test(unit): migrate batch2 document and parser suites to vitest

tests/unit/canonicalize-epub-segment.vitest.spec.ts

@@ -1,4 +1,4 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 
 import {
   canonicalizeEpubSegmentAgainstSpineText,
@@ -6,7 +6,7 @@ import {
 } from '../../src/lib/client/epub/canonicalize-epub-segment';
 import { planCanonicalTtsSegments } from '../../src/lib/shared/tts-segment-plan';
 
-test.describe('canonicalizeEpubSegmentAgainstSpineText', () => {
+describe('canonicalizeEpubSegmentAgainstSpineText', () => {
   test('maps an exact sentence to the canonical segment identity', () => {
     const spineText = [
       'First section sentence with enough words to stand alone.',
@@ -95,7 +95,7 @@ test.describe('canonicalizeEpubSegmentAgainstSpineText', () => {
   });
 });
 
-test.describe('canonicalizeEpubSegmentsAgainstSpineText', () => {
+describe('canonicalizeEpubSegmentsAgainstSpineText', () => {
   test('maps overlap-boundary drift sentences to forward canonical segments', () => {
     const sourceSentences = [
       'The star was particularly bright when the station lights switched off for cycle night.',

tests/unit/docstore.vitest.spec.ts

@@ -1,10 +1,10 @@
-import { test, expect } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { getMigratedDocumentFileName } from '../../src/lib/server/storage/docstore-legacy';
 
-test.describe('Docstore Filename Safety', () => {
+describe('docstore filename safety', () => {
   const id = 'a'.repeat(64); // Simulate sha256 hex ID
 
-  test('should generate standard filename for short names', () => {
+  test('generates stable file names for short inputs', () => {
     const name = 'test-file.pdf';
     const result = getMigratedDocumentFileName(id, name);
     expect(result).toBe(`${id}__test-file.pdf`);
@@ -50,4 +50,13 @@ test.describe('Docstore Filename Safety', () => {
      expect(result.length).toBe(240);
      expect(result).not.toContain('truncated-');
   });
+
+  test('drops path traversal and null-byte fragments from migrated file names', () => {
+    const dirtyName = '../nested/\u0000financial-report.pdf';
+    const result = getMigratedDocumentFileName(id, dirtyName);
+
+    expect(result).toBe(`${id}__financial-report.pdf`);
+    expect(result).not.toContain('..');
+    expect(result).not.toContain('\u0000');
+  });
 });

tests/unit/document-cache.vitest.spec.ts

@@ -1,16 +1,17 @@
-import { test, expect } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import type { BaseDocument, EPUBDocument, HTMLDocument, PDFDocument } from '../../src/types/documents';
 import { ensureCachedDocumentCore } from '../../src/lib/client/cache/documents';
+import { makeBaseDocument } from './support/document-fixtures';
 
-test.describe('document-cache-core', () => {
+describe('document-cache-core', () => {
   test('returns cached PDF without downloading', async () => {
-    const meta: BaseDocument = {
+    const meta: BaseDocument = makeBaseDocument({
       id: 'pdf1',
       name: 'a.pdf',
       size: 10,
-      lastModified: Date.now(),
+      lastModified: 1_700_000_000_001,
       type: 'pdf',
-    };
+    });
 
     let downloads = 0;
     const cached: PDFDocument = { ...meta, type: 'pdf', data: new ArrayBuffer(1) };
@@ -32,13 +33,13 @@ test.describe('document-cache-core', () => {
   });
 
   test('downloads and stores on cache miss (EPUB)', async () => {
-    const meta: BaseDocument = {
+    const meta: BaseDocument = makeBaseDocument({
       id: 'epub1',
       name: 'b.epub',
       size: 10,
-      lastModified: Date.now(),
+      lastModified: 1_700_000_000_002,
       type: 'epub',
-    };
+    });
 
     const store = new Map<string, PDFDocument | EPUBDocument | HTMLDocument>();
     let downloads = 0;
@@ -63,13 +64,13 @@ test.describe('document-cache-core', () => {
   });
 
   test('downloads, decodes, and stores HTML on cache miss', async () => {
-    const meta: BaseDocument = {
+    const meta: BaseDocument = makeBaseDocument({
       id: 'html1',
       name: 'c.txt',
       size: 5,
-      lastModified: Date.now(),
+      lastModified: 1_700_000_000_003,
       type: 'html',
-    };
+    });
 
     const store = new Map<string, PDFDocument | EPUBDocument | HTMLDocument>();
     let decodedCalls = 0;
@@ -92,4 +93,22 @@ test.describe('document-cache-core', () => {
     expect(result.type).toBe('html');
     expect((result as HTMLDocument).data).toBe('hello');
   });
+
+  test('throws deterministic cache error when download succeeds but backend misses persisted PDF', async () => {
+    const meta = makeBaseDocument({
+      id: 'pdf-cache-miss',
+      type: 'pdf',
+    });
+
+    await expect(
+      ensureCachedDocumentCore(meta, {
+        get: async () => null,
+        putPdf: async () => { /* simulate write path without persisted row */ },
+        putEpub: async () => { /* unused */ },
+        putHtml: async () => { /* unused */ },
+        download: async () => new Uint8Array([1, 2, 3]).buffer,
+        decodeText: () => '',
+      }),
+    ).rejects.toThrow('Failed to cache PDF');
+  });
 });

tests/unit/html-audiobook-adapter.vitest.spec.ts

@@ -1,11 +1,11 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { createHtmlAudiobookSourceAdapter } from '../../src/lib/client/audiobooks/adapters/html';
 import { parseHtmlBlocks, type HtmlBlock } from '../../src/lib/client/html/blocks';
 
 const blocksFromMd = (src: string): HtmlBlock[] => parseHtmlBlocks(src, false);
 const blocksFromTxt = (src: string): HtmlBlock[] => parseHtmlBlocks(src, true);
 
-test.describe('createHtmlAudiobookSourceAdapter (markdown chapter splitting)', () => {
+describe('createHtmlAudiobookSourceAdapter (markdown chapter splitting)', () => {
   test('starts a new chapter at each h1/h2 heading by default', async () => {
     const blocks = blocksFromMd(
       [
@@ -95,7 +95,7 @@ test.describe('createHtmlAudiobookSourceAdapter (markdown chapter splitting)', (
   });
 });
 
-test.describe('createHtmlAudiobookSourceAdapter (txt chapter splitting)', () => {
+describe('createHtmlAudiobookSourceAdapter (txt chapter splitting)', () => {
   test('chunks TXT documents into "Part N" of 50 blocks by default', async () => {
     const blocks = blocksFromTxt(
       Array.from({ length: 120 }, (_, i) => `Block ${i + 1}.`).join('\n\n'),
@@ -130,7 +130,7 @@ test.describe('createHtmlAudiobookSourceAdapter (txt chapter splitting)', () =>
   });
 });
 
-test.describe('createHtmlAudiobookSourceAdapter — prepareChapter', () => {
+describe('createHtmlAudiobookSourceAdapter — prepareChapter', () => {
   test('returns the same chapter by index that prepareChapters lists', async () => {
     const blocks = blocksFromMd(['# A', '', 'one', '', '## B', '', 'two'].join('\n'));
     const adapter = createHtmlAudiobookSourceAdapter({

tests/unit/html-blocks.vitest.spec.ts

@@ -1,4 +1,4 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import {
   mdToPlainText,
   parseHtmlBlocks,
@@ -6,7 +6,7 @@ import {
   splitTxtBlocks,
 } from '../../src/lib/client/html/blocks';
 
-test.describe('parseHtmlBlocks (markdown)', () => {
+describe('parseHtmlBlocks (markdown)', () => {
   test('splits headings, paragraphs, and lists into separate blocks', () => {
     const src = [
       '# Title',
@@ -55,7 +55,7 @@ test.describe('parseHtmlBlocks (markdown)', () => {
   });
 });
 
-test.describe('parseHtmlBlocks (txt)', () => {
+describe('parseHtmlBlocks (txt)', () => {
   test('splits on blank-line boundaries and preserves intra-block whitespace', () => {
     const src = 'First block\nmore.\n\nSecond block.\n\n\nThird block.';
     const blocks = parseHtmlBlocks(src, true);
@@ -73,7 +73,7 @@ test.describe('parseHtmlBlocks (txt)', () => {
   });
 });
 
-test.describe('mdToPlainText badge/image stripping', () => {
+describe('mdToPlainText badge/image stripping', () => {
   // The bug: badge alt text was being kept in plainText. Since the rendered
   // DOM is just an <img> with no text node, the sentence-highlight pattern
   // matcher couldn't find those words and the WHOLE first-segment match
@@ -124,7 +124,7 @@ test.describe('mdToPlainText badge/image stripping', () => {
   });
 });
 
-test.describe('buildFullDocumentText-style integration (badge-only blocks)', () => {
+describe('buildFullDocumentText-style integration (badge-only blocks)', () => {
   // If a paragraph is composed only of badges, its plainText is empty after
   // stripping. The reader filters empty plainText out of the TTS source
   // (`useHtmlDocument#buildFullDocumentText`), so badge blocks don't generate

tests/unit/pdf-audiobook-adapter.vitest.spec.ts

@@ -1,15 +1,15 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { createPdfAudiobookSourceAdapter } from '../../src/lib/client/audiobooks/adapters/pdf';
 import type { ParsedPdfDocument } from '../../src/types/parsed-pdf';
 import type { DocumentSettings } from '../../src/types/document-settings';
 
-test.describe('pdf audiobook adapter', () => {
+describe('pdf audiobook adapter', () => {
   test('builds chapters from paragraph titles and filters skipped kinds', async () => {
     const parsed: ParsedPdfDocument = {
       schemaVersion: 1,
       documentId: 'doc-1',
       parserVersion: 'test',
-      parsedAt: Date.now(),
+      parsedAt: 1_700_000_000_000,
       pages: [
         {
           pageNumber: 1,
@@ -77,7 +77,7 @@ test.describe('pdf audiobook adapter', () => {
       schemaVersion: 1,
       documentId: 'doc-2',
       parserVersion: 'test',
-      parsedAt: Date.now(),
+      parsedAt: 1_700_000_000_001,
       pages: [
         {
           pageNumber: 1,

tests/unit/pdf-build-page-text.vitest.spec.ts

@@ -1,8 +1,8 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { buildPageTextFromBlocks } from '../../src/lib/client/pdf-block-text';
 import type { ParsedPdfPage } from '../../src/types/parsed-pdf';
 
-test.describe('buildPageTextFromBlocks', () => {
+describe('buildPageTextFromBlocks', () => {
   test('filters skipped kinds and preserves reading order', () => {
     const page: ParsedPdfPage = {
       pageNumber: 1,

tests/unit/pdf-force-reparse.vitest.spec.ts

@@ -1,4 +1,4 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import {
   FORCE_REPARSE_CONFIRM_MESSAGE,
   FORCE_REPARSE_CONFIRM_TEXT,
@@ -6,7 +6,7 @@ import {
   isForceReparseDisabled,
 } from '../../src/lib/client/pdf/force-reparse';
 
-test.describe('pdf force reparse controls', () => {
+describe('pdf force reparse controls', () => {
   test('disables action while parse is pending or running', () => {
     expect(isForceReparseDisabled('pending')).toBeTruthy();
     expect(isForceReparseDisabled('running')).toBeTruthy();

tests/unit/pdf-merge-text-with-regions.vitest.spec.ts

@@ -1,7 +1,7 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { mergeTextWithRegions } from '@openreader/compute-core';
 
-test.describe('mergeTextWithRegions', () => {
+describe('mergeTextWithRegions', () => {
   test('assigns text items to containing regions by centroid and joins in reading order', () => {
     const regions = [
       { bbox: [0, 0, 100, 50] as [number, number, number, number], label: 'text' as const },
@@ -20,4 +20,18 @@ test.describe('mergeTextWithRegions', () => {
     expect(merged[0].text).toBe('hello world');
     expect(merged[1].text).toBe('Figure 1.2');
   });
+
+  test('drops text whose centroid is outside every region', () => {
+    const regions = [
+      { bbox: [0, 0, 50, 50] as [number, number, number, number], label: 'text' as const },
+    ];
+    const textItems = [
+      { text: 'inside', x: 10, y: 10, width: 10, height: 8 },
+      { text: 'outside', x: 80, y: 80, width: 12, height: 8 },
+    ];
+
+    const merged = mergeTextWithRegions(regions, textItems);
+    expect(merged).toHaveLength(1);
+    expect(merged[0].text).toBe('inside');
+  });
 });

tests/unit/pdf-parse-normalize-text-items.vitest.spec.ts

@@ -1,4 +1,4 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 
 import { normalizeTextItemsForLayout } from '@openreader/compute-core';
 import type { TextItem } from 'pdfjs-dist/types/src/display/api';
@@ -19,7 +19,7 @@ function makeTextItem(
   } as unknown as TextItem;
 }
 
-test.describe('normalizeTextItemsForLayout', () => {
+describe('normalizeTextItemsForLayout', () => {
   test('keeps horizontal body text and drops rotated/skewed margin text', () => {
     const horizontal = makeTextItem(
       'Powered by large language models',
@@ -36,4 +36,12 @@ test.describe('normalizeTextItemsForLayout', () => {
     expect(normalized).toHaveLength(1);
     expect(normalized[0]?.text).toBe('Powered by large language models');
   });
+
+  test('drops malformed/vertical-only runs so downstream layout planning sees no body text', () => {
+    const vertical = makeTextItem('Side label', [0, 10, -10, 0, 30, 200]);
+    const skewed = makeTextItem('Watermark', [10, 5, 2, 10, 200, 500]);
+
+    const normalized = normalizeTextItemsForLayout([vertical, skewed], 800);
+    expect(normalized).toEqual([]);
+  });
 });

tests/unit/pdf-stitch-cross-page-blocks.vitest.spec.ts

@@ -1,6 +1,7 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 import { stitchCrossPageBlocks } from '@openreader/compute-core';
 import type { ParsedPdfBlock, ParsedPdfDocument, ParsedPdfBlockKind } from '../../src/types/parsed-pdf';
+import { makeParsedPdfDocument, makeParsedPdfPage } from './support/document-fixtures';
 
 function makeBlock(
   id: string,
@@ -23,19 +24,13 @@ function makeBlock(
 }
 
 function makeDoc(page1Blocks: ParsedPdfBlock[], page2Blocks: ParsedPdfBlock[]): ParsedPdfDocument {
-  return {
-    schemaVersion: 1,
-    documentId: 'doc',
-    parserVersion: 'test',
-    parsedAt: 0,
-    pages: [
-      { pageNumber: 1, width: 100, height: 100, blocks: page1Blocks },
-      { pageNumber: 2, width: 100, height: 100, blocks: page2Blocks },
-    ],
-  };
+  return makeParsedPdfDocument([
+    makeParsedPdfPage(1, page1Blocks),
+    makeParsedPdfPage(2, page2Blocks),
+  ]);
 }
 
-test.describe('stitchCrossPageBlocks', () => {
+describe('stitchCrossPageBlocks', () => {
   test('stitches paragraph continuation across footer/header noise', () => {
     const doc = makeDoc(
       [

tests/unit/pdf-tts-planning.vitest.spec.ts

@@ -1,4 +1,4 @@
-import { expect, test } from '@playwright/test';
+import { describe, expect, test } from 'vitest';
 
 import { buildPdfPageSourceUnits, buildPdfPrefetchPayload } from '../../src/lib/client/pdf-tts-planning';
 import type { ParsedPdfPage } from '../../src/types/parsed-pdf';
@@ -39,7 +39,7 @@ function buildPage(pageNumber: number): ParsedPdfPage {
   };
 }
 
-test.describe('pdf tts planning helpers', () => {
+describe('pdf tts planning helpers', () => {
   test('buildPdfPageSourceUnits uses parsed blocks, honors skip kinds, and pins locator page', () => {
     const page = buildPage(2);
     const units = buildPdfPageSourceUnits(page, 2, ['header']);

tests/unit/support/document-fixtures.ts

@@ -0,0 +1,54 @@
+import type { BaseDocument } from '../../../src/types/documents';
+import type { ParsedPdfBlock, ParsedPdfBlockKind, ParsedPdfDocument, ParsedPdfPage } from '../../../src/types/parsed-pdf';
+
+export function makeBaseDocument(overrides: Partial<BaseDocument> = {}): BaseDocument {
+  return {
+    id: 'doc-1',
+    name: 'document.pdf',
+    size: 1_024,
+    lastModified: 1_700_000_000_000,
+    type: 'pdf',
+    ...overrides,
+  };
+}
+
+export function makeParsedPdfBlock(input: {
+  id: string;
+  kind: ParsedPdfBlockKind;
+  text: string;
+  page: number;
+  readingOrder: number;
+}): ParsedPdfBlock {
+  return {
+    id: input.id,
+    kind: input.kind,
+    text: input.text,
+    fragments: [
+      {
+        page: input.page,
+        bbox: [0, 0, 100, 10],
+        text: input.text,
+        readingOrder: input.readingOrder,
+      },
+    ],
+  };
+}
+
+export function makeParsedPdfPage(pageNumber: number, blocks: ParsedPdfBlock[]): ParsedPdfPage {
+  return {
+    pageNumber,
+    width: 800,
+    height: 1_200,
+    blocks,
+  };
+}
+
+export function makeParsedPdfDocument(pages: ParsedPdfPage[]): ParsedPdfDocument {
+  return {
+    schemaVersion: 1,
+    documentId: 'doc-fixture',
+    parserVersion: 'test',
+    parsedAt: 1_700_000_000_000,
+    pages,
+  };
+}

vitest.config.ts

@@ -2,7 +2,17 @@ import { defineConfig } from 'vitest/config';
 import { fileURLToPath } from 'node:url';
 
 const srcDir = fileURLToPath(new URL('./src/', import.meta.url));
+const computeCoreIndex = fileURLToPath(new URL('./compute/core/src/index.ts', import.meta.url));
+const computeCoreApiContracts = fileURLToPath(new URL('./compute/core/src/api-contracts/index.ts', import.meta.url));
+const computeCoreControlPlane = fileURLToPath(new URL('./compute/core/src/control-plane/index.ts', import.meta.url));
+const computeCoreLocalRuntime = fileURLToPath(new URL('./compute/core/src/local-runtime.ts', import.meta.url));
+const computeCoreTypes = fileURLToPath(new URL('./compute/core/src/types/index.ts', import.meta.url));
 const alias = [
+  { find: /^@openreader\/compute-core$/, replacement: computeCoreIndex },
+  { find: /^@openreader\/compute-core\/api-contracts$/, replacement: computeCoreApiContracts },
+  { find: /^@openreader\/compute-core\/control-plane$/, replacement: computeCoreControlPlane },
+  { find: /^@openreader\/compute-core\/local-runtime$/, replacement: computeCoreLocalRuntime },
+  { find: /^@openreader\/compute-core\/types$/, replacement: computeCoreTypes },
   { find: /^@\//, replacement: `${srcDir}` },
   { find: '@', replacement: srcDir },
 ];