else the rule refuses (a string such as `'yes'`, `'TRUE'` or `''`, a number other than `1` / `0`) drops the policy as a refused comparand: the read is filtered by the deny sentinel, the write is refused 403, and the WARN line names the clause, the field and the position. Before, `record.flag != 'true'` kept every row on SQLite and the write check admitted every row, so the exclusion the author wrote was not applied.
+- **Analytics native SQL (`@objectstack/service-analytics`).** The query's `where` (and the dataset query's `runtimeFilter`, which is merged into it), each measure's own `filter` and a dataset's own `filter` are judged by the same rule before the statement compiles. An accepted spelling is read as its boolean, and anything else the rule refuses is refused `INVALID_FILTER` / 400 with the rule's own message, before any statement runs. The native strategy now answers what the engine-aggregate strategy answers. Before, `{ flag: 'true' }` counted no rows on SQLite, `{ flag: { $ne: 'true' } }` counted every row, and `{ flag: 'yes' }` answered 200.
+- **What you may notice.** A policy or analytics filter that compared a boolean field with a value outside the accepted set now refuses instead of answering. Write `true` / `false`. A policy `record.flag == 1` now admits writing a `true` row, which its read already showed.
+- **Unchanged.** A boolean literal, a column that is not boolean, a `{ $field }` reference, and an object whose declaration cannot be read (nothing is judged without one).
@@ -197,14 +197,23 @@ Layer 1 (business RLS)       ─┘
 
 ## The fail-closed contract
 
-Four ways a policy denies rather than leaks:
+Five ways a policy denies rather than leaks:
 
 1. A policy exists but **every** applicable expression fails to compile → a
    deny-everything filter.
 2. A referenced context variable is missing, null, or an empty array → that
    policy drops out (it cannot match).
 3. A policy references a column the object doesn't have → deny.
-4. `check` is omitted → `using` stands in as the `check`. The choice is
+4. A policy compares a column with a literal its declared type cannot be
+   compared with → that policy drops out, for `using` and `check` alike: on a
+   `boolean` / `toggle` column, anything but `true` / `false` and the
+   spellings `'true'` / `'false'`, `1` / `0` and `'1'` / `'0'`
+   (`active == 'yes'`, `active != 'TRUE'`, `active == 2`); on a numeric
+   column, a comparand that is not a number (a string with no numeric reading,
+   a boolean). These are the comparisons a caller's `where` is refused for
+   (`INVALID_FILTER`). An accepted spelling is read as the value it names, so
+   `active != 'true'` hides the `true` rows exactly as `active != true` does.
+5. `check` is omitted → `using` stands in as the `check`. The choice is
    made per operation across all the applicable policies, not policy by
    policy: when any of them declares a `check`, only the declared checks
    decide, and a USING-only sibling's `using` is not part of the check.
 
@@ -0,0 +1,300 @@
+// Copyright (c) 2026 ObjectStack. Licensed under the Apache-2.0 license.
+
+/**
+ * [#21376] A row-level policy comparing a BOOLEAN column with a comparand that
+ * is not a boolean is refused at the RLS compile seam, and an accepted
+ * spelling is narrowed to the boolean it names — as the engine's `where` door
+ * answers the same comparison — so the read, the write and a caller's `where`
+ * give one answer.
+ *
+ * Driven through the real engine, the real plugin and `SqlDriver`, as a member
+ * resolving a permission set whose `using` and `check` are the same predicate,
+ * over two rows (`t` stores `true`, `f` stores `false`). Measured before the
+ * seam ran the spec's boolean-comparand verdict (`booleanComparandDoorVerdict`,
+ * the one the engine's `where` door consults beside the number one):
+ *
+ * | predicate | read, SQLite | read, PostgreSQL 16 | write `true` / `false` | `where` twin |
+ * |---|---|---|---|---|
+ * | `record.flag == true` | `t` | `t` | admitted / 403 | `t` |
+ * | `record.flag == 'true'` | none | `t` | 403 / 403 | `t` |
+ * | `record.flag != 'true'` | **`f`, `t`** | `f` | **admitted** / admitted | `f` |
+ * | `record.flag == 'yes'` | none | **`t`** | 403 / 403 | `INVALID_FILTER` / 400 |
+ * | `record.flag == 1` | `t` | `t` | 403 / 403 | `t` |
+ * | `record.flag == '1'` | `t` | `t` | 403 / 403 | `t` |
+ *
+ * The negation was fail-open on both faces: the read kept the row the author
+ * excluded, and the write check admitted it. The policy's comparand is now
+ * judged by the spec's verdict in the same walk as the number arm: `'true'` /
+ * `'false'`, `'1'` / `'0'` and `1` / `0` narrow to the boolean each names, so
+ * every cell answers the `where` twin's rows and the write check admits
+ * exactly what the read shows; anything else the verdict refuses drops the
+ * policy through the `refused-comparand` route for both clauses (the read gets
+ * the deny sentinel, the write a 403). `record.flag == true` is the control
+ * and does not move.
+ *
+ * The compiled policy filter is deep-frozen as `compileCelToFilter` returns it
+ * (the mock below), so every cell also holds the narrowing to copy-on-write:
+ * an edit in place would throw. The last block reads the frozen filter back.
+ *
+ * The PostgreSQL cell runs where `OS_TEST_POSTGRES_URL` is set and is a named
+ * skip otherwise; no CI step provisions that variable for this package. Each
+ * live table is dropped after its case.
+ */
+
+import { describe, it, expect, afterEach, vi } from 'vitest';
+import { ObjectQL } from '@objectstack/objectql';
+import { SqlDriver } from '@objectstack/driver-sql';
+import { PermissionSetSchema } from '@objectstack/spec/security';
+import { SecurityPlugin } from './security-plugin.js';
+import { defaultPermissionSets } from './objects/default-permission-sets.js';
+
+/** Every compiled policy filter `compileCelToFilter` handed the seam, deep-frozen, by predicate. */
+const compiled = new Map<string, unknown[]>();
+
+function deepFreeze<T>(value: T): T {
+  if (value !== null && typeof value === 'object' && !Object.isFrozen(value)) {
+    Object.freeze(value);
+    for (const child of Object.values(value as Record<string, unknown>)) deepFreeze(child);
+  }
+  return value;
+}
+
+vi.mock('@objectstack/formula', async (importOriginal) => {
+  const real = await importOriginal<typeof import('@objectstack/formula')>();
+  return {
+    ...real,
+    compileCelToFilter: ((expression, options) => {
+      const result = real.compileCelToFilter(expression, options);
+      if (result.ok) {
+        deepFreeze(result.filter);
+        const key = typeof expression === 'string' ? expression : (expression.source ?? '');
+        compiled.set(key, [...(compiled.get(key) ?? []), result.filter]);
+      }
+      return result;
+    }) as typeof real.compileCelToFilter,
+  };
+});
+
+const SYS_CTX = { isSystem: true, userId: 'usr_system' };
+const MEMBER_DEFAULT = defaultPermissionSets.find((p) => p.name === 'member_default')!;
+
+interface Cell {
+  id: 'sqlite' | 'pg';
+  label: string;
+  config: () => Record<string, unknown> | null;
+}
+
+const DRIVER_CELLS: readonly Cell[] = [
+  { id: 'sqlite', label: 'SqlDriver, SQLite', config: () => ({ client: 'better-sqlite3', connection: { filename: ':memory:' }, useNullAsDefault: true }) },
+  {
+    id: 'pg',
+    label: 'SqlDriver, live PostgreSQL',
+    config: () => (process.env.OS_TEST_POSTGRES_URL ? { client: 'pg', connection: process.env.OS_TEST_POSTGRES_URL } : null),
+  },
+];
+
+const cleanups: Array<() => Promise<void>> = [];
+afterEach(async () => {
+  while (cleanups.length) {
+    try { await cleanups.pop()!(); } catch { /* noop */ }
+  }
+});
+
+let seq = 0;
+/** One engine and plugin over `config`, with ONE policy whose `using` and `check` are the same predicate. */
+async function boot(config: Record<string, unknown>, predicate: string) {
+  const OBJ = `qa_rls_boolean_${process.pid}_${++seq}`;
+  const driver = new SqlDriver(config as never);
+  const engine = new ObjectQL();
+  engine.registerDriver(driver as never, true);
+  await engine.init();
+  engine.registerApp({
+    id: `com.objectstack.qa.rls-boolean-comparand-${seq}`,
+    name: 'RLS boolean comparand',
+    version: '1.0.0',
+    type: 'plugin',
+    scope: 'system',
+    objects: [
+      {
+        name: OBJ,
+        label: 'Flagged line',
+        sharingModel: 'public_read_write',
+        fields: {
+          id: { name: 'id', type: 'text', primaryKey: true },
+          flag: { name: 'flag', type: 'boolean' },
+        },
+      },
+    ],
+  } as never);
+  await engine.syncSchemas();
+  cleanups.push(async () => {
+    if (config.client === 'pg') await (driver as unknown as { execute(sql: string): Promise<unknown> }).execute(`drop table if exists ${OBJ}`);
+    await engine.destroy();
+  });
+
+  const set = PermissionSetSchema.parse({
+    name: 'qa_boolean_guard',
+    objects: { [OBJ]: { allowRead: true, allowCreate: true, allowEdit: true, allowDelete: true } },
+    rowLevelSecurity: [{ name: 'boolean_guard', object: OBJ, operation: 'all', using: predicate, check: predicate }],
+  });
+  const services: Record<string, unknown> = {
+    manifest: { register: vi.fn() },
+    objectql: engine,
+    metadata: {
+      get: async (_type: string, name: string) => engine.getSchema(name) ?? null,
+      list: async () => [MEMBER_DEFAULT, set],
+    },
+  };
+  const warn = vi.fn();
+  const ctx = {
+    logger: { info: vi.fn(), warn, error: vi.fn(), debug: vi.fn() },
+    registerService: vi.fn(),
+    getService: (name: string) => {
+      if (!(name in services)) throw new Error(`service not registered: ${name}`);
+      return services[name];
+    },
+  };
+  const plugin = new SecurityPlugin({ fallbackPermissionSet: 'member_default' });
+  await plugin.init(ctx as never);
+  await plugin.start(ctx as never);
+  vi.spyOn((engine as unknown as { logger: { warn: () => void } }).logger, 'warn').mockImplementation(() => undefined);
+
+  await engine.insert(OBJ, { id: 't', flag: true }, { context: SYS_CTX } as never);
+  await engine.insert(OBJ, { id: 'f', flag: false }, { context: SYS_CTX } as never);
+
+  const caller = { userId: 'usr_member', positions: ['qa_pos'], permissions: [set.name], posture: 'MEMBER' };
+  const ids = (rows: unknown) => (rows as Array<{ id: string }>).map((r) => r.id).sort();
+  /** The ids the member's read shows. */
+  const shown = async () => ids(await engine.find(OBJ, { context: caller } as never));
+  /** The engine's `where` door on the same comparison: the rows it serves, or its refusal envelope. */
+  const whereTwin = (where: Record<string, unknown>) =>
+    engine.find(OBJ, { where, context: SYS_CTX } as never).then(ids, (e: unknown) => envelopeOf(e));
+  const write = (id: string, flag: boolean) =>
+    outcome(engine.insert(OBJ, { id, flag }, { context: caller } as never));
+  /** `clause:reason` of every fail-closed drop of this policy the compile seam logged. */
+  const drops = () => [
+    ...new Set(
+      warn.mock.calls
+        .map((call) => call[1] as { reason?: string; clause?: string; policy?: string } | undefined)
+        .filter((meta) => meta?.policy === 'boolean_guard')
+        .map((meta) => `${meta!.clause}:${meta!.reason}`),
+    ),
+  ].sort();
+  /** The `detail` of each fail-closed drop of this policy, by clause. */
+  const details = () =>
+    Object.fromEntries(
+      warn.mock.calls
+        .map((call) => call[1] as { clause?: string; policy?: string; detail?: string } | undefined)
+        .filter((meta) => meta?.policy === 'boolean_guard')
+        .map((meta) => [meta!.clause!, meta!.detail!]),
+    ) as Record<string, string>;
+  return { shown, whereTwin, write, drops, details };
+}
+
+type Envelope = { code: string; status: number };
+const DENIED: Envelope = { code: 'PERMISSION_DENIED', status: 403 };
+const REFUSED: Envelope = { code: 'INVALID_FILTER', status: 400 };
+const envelopeOf = (e: unknown): Envelope => {
+  const x = e as { code?: string; status?: number; statusCode?: number };
+  return { code: String(x?.code), status: Number(x?.statusCode ?? x?.status) };
+};
+const outcome = (p: Promise<unknown>): Promise<'admitted' | Envelope> =>
+  p.then(() => 'admitted' as const, (e: unknown) => envelopeOf(e));
+
+interface Row {
+  predicate: string;
+  /** The same comparison as a caller's `where`. */
+  where: Record<string, unknown>;
+  /** What the member's read shows — the `where` twin's rows. */
+  shown: string[];
+  /** The write check on a `true` row and on a `false` row. */
+  writes: readonly ['admitted' | Envelope, 'admitted' | Envelope];
+}
+
+/** Each cell answers the engine-door column; the write check admits what the read shows. */
+const NARROWED: readonly Row[] = [
+  // The control: a boolean literal. Unchanged on every face.
+  { predicate: 'record.flag == true', where: { flag: true }, shown: ['t'], writes: ['admitted', DENIED] },
+  { predicate: "record.flag == 'true'", where: { flag: 'true' }, shown: ['t'], writes: ['admitted', DENIED] },
+  // The negation hides the row the author excluded, and the write check refuses it.
+  { predicate: "record.flag != 'true'", where: { flag: { $ne: 'true' } }, shown: ['f'], writes: [DENIED, 'admitted'] },
+  { predicate: "record.flag == 'false'", where: { flag: 'false' }, shown: ['f'], writes: [DENIED, 'admitted'] },
+  // The number spelling: the read was already right on both dialects; the
+  // write check compared the stored `true` with `1` and refused it.
+  { predicate: 'record.flag == 1', where: { flag: 1 }, shown: ['t'], writes: ['admitted', DENIED] },
+  { predicate: "record.flag == '1'", where: { flag: '1' }, shown: ['t'], writes: ['admitted', DENIED] },
+  { predicate: "record.flag in ['true']", where: { flag: { $in: ['true'] } }, shown: ['t'], writes: ['admitted', DENIED] },
+];
+
+/** Each refused: no row shown, both writes 403, both clauses dropped, and the twin is the engine's 400. */
+const REFUSED_ROWS: ReadonlyArray<readonly [predicate: string, where: Record<string, unknown>]> = [
+  ["record.flag == 'yes'", { flag: 'yes' }],
+  ["record.flag != 'yes'", { flag: { $ne: 'yes' } }],
+  ["record.flag == 'TRUE'", { flag: 'TRUE' }],
+  ["record.flag in [true, 'on']", { flag: { $in: [true, 'on'] } }],
+  // A number other than 1 / 0 is the verdict's refusal too: the seam carries no
+  // table of its own, so it answers whatever the spec's verdict answers.
+  ['record.flag == 2', { flag: 2 }],
+];
+
+for (const cell of DRIVER_CELLS) {
+  const config = cell.config();
+  const suffix = config ? '' : ' (skipped: set OS_TEST_POSTGRES_URL to run this cell)';
+
+  describe.skipIf(!config)(`[#21376] a boolean comparand is narrowed at the RLS seam as the where door narrows it — ${cell.label}${suffix}`, () => {
+    for (const row of NARROWED) {
+      it(`${row.predicate}: the read shows [${row.shown.join(', ')}], the where twin's rows, and the write check agrees`, async () => {
+        const r = await boot(config!, row.predicate);
+        expect(await r.whereTwin(row.where)).toEqual(row.shown);
+        expect(await r.shown()).toEqual(row.shown);
+        expect(await r.write('wt', true)).toEqual(row.writes[0]);
+        expect(await r.write('wf', false)).toEqual(row.writes[1]);
+        expect(r.drops()).toEqual([]);
+      });
+    }
+  });
+
+  describe.skipIf(!config)(`[#21376] a string that names no boolean is refused at the RLS seam, read and write alike — ${cell.label}${suffix}`, () => {
+    for (const [predicate, where] of REFUSED_ROWS) {
+      it(`${predicate}: no row is shown, both writes are 403, and the where twin is INVALID_FILTER / 400`, async () => {
+        const r = await boot(config!, predicate);
+        expect(await r.whereTwin(where)).toEqual(REFUSED);
+        expect(await r.shown()).toEqual([]);
+        expect(await r.write('wt', true)).toEqual(DENIED);
+        expect(await r.write('wf', false)).toEqual(DENIED);
+        // Both clauses dropped the policy through the shared-face route.
+        expect(r.drops()).toEqual(['check:refused-comparand', 'using:refused-comparand']);
+      });
+    }
+  });
+}
+
+describe('[#21376] the refusal names its clause and the comparand\'s position', () => {
+  it("record.flag != 'yes': each clause's detail is rooted at that clause", async () => {
+    const r = await boot(DRIVER_CELLS[0].config()!, "record.flag != 'yes'");
+    expect(await r.shown()).toEqual([]);
+    expect(await r.write('wt', true)).toEqual(DENIED);
+    const { using, check } = r.details();
+    expect(using).toContain('`using` predicate compares a boolean column');
+    expect(using).toContain('using.flag.$ne');
+    expect(check).toContain('`check` predicate compares a boolean column');
+    expect(check).toContain('check.flag.$ne');
+  });
+});
+
+describe('[#21376] narrowing is copy-on-write: the compiled policy filter is never edited', () => {
+  it("record.flag != 'true': the filter the compiler returned still holds the string after the read and the write", async () => {
+    const predicate = "record.flag != 'true'";
+    compiled.delete(predicate);
+    const r = await boot(DRIVER_CELLS[0].config()!, predicate);
+    expect(await r.shown()).toEqual(['f']);
+    expect(await r.write('wt', true)).toEqual(DENIED);
+    const filters = compiled.get(predicate) ?? [];
+    // One compile per read and per write check, every one frozen and unedited.
+    expect(filters.length).toBeGreaterThanOrEqual(2);
+    for (const filter of filters) {
+      expect(Object.isFrozen(filter)).toBe(true);
+      expect(JSON.stringify(filter)).toContain('"true"');
+    }
+  });
+});