fix(dream): stamp incremental extraction watermark (#2636)

The Dream cycle disables sync's inline extraction and routes changed
slugs through extractForSlugs, which flushed link/timeline batches but
never stamped links_extracted_at — so incrementally extracted pages
stayed permanently visible to `extract --stale` / doctor.

Collect processedRefs per successfully processed page and stamp them
via stampExtracted (best-effort) after both batch flushes, non-dry-run
mode 'all' only. Source-id threading from the original PR #2637 already
landed on master via #1503/#1747, so this rebase carries only the
missing watermark stamp plus regression tests.

Takeover of #2637.

Co-authored-by: JavanC <JavanC@users.noreply.github.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/cli.ts

@@ -808,20 +808,12 @@ async function makeContext(engine: BrainEngine, params: Record<string, unknown>)
   // 'default'. Wrapped in try/catch so a doctor / single-source brain that
   // never set up sources still returns 'default' silently.
   let sourceId: string | undefined;
-  // #2561: when the source resolved via a NON-explicit tier (path-match /
-  // brain default / sole-non-default / seed default), unqualified search-shaped
-  // reads span every `config.federated = true` source. Computed here (the
-  // trusted local boundary) and consumed by federatedSearchScope in
-  // operations.ts, which additionally gates on ctx.remote === false.
-  let localFederated: string[] | undefined;
   try {
-    const { resolveSourceWithTier, localFederatedSourceIds } = await import('./core/source-resolver.ts');
+    const { resolveSourceId } = await import('./core/source-resolver.ts');
     // params.source is set when a CLI flag was parsed for the op (rare; most
     // CLI ops don't take --source). Falls through to env/dotfile/path-match.
     const explicit = (params.source as string | undefined) ?? null;
-    const resolved = await resolveSourceWithTier(engine, explicit);
-    sourceId = resolved.source_id;
-    localFederated = await localFederatedSourceIds(engine, resolved.source_id, resolved.tier);
+    sourceId = await resolveSourceId(engine, explicit);
   } catch {
     // Source resolution failed (e.g. sources table doesn't exist on a fresh
     // pre-init brain). Leave sourceId unset; engine read methods fall through
@@ -842,7 +834,6 @@ async function makeContext(engine: BrainEngine, params: Record<string, unknown>)
     // table). Matches dispatch.ts's auto-fill so the contract holds across
     // every transport.
     sourceId: sourceId ?? 'default',
-    ...(localFederated ? { localFederatedSourceIds: localFederated } : {}),
   };
 }
 

src/commands/extract.ts

@@ -1025,6 +1025,10 @@ async function extractForSlugs(
   let linksCreated = 0;
   let timelineCreated = 0;
   let pagesProcessed = 0;
+  // #2636: successfully processed pages get their extraction watermark
+  // stamped after the final flush (mode 'all' only — a partial-mode run
+  // hasn't done the full extraction the watermark asserts).
+  const processedRefs: Array<{ slug: string; source_id: string }> = [];
 
   // Issue #972: read the basename flag once per extract run.
   const globalBasename = await isGlobalBasenameEnabled(engine);
@@ -1113,6 +1117,7 @@ async function extractForSlugs(
         }
 
         pagesProcessed++;
+        if (!dryRun) processedRefs.push({ slug, source_id: sourceId ?? 'default' });
       } catch { /* skip unreadable */ }
       progress.tick(1);
     },
@@ -1120,6 +1125,13 @@ async function extractForSlugs(
 
   await flushLinks();
   await flushTimeline();
+  // #2636: the Dream cycle disables sync's inline extraction and routes
+  // changed slugs through this incremental path — without a stamp here,
+  // those pages never get links_extracted_at and stay permanently visible
+  // to `extract --stale` / doctor. Stamp only after BOTH batches flushed.
+  if (!dryRun && mode === 'all') {
+    await stampExtracted(engine, processedRefs);
+  }
   progress.finish();
 
   if (!jsonMode) {

src/core/operations.ts

@@ -424,23 +424,6 @@ export interface OperationContext {
    * satisfied even on single-source brains.
    */
   sourceId: string;
-  /**
-   * #2561 — federated read scope for UNQUALIFIED local CLI reads.
-   *
-   * Set ONLY by the local CLI's context builder (src/cli.ts makeContext), and
-   * only when the source resolved via a non-explicit tier (local_path /
-   * brain_default / sole_non_default / seed_default — NOT --source, NOT
-   * GBRAIN_SOURCE, NOT a .gbrain-source dotfile). Contains the resolved
-   * source first, then every other `config.federated = true` source, so an
-   * unqualified `gbrain search "X"` spans federated sources as
-   * docs/guides/multi-source-brains.md promises.
-   *
-   * Consumed exclusively by `federatedSearchScope` and ONLY when
-   * `ctx.remote === false` — a remote caller's scope stays governed by
-   * `ctx.auth.allowedSources` / scalar `ctx.sourceId` (source-isolation
-   * invariant, fail-closed).
-   */
-  localFederatedSourceIds?: string[];
 }
 
 /**
@@ -556,45 +539,6 @@ export function resolveRequestedScope(
   return sourceScopeOpts(ctx);
 }
 
-/**
- * #2561 — source scope for the search-shaped read ops (`search`, `query`).
- *
- * Delegates to `resolveRequestedScope` (the single trust+grant resolver), then
- * widens an UNQUALIFIED trusted-local scalar scope to the CLI-computed
- * federated set (`ctx.localFederatedSourceIds`, resolved source first). This is
- * what makes `sources add --federated` mean something for local search: a
- * federated source participates in unqualified `gbrain search "X"` results.
- *
- * The expansion NEVER applies when:
- *   - the caller is not strictly trusted-local (`ctx.remote !== false`) —
- *     remote scope stays grant-governed (fail-closed source isolation);
- *   - a per-call `source_id` was passed (explicit wins, including `__all__`);
- *   - the resolver already produced a federated array (OAuth grant);
- *   - the CLI resolved the source from an explicit signal (--source / env /
- *     dotfile) — makeContext leaves `localFederatedSourceIds` unset then.
- *
- * Deliberately NOT inside `sourceScopeOpts`: code-intel ops collapse a
- * multi-element scope to an error (`resolveCodeIntelScope`), and non-search
- * reads (get_page, get_links, …) keep their long-standing scalar behavior.
- */
-export function federatedSearchScope(
-  ctx: OperationContext,
-  sourceIdParam?: string,
-): { sourceId?: string; sourceIds?: string[] } {
-  const scope = resolveRequestedScope(ctx, sourceIdParam);
-  if (
-    ctx.remote === false &&
-    sourceIdParam === undefined &&
-    scope.sourceId !== undefined &&
-    scope.sourceIds === undefined &&
-    ctx.localFederatedSourceIds !== undefined &&
-    ctx.localFederatedSourceIds.length > 1
-  ) {
-    return { sourceIds: ctx.localFederatedSourceIds };
-  }
-  return scope;
-}
-
 /**
  * Code-intel adapter for `resolveRequestedScope`. Graph traversal
  * (code_callers/code_callees/code_blast/code_flow) is single-source by design —
@@ -1504,8 +1448,7 @@ const search: Operation = {
     const queryText = p.query as string;
     const limit = (p.limit as number) || 20;
     const offset = (p.offset as number) || 0;
-    // #2561: unqualified trusted-local search spans federated sources.
-    const scope = federatedSearchScope(ctx);
+    const scope = sourceScopeOpts(ctx);
 
     // T4/D5 — per-call mode honored ONLY for trusted/local callers so a remote
     // OAuth client can't escalate to the costly tokenmax bundle. Local + unknown
@@ -1667,9 +1610,7 @@ const query: Operation = {
     // is spread into BOTH the image-similarity searchVector path and the text
     // hybridSearch path below, so both honor the same grant.
     const sourceIdParam = typeof p.source_id === 'string' ? p.source_id : undefined;
-    // #2561: unqualified trusted-local query spans federated sources (per-call
-    // source_id / remote grants still resolve through resolveRequestedScope).
-    const querySourceScope = federatedSearchScope(ctx, sourceIdParam);
+    const querySourceScope = resolveRequestedScope(ctx, sourceIdParam);
 
     // v0.27.1: image-similarity branch. Bypasses hybridSearch (which is
     // text-only); embeds the image via embedMultimodal and runs a direct

src/core/source-resolver.ts

@@ -353,45 +353,6 @@ export async function resolveSourceWithTier(
   return { source_id: 'default', tier: 'seed_default' };
 }
 
-/**
- * #2561 — compute the federated read scope for an UNQUALIFIED local CLI call.
- *
- * `sources add --federated` promises that a `config.federated = true` source
- * "participates in unqualified `gbrain search` results"
- * (docs/guides/multi-source-brains.md). This helper turns that promise into a
- * scope: given the resolved source and WHICH tier resolved it, return
- * `[resolvedSource, ...other federated source ids]` — or `undefined` when the
- * expansion must not apply:
- *
- *   - explicit tiers (`flag` / `env` / `dotfile`): the user named a source;
- *     scalar scope stands (that IS the qualified case);
- *   - no other federated source exists: keep the scalar fast path unchanged.
- *
- * Archived sources are excluded (same rationale as pickSoleNonDefaultSource);
- * the archived column is v34+, so fall back to the un-archived query on older
- * brains. Callers put the result on `OperationContext.localFederatedSourceIds`
- * — consumed only by `federatedSearchScope` and only when `remote === false`.
- */
-export async function localFederatedSourceIds(
-  engine: BrainEngine,
-  sourceId: string,
-  tier: SourceTier,
-): Promise<string[] | undefined> {
-  if (tier === 'flag' || tier === 'env' || tier === 'dotfile') return undefined;
-  let rows: Array<{ id: string }>;
-  try {
-    rows = await engine.executeRaw<{ id: string }>(
-      `SELECT id FROM sources WHERE config->>'federated' = 'true' AND archived = false ORDER BY id`,
-    );
-  } catch {
-    rows = await engine.executeRaw<{ id: string }>(
-      `SELECT id FROM sources WHERE config->>'federated' = 'true' ORDER BY id`,
-    );
-  }
-  const ids = [sourceId, ...rows.map((r) => r.id).filter((id) => id !== sourceId)];
-  return ids.length > 1 ? ids : undefined;
-}
-
 /** Exposed for tests. */
 export const __testing = {
   readDotfileWalk,

test/extract-incremental.test.ts

@@ -60,6 +60,51 @@ async function seedPage(slug: string, body: string): Promise<void> {
 }
 
 describe('runExtractCore — incremental cycle path (#417)', () => {
+  test('Dream incremental all-mode stamps the source-scoped extraction watermark (#2636)', async () => {
+    await engine.executeRaw(
+      `INSERT INTO sources (id, name, local_path) VALUES ($1, $2, $3)`,
+      ['repo-a', 'repo-a', tempDir],
+    );
+    await engine.putPage('people/alice-example', {
+      type: 'person',
+      title: 'alice-example',
+      compiled_truth: '# alice',
+      timeline: '',
+      frontmatter: {},
+      content_hash: 'h',
+    }, { sourceId: 'repo-a' });
+    writeFileSync(join(tempDir, 'people/alice-example.md'), '# alice');
+
+    await runExtractCore(engine as unknown as BrainEngine, {
+      mode: 'all',
+      dir: tempDir,
+      slugs: ['people/alice-example'],
+      sourceId: 'repo-a',
+    });
+
+    const rows = await engine.executeRaw<{ links_extracted_at: string | null }>(
+      `SELECT links_extracted_at FROM pages WHERE slug = $1 AND source_id = $2`,
+      ['people/alice-example', 'repo-a'],
+    );
+    expect(rows[0]?.links_extracted_at).not.toBeNull();
+    expect(await engine.countStalePagesForExtraction({ sourceId: 'repo-a' })).toBe(0);
+  });
+
+  test('Dream incremental dry-run does NOT stamp the watermark', async () => {
+    await seedPage('people/alice-example', '# alice');
+    await runExtractCore(engine as unknown as BrainEngine, {
+      mode: 'all',
+      dir: tempDir,
+      slugs: ['people/alice-example'],
+      dryRun: true,
+    });
+    const rows = await engine.executeRaw<{ links_extracted_at: string | null }>(
+      `SELECT links_extracted_at FROM pages WHERE slug = $1`,
+      ['people/alice-example'],
+    );
+    expect(rows[0]?.links_extracted_at ?? null).toBeNull();
+  });
+
   test('1. slugs: [] returns immediately with zero counts (early-return path)', async () => {
     await seedPage('people/alice-example', '# alice');
     const result = await runExtractCore(engine as unknown as BrainEngine, {

test/local-federated-search-scope.test.ts

@@ -1,154 +0,0 @@
-/**
- * #2561 — sources.config.federated participates in UNQUALIFIED local CLI
- * search/query.
- *
- * Pre-fix: the local CLI always emitted a scalar `{sourceId}` scope (required
- * field, auto-filled 'default'), so a source registered with
- * `gbrain sources add --federated` was invisible to an unqualified
- * `gbrain search "X"` — contradicting docs/guides/multi-source-brains.md
- * ("Source participates in unqualified `gbrain search` results").
- *
- * Fix: the CLI context builder computes `ctx.localFederatedSourceIds`
- * (resolved source + every other federated source) whenever the source
- * resolved via a NON-explicit tier; `federatedSearchScope` widens the scalar
- * scope to that set for the `search` / `query` ops — trusted-local only
- * (`ctx.remote === false`), never for remote callers, never when a per-call
- * `source_id` or an explicit --source/env/dotfile was given.
- */
-import { describe, test, expect, beforeAll, afterAll } from 'bun:test';
-import { PGLiteEngine } from '../src/core/pglite-engine.ts';
-import { localFederatedSourceIds } from '../src/core/source-resolver.ts';
-import {
-  federatedSearchScope,
-  operations,
-  type OperationContext,
-} from '../src/core/operations.ts';
-
-let engine: PGLiteEngine;
-const search = operations.find((o) => o.name === 'search')!;
-
-function ctxOf(overrides: Partial<OperationContext> = {}): OperationContext {
-  return {
-    engine: engine as any,
-    config: {} as any,
-    logger: console as any,
-    dryRun: false,
-    remote: false,
-    sourceId: 'default',
-    ...overrides,
-  };
-}
-
-beforeAll(async () => {
-  engine = new PGLiteEngine();
-  await engine.connect({});
-  await engine.initSchema();
-  // Seeded 'default' source is federated=true. Add:
-  //   wiki    — federated (must join unqualified search)
-  //   private — NOT federated (must stay invisible unless explicitly named)
-  //   oldnews — federated but archived (must stay excluded)
-  await engine.executeRaw(
-    `INSERT INTO sources (id, name, local_path, config) VALUES ('wiki', 'wiki', '/tmp/wiki', '{"federated": true}'::jsonb)`,
-  );
-  await engine.executeRaw(
-    `INSERT INTO sources (id, name, local_path, config) VALUES ('private', 'private', '/tmp/private', '{}'::jsonb)`,
-  );
-  await engine.executeRaw(
-    `INSERT INTO sources (id, name, local_path, config, archived) VALUES ('oldnews', 'oldnews', '/tmp/oldnews', '{"federated": true}'::jsonb, true)`,
-  );
-  const pages: Array<[slug: string, sourceId: string, where: string]> = [
-    ['notes/home', 'default', 'default'],
-    ['wiki/topic', 'wiki', 'wiki'],
-    ['private/topic', 'private', 'private'],
-    ['old/topic', 'oldnews', 'oldnews'],
-  ];
-  for (const [slug, sourceId, where] of pages) {
-    await engine.putPage(slug, {
-      type: 'note', title: `Topic in ${where}`, compiled_truth: `the zebra telescope in ${where}`, frontmatter: {},
-    }, { sourceId });
-    await engine.upsertChunks(slug, [
-      { chunk_index: 0, chunk_text: `the zebra telescope in ${where}`, chunk_source: 'compiled_truth' },
-    ], { sourceId });
-  }
-  // Keyword-only search path: no embedding provider needed in tests.
-  await engine.setConfig('search.mcp_keyword_only', 'true');
-}, 60_000);
-
-afterAll(async () => {
-  if (engine) await engine.disconnect();
-}, 60_000);
-
-describe('localFederatedSourceIds — CLI-side scope computation', () => {
-  test('non-explicit tier: resolved source first, then other federated, archived excluded', async () => {
-    expect(await localFederatedSourceIds(engine, 'default', 'seed_default')).toEqual(['default', 'wiki']);
-  });
-
-  test('non-federated resolved source still joins its own scope', async () => {
-    expect(await localFederatedSourceIds(engine, 'private', 'brain_default')).toEqual(['private', 'default', 'wiki']);
-  });
-
-  test('explicit tiers (--source / env / dotfile) never expand', async () => {
-    expect(await localFederatedSourceIds(engine, 'default', 'flag')).toBeUndefined();
-    expect(await localFederatedSourceIds(engine, 'default', 'env')).toBeUndefined();
-    expect(await localFederatedSourceIds(engine, 'default', 'dotfile')).toBeUndefined();
-  });
-
-  test('single federated source (the resolved one) keeps the scalar fast path', async () => {
-    const solo = { executeRaw: async () => [{ id: 'default' }] } as any;
-    expect(await localFederatedSourceIds(solo, 'default', 'seed_default')).toBeUndefined();
-  });
-});
-
-describe('federatedSearchScope — trust + explicitness matrix', () => {
-  test('trusted local + unqualified widens to the federated set', () => {
-    const ctx = ctxOf({ localFederatedSourceIds: ['default', 'wiki'] });
-    expect(federatedSearchScope(ctx)).toEqual({ sourceIds: ['default', 'wiki'] });
-  });
-
-  test('remote caller NEVER widens (fail-closed), even if the field is set', () => {
-    const ctx = ctxOf({ remote: true, localFederatedSourceIds: ['default', 'wiki'] });
-    expect(federatedSearchScope(ctx)).toEqual({ sourceId: 'default' });
-  });
-
-  test('per-call source_id wins over the federated set', () => {
-    const ctx = ctxOf({ localFederatedSourceIds: ['default', 'wiki'] });
-    expect(federatedSearchScope(ctx, 'wiki')).toEqual({ sourceId: 'wiki' });
-  });
-
-  test('per-call __all__ keeps the whole-brain semantics for trusted local', () => {
-    const ctx = ctxOf({ localFederatedSourceIds: ['default', 'wiki'] });
-    expect(federatedSearchScope(ctx, '__all__')).toEqual({});
-  });
-
-  test('a federated OAuth grant wins over the local set', () => {
-    const ctx = ctxOf({
-      localFederatedSourceIds: ['default', 'wiki'],
-      auth: { allowedSources: ['a', 'b'] } as OperationContext['auth'],
-    });
-    expect(federatedSearchScope(ctx)).toEqual({ sourceIds: ['a', 'b'] });
-  });
-
-  test('no local federated set → unchanged scalar scope', () => {
-    expect(federatedSearchScope(ctxOf())).toEqual({ sourceId: 'default' });
-  });
-});
-
-describe('search op — unqualified local search spans federated sources', () => {
-  test('federated source results appear; non-federated + archived stay invisible', async () => {
-    const ctx = ctxOf({
-      localFederatedSourceIds: await localFederatedSourceIds(engine, 'default', 'seed_default'),
-    });
-    const results = (await search.handler(ctx, { query: 'zebra telescope' })) as Array<{ slug: string }>;
-    const slugs = results.map((r) => r.slug);
-    expect(slugs).toContain('notes/home');
-    expect(slugs).toContain('wiki/topic'); // pre-#2561 this was missing
-    expect(slugs).not.toContain('private/topic');
-    expect(slugs).not.toContain('old/topic');
-  });
-
-  test('explicit source resolution (no federated set on ctx) stays single-source', async () => {
-    const results = (await search.handler(ctxOf(), { query: 'zebra telescope' })) as Array<{ slug: string }>;
-    const slugs = results.map((r) => r.slug);
-    expect(slugs).toEqual(['notes/home']);
-  });
-});