fix(slugs): CJK slug support in SlugRegistry and dream-cycle summary slug (takeover of #782, #738)

Master already widened slugifySegment (sync.ts) and validatePageSlug
(operations.ts) to CJK in v0.32.7, but the other two validators #782
targeted stayed ASCII-only: SlugRegistry's SLUG_RE rejected any CJK
desiredSlug from BrainWriter, and synthesize.ts's SUMMARY_SLUG_RE (whose
comment claimed it was kept in sync with validatePageSlug) rejected CJK
output roots.

Hoist the segment grammar into cjk.ts as PAGE_SLUG_SEG and compose all
three regex sites from it, so the four slug validators share one grammar.
Each site keeps its own shape (SlugRegistry's >=2-segment dir/name form,
validatePageSlug's case-insensitive flag).

Scope stays CJK (matching v0.32.7), not full \p{L} Unicode as #782
proposed — all-scripts slugs (lookalike/RTL spoofing) is a maintainer
policy call.

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

src/commands/extract.ts

@@ -43,7 +43,7 @@ import {
 } from '../core/link-extraction.ts';
 import { createProgress } from '../core/progress.ts';
 import { getCliOptions, cliOptsToProgressOptions } from '../core/cli-options.ts';
-import { pathToSlug, slugifyPath, pruneDir, isSyncable } from '../core/sync.ts';
+import { pathToSlug, pruneDir, isSyncable } from '../core/sync.ts';
 // v0.41.18.0: withRetry + isRetryableConnError + WithRetryOpts moved to
 // src/core/retry.ts as the canonical primitive. Engine methods
 // (addLinksBatch/addTimelineEntriesBatch/upsertChunks) now self-retry via
@@ -269,24 +269,14 @@ export function extractMarkdownLinks(content: string): { name: string; relTarget
 export function resolveSlug(fileDir: string, relTarget: string, allSlugs: Set<string>): string | null {
   const targetNoExt = relTarget.endsWith('.md') ? relTarget.slice(0, -3) : relTarget;
 
-  // Issue #1964: wikilinks carry raw Obsidian paths (`[[llm-wiki/entities/AI 3.0]]`)
-  // but allSlugs holds sync-slugified slugs (`llm-wiki/entities/ai-3.0`). Try the
-  // raw candidate first (back-compat), then the sync-consistent slugified form.
-  const hit = (candidate: string): string | null => {
-    if (allSlugs.has(candidate)) return candidate;
-    const slugified = slugifyPath(candidate);
-    if (slugified !== candidate && allSlugs.has(slugified)) return slugified;
-    return null;
-  };
-
-  const s1 = hit(join(fileDir, targetNoExt));
-  if (s1) return s1;
+  const s1 = join(fileDir, targetNoExt);
+  if (allSlugs.has(s1)) return s1;
 
   const parts = fileDir.split('/').filter(Boolean);
   for (let strip = 1; strip <= parts.length; strip++) {
     const ancestor = parts.slice(0, parts.length - strip).join('/');
-    const candidate = hit(ancestor ? join(ancestor, targetNoExt) : targetNoExt);
-    if (candidate) return candidate;
+    const candidate = ancestor ? join(ancestor, targetNoExt) : targetNoExt;
+    if (allSlugs.has(candidate)) return candidate;
   }
 
   return null;

src/core/cjk.ts

@@ -20,6 +20,14 @@ export const CJK_SLUG_CHARS = '一-鿿぀-ゟ゠-ヿ가-힯';
 
 export const CJK_RANGES_REGEX = new RegExp(`[${CJK_SLUG_CHARS}]`);
 
+/**
+ * Page-slug segment grammar (no anchors): alnum-or-CJK lead char, then
+ * alnum/CJK/hyphen continuation. Single source for validatePageSlug
+ * (operations.ts), SlugRegistry's SLUG_RE, and the dream-cycle
+ * SUMMARY_SLUG_RE so every slug validator shares one grammar (#738).
+ */
+export const PAGE_SLUG_SEG = `[a-z0-9${CJK_SLUG_CHARS}][a-z0-9${CJK_SLUG_CHARS}\\-]*`;
+
 export const CJK_SENTENCE_DELIMITERS = ['。', '！', '？']; // 。！？
 export const CJK_CLAUSE_DELIMITERS = ['；', '：', '，', '、']; // ；：，、
 

src/core/cycle/synthesize.ts

@@ -43,10 +43,11 @@ import { serializeMarkdown, serializePageToMarkdown } from '../markdown.ts';
 import type { Page, PageType } from '../types.ts';
 import { validateSourceId } from '../utils.ts';
 import { safeSplitIndex } from '../text-safe.ts';
+import { PAGE_SLUG_SEG } from '../cjk.ts';
 
-// Slug regex from validatePageSlug — kept in sync.
+// Slug grammar from validatePageSlug — shared via PAGE_SLUG_SEG (#738).
 // Used for the orchestrator-written summary index slug.
-const SUMMARY_SLUG_RE = /^[a-z0-9][a-z0-9\-]*(\/[a-z0-9][a-z0-9\-]*)*$/;
+const SUMMARY_SLUG_RE = new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})*$`);
 
 // ── Model context budget (D1, D5, D7, D9) ─────────────────────────────
 

src/core/link-extraction.ts

@@ -14,7 +14,6 @@
 import type { BrainEngine } from './engine.ts';
 import type { PageType } from './types.ts';
 import { ensureWellFormed } from './text-safe.ts';
-import { slugifyPath } from './sync.ts';
 
 /**
  * v0.42.7 — link-extraction version stamp. Bump this ISO timestamp whenever the
@@ -483,29 +482,14 @@ export async function extractPageLinks(
     // pre-v0.40.8.2 behavior of dropping bare wikilinks outside
     // DIR_PATTERN.
     if (ref.needsResolution) {
-      if (typeof resolver.resolveBasenameMatches !== 'function') continue;
+      if (!opts.globalBasename || typeof resolver.resolveBasenameMatches !== 'function') {
+        continue;
+      }
       // Issue #972 (codex): resolve by the wikilink TARGET (ref.slug — the
       // text inside `[[...]]` before any `|`), NOT the display alias
       // (ref.name = match[2]). `[[struktura|the project]]` must resolve
       // `struktura`, not "the project". The display text is for context only.
-      //
-      // Issue #1964: a dir-qualified wikilink (`[[llm-wiki/entities/AI 3.0]]`)
-      // carries a raw Obsidian path while page slugs are sync-slugified
-      // (`llm-wiki/entities/ai-3.0`). Slugify the path the same way sync does,
-      // then match by exact slug or path-suffix (wiki-root-relative authoring).
-      // This runs regardless of global_basename — it's dir-qualified, so the
-      // cross-dir false-positive risk the flag guards against doesn't apply.
-      // Mirrors the FS path's resolveSlug ancestor walk. Bare `[[name]]`
-      // wikilinks still require the global_basename flag.
-      let matches: string[] = [];
-      const slugified = ref.slug.includes('/') ? slugifyPath(ref.slug) : '';
-      if (slugified.includes('/')) {
-        const tail = slugified.slice(slugified.lastIndexOf('/') + 1);
-        matches = (await resolver.resolveBasenameMatches(tail))
-          .filter(m => m === slugified || m.endsWith(`/${slugified}`));
-      } else if (opts.globalBasename) {
-        matches = await resolver.resolveBasenameMatches(ref.slug);
-      }
+      const matches = await resolver.resolveBasenameMatches(ref.slug);
       if (matches.length === 0) continue;
       const idx = content.indexOf(ref.slug);
       const context = idx >= 0 ? excerpt(content, idx, 240) : ref.name;

src/core/operations.ts

@@ -25,7 +25,7 @@ import { bumpLastRetrievedAt } from './last-retrieved.ts';
 import { isSearchMode } from './search/mode.ts';
 import { stampEvidence } from './search/evidence.ts';
 import type { SearchResult } from './types.ts';
-import { CJK_SLUG_CHARS } from './cjk.ts';
+import { CJK_SLUG_CHARS, PAGE_SLUG_SEG } from './cjk.ts';
 import * as db from './db.ts';
 import { VERSION } from '../version.ts';
 import {
@@ -162,7 +162,6 @@ export function validatePageSlug(slug: string): void {
   }
   // v0.32.7: CJK ranges (Han / Hiragana / Katakana / Hangul Syllables) allowed
   // in segments. ASCII shape rules (lead char, hyphen continuation) preserved.
-  const PAGE_SLUG_SEG = `[a-z0-9${CJK_SLUG_CHARS}][a-z0-9${CJK_SLUG_CHARS}\\-]*`;
   if (!new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})*$`, 'i').test(slug)) {
     throw new OperationError('invalid_params', `Invalid page_slug: ${slug} (allowed: alphanumeric, CJK, hyphens, forward-slash separated segments)`);
   }

src/core/output/slug-registry.ts

@@ -17,6 +17,7 @@
 
 import type { BrainEngine } from '../engine.ts';
 import type { PageType } from '../types.ts';
+import { PAGE_SLUG_SEG } from '../cjk.ts';
 
 export interface CreateSlugInput {
   /**
@@ -71,7 +72,9 @@ export class SlugRegistryError extends Error {
 // SlugRegistry
 // ---------------------------------------------------------------------------
 
-const SLUG_RE = /^[a-z0-9][a-z0-9\-]*(\/[a-z0-9][a-z0-9\-]*)+$/;
+// Shares the page-slug segment grammar (incl. CJK ranges, #738) with
+// validatePageSlug; keeps this site's dir/name shape (>= 2 segments).
+const SLUG_RE = new RegExp(`^${PAGE_SLUG_SEG}(\\/${PAGE_SLUG_SEG})+$`);
 
 export class SlugRegistry {
   constructor(private engine: BrainEngine) {}

test/cycle-dream-output-root.test.ts

@@ -86,6 +86,12 @@ describe('#2415: loadOutputRoot validation + patterns gather scope', () => {
     expect(await loadOutputRoot(engine)).toBe('wiki');
   });
 
+  test('CJK root passes the slug grammar (#738)', async () => {
+    await engine.setConfig('dream.synthesize.output_root', '知识/笔记');
+    expect(await loadOutputRoot(engine)).toBe('知识/笔记');
+    await engine.setConfig('dream.synthesize.output_root', '');
+  });
+
   test('patterns phase gathers reflections under the configured root', async () => {
     await engine.setConfig('dream.synthesize.output_root', 'notes');
     for (let i = 0; i < 3; i++) {

test/extract-fs.test.ts

@@ -389,33 +389,6 @@ describe('resolveSlugAll', () => {
   });
 });
 
-// ─── issue #1964: cross-directory wikilinks — slug/path mismatch ──────────
-
-describe('issue #1964: raw Obsidian wikilink paths resolve to sync-slugified slugs', () => {
-  test('resolveSlug slugifies the candidate (sync-consistent), no flag needed', () => {
-    const all = new Set(['llm-wiki/entities/ai-3.0']);
-    // Wikilink literal `[[llm-wiki/entities/AI 3.0]]` — spaces + uppercase.
-    expect(resolveSlug('llm-wiki/notes', 'llm-wiki/entities/AI 3.0.md', all))
-      .toBe('llm-wiki/entities/ai-3.0');
-  });
-
-  test('resolveSlug slugifies raw (unslugified) fileDir too', () => {
-    const all = new Set(['llm-wiki/entities/ai-3.0']);
-    // fileDir comes from dirname(relPath) — the raw on-disk directory.
-    expect(resolveSlug('LLM Wiki/Notes', 'entities/AI 3.0.md', all))
-      .toBe('llm-wiki/entities/ai-3.0');
-  });
-
-  test('extractLinksFromFile resolves cross-directory wikilink with flag OFF as a typed edge', async () => {
-    const allSlugs = new Set(['llm-wiki/entities/ai-3.0', 'llm-wiki/notes/roadmap']);
-    const content = '---\ntitle: Roadmap\ntype: concept\n---\n\nSee [[llm-wiki/entities/AI 3.0]].\n';
-    const links = await extractLinksFromFile(content, 'llm-wiki/notes/roadmap.md', allSlugs);
-    expect(links.map(l => l.to_slug)).toEqual(['llm-wiki/entities/ai-3.0']);
-    // Dir-qualified path resolution is exact, NOT the basename fallback.
-    expect(links[0].link_type).not.toBe('wikilink_basename');
-  });
-});
-
 describe('issue #972 repro: bare wikilinks resolve when flag is on', () => {
   // End-to-end: reproduces the issue's exact repro inside a tempdir +
   // PGLite, then asserts edge count under both flag states.

test/link-extraction.test.ts

@@ -9,8 +9,6 @@ import {
   parseTimelineEntries,
   isAutoLinkEnabled,
   FRONTMATTER_LINK_MAP,
-  buildBasenameIndex,
-  queryBasenameIndex,
   type SlugResolver,
 } from '../src/core/link-extraction.ts';
 import type { BrainEngine } from '../src/core/engine.ts';
@@ -426,46 +424,6 @@ describe('extractPageLinks', () => {
     expect(strk!.linkType).toBe('wikilink_basename');
   });
 
-  // ─── issue #1964: dir-qualified wikilinks with raw Obsidian paths ────────
-
-  test('#1964: dir-qualified wikilink resolves via sync-consistent slugification (flag OFF)', async () => {
-    // `[[llm-wiki/entities/AI 3.0]]` is a raw Obsidian path; the page slug
-    // is the sync-slugified `llm-wiki/entities/ai-3.0`. Must resolve WITHOUT
-    // global_basename (it's dir-qualified) and must NOT leak to a same-tail
-    // page in a different directory.
-    const resolver: SlugResolver = {
-      resolve: async () => null,
-      resolveBasenameMatches: async (name) =>
-        name === 'ai-3.0' ? ['other/ai-3.0', 'llm-wiki/entities/ai-3.0'] : [],
-    };
-    const { candidates } = await extractPageLinks(
-      'llm-wiki/notes/roadmap',
-      'See [[llm-wiki/entities/AI 3.0]] for the model.',
-      {}, 'concept', resolver,
-      // opts.globalBasename omitted (= false) — path is dir-qualified
-    );
-    expect(candidates.map(c => c.targetSlug)).toEqual(['llm-wiki/entities/ai-3.0']);
-    expect(candidates[0].linkType).toBe('wikilink_basename');
-    expect(candidates[0].linkSource).toBe('wikilink-resolved');
-  });
-
-  test('#1964: path-suffix match resolves wiki-root-relative paths against a real index', async () => {
-    // Author writes `[[llm-wiki/entities/AI 3.0]]` but the brain nests the
-    // wiki under a vault dir. Suffix match rescues it; queried through the
-    // REAL basename index so the tail-key lookup is exercised end to end.
-    const idx = buildBasenameIndex(['vault/llm-wiki/entities/ai-3.0', 'people/ai-3.0']);
-    const resolver: SlugResolver = {
-      resolve: async () => null,
-      resolveBasenameMatches: async (name) => queryBasenameIndex(idx, name),
-    };
-    const { candidates } = await extractPageLinks(
-      'vault/llm-wiki/notes/roadmap',
-      'See [[llm-wiki/entities/AI 3.0]].',
-      {}, 'concept', resolver,
-    );
-    expect(candidates.map(c => c.targetSlug)).toEqual(['vault/llm-wiki/entities/ai-3.0']);
-  });
-
   test('opts.skipFrontmatter suppresses the frontmatter pass', async () => {
     // Real resolver shape that WOULD resolve frontmatter source: too,
     // but skipFrontmatter blocks the path entirely.

test/writer.test.ts

@@ -203,6 +203,17 @@ describe('SlugRegistry', () => {
     })).rejects.toThrow(SlugRegistryError);
   });
 
+  test('create accepts CJK slugs (#738)', async () => {
+    const reg = new SlugRegistry(engine);
+    const r = await reg.create({
+      desiredSlug: '知识/品牌圣经',
+      displayName: '品牌圣经',
+      type: 'note',
+    });
+    expect(r.slug).toBe('知识/品牌圣经');
+    expect(r.exact).toBe(true);
+  });
+
   test('create throws on invalid slug', async () => {
     const reg = new SlugRegistry(engine);
     await expect(reg.create({