@bendyline/squisq-formats 2.0.1 → 2.2.0

package/src/__tests__/htmlImport.test.ts

@@ -1,57 +0,0 @@
-import { describe, expect, it } from 'vitest';
-import { htmlToMarkdown, htmlToMarkdownDocSync } from '../html/import.js';
-
-describe('htmlToMarkdown', () => {
-  it('converts headings and paragraphs', () => {
-    const md = htmlToMarkdown('<h1>Title</h1><p>Hello <strong>world</strong>.</p>');
-    expect(md).toContain('# Title');
-    expect(md).toContain('Hello **world**.');
-  });
-
-  it('converts links and inline code', () => {
-    const md = htmlToMarkdown(
-      '<p>See <a href="https://x.test">site</a> and <code>npm i</code></p>',
-    );
-    expect(md).toContain('[site](https://x.test)');
-    expect(md).toContain('`npm i`');
-  });
-
-  it('converts unordered and ordered lists', () => {
-    const md = htmlToMarkdown('<ul><li>a</li><li>b</li></ul><ol><li>one</li><li>two</li></ol>');
-    expect(md).toMatch(/[-*]\s+a/);
-    expect(md).toMatch(/[-*]\s+b/);
-    expect(md).toMatch(/1\.\s+one/);
-  });
-
-  it('converts blockquotes and tables', () => {
-    const md = htmlToMarkdown(
-      '<blockquote><p>quoted</p></blockquote>' +
-        '<table><thead><tr><th>H</th></tr></thead><tbody><tr><td>cell</td></tr></tbody></table>',
-    );
-    expect(md).toContain('> quoted');
-    expect(md).toContain('cell');
-  });
-
-  it('drops scripts and styles (sanitized by default)', () => {
-    const md = htmlToMarkdown('<p>safe</p><script>alert(1)</script><style>.x{color:red}</style>');
-    expect(md).toContain('safe');
-    expect(md).not.toContain('alert(1)');
-    expect(md).not.toContain('color:red');
-  });
-
-  it('strips javascript: links via sanitizer', () => {
-    const md = htmlToMarkdown('<a href="javascript:alert(1)">click</a>');
-    expect(md).not.toContain('javascript:alert(1)');
-  });
-
-  it('builds a document node tree', () => {
-    const doc = htmlToMarkdownDocSync('<h2>Hi</h2>');
-    expect(doc.type).toBe('document');
-    expect(doc.children[0]?.type).toBe('heading');
-  });
-
-  it('handles unwrapped div/span containers', () => {
-    const md = htmlToMarkdown('<div><span>plain </span><em>text</em></div>');
-    expect(md).toContain('plain *text*');
-  });
-});

package/src/__tests__/inferTheme.test.ts

@@ -1,135 +0,0 @@
-/**
- * Tests for `inferThemeFromFile`: format sniffing, per-format extraction
- * (incl. dark-deck clrMap inversion), theme compilation, and error paths.
- */
-
-import { describe, expect, it } from 'vitest';
-import { ConversionError } from '../registry/errors';
-import { inferThemeFromFile } from '../infer/index';
-import {
-  buildThemeXml,
-  buildThemedDocx,
-  buildThemedPptx,
-  buildThemedXlsx,
-} from './pptxInferFixtures';
-
-describe('inferThemeFromFile — PPTX', () => {
-  it('maps a light deck onto seeds, palette, schemes, and fonts', async () => {
-    const data = await buildThemedPptx();
-    const result = await inferThemeFromFile(data);
-
-    expect(result.extraction.sourceFormat).toBe('pptx');
-    expect(result.extraction.colorMap).toEqual({ bg1: 'lt1', tx1: 'dk1' });
-
-    const { theme } = result;
-    expect(theme.seedColors).toMatchObject({
-      primary: '#4472c4',
-      secondary: '#ed7d31',
-      accent: '#a5a5a5',
-      background: '#fdfdf8',
-      text: '#1a1a2e',
-    });
-    expect(theme.colors.background).toBe('#fdfdf8');
-    expect(theme.colors.text).toBe('#1a1a2e');
-    // Companion slot (lt2) becomes backgroundLight; hlink becomes highlight.
-    expect(theme.colors.backgroundLight).toBe('#efefe4');
-    expect(theme.colors.highlight).toBe('#0563c1');
-    // Six accents → six color schemes keyed by slot.
-    expect(Object.keys(theme.colorSchemes)).toEqual([
-      'accent1',
-      'accent2',
-      'accent3',
-      'accent4',
-      'accent5',
-      'accent6',
-    ]);
-    // Fonts: curated stack match + custom preservation.
-    expect(theme.typography.titleFont).toEqual({ stackId: 'playfair' });
-    expect(theme.typography.bodyFont).toEqual({
-      custom: { name: 'Aptos', fallback: 'sans-serif' },
-    });
-    // Compiled + validated, named from the theme part, re-editable via seeds.
-    expect(theme.name).toBe('Fixture Theme');
-    expect(theme.id).toBe('custom-fixture-theme');
-  });
-
-  it('inverts background/text for a dark deck (clrMap bg1="dk1")', async () => {
-    const data = await buildThemedPptx({
-      clrMapAttrs: 'bg1="dk1" tx1="lt1" bg2="dk2" tx2="lt2"',
-    });
-    const result = await inferThemeFromFile(data);
-    expect(result.theme.seedColors?.background).toBe('#1a1a2e');
-    expect(result.theme.seedColors?.text).toBe('#fdfdf8');
-    // Dark companion slot backs backgroundLight.
-    expect(result.theme.colors.backgroundLight).toBe('#30304a');
-  });
-
-  it('respects a nameHint over the file theme name', async () => {
-    const result = await inferThemeFromFile(await buildThemedPptx(), { nameHint: 'Q3 Deck' });
-    expect(result.theme.name).toBe('Q3 Deck');
-    expect(result.theme.id).toBe('custom-q3-deck');
-  });
-
-  it('throws invalid-input when the file has no theme part', async () => {
-    const data = await buildThemedPptx({ themeXml: null });
-    await expect(inferThemeFromFile(data)).rejects.toMatchObject({
-      name: 'ConversionError',
-      code: 'invalid-input',
-    });
-  });
-});
-
-describe('inferThemeFromFile — DOCX / XLSX', () => {
-  it('extracts from a DOCX theme part', async () => {
-    const result = await inferThemeFromFile(await buildThemedDocx());
-    expect(result.extraction.sourceFormat).toBe('docx');
-    expect(result.theme.seedColors?.primary).toBe('#4472c4');
-    expect(result.theme.colors.background).toBe('#fdfdf8');
-  });
-
-  it('extracts from an XLSX theme part', async () => {
-    const result = await inferThemeFromFile(await buildThemedXlsx());
-    expect(result.extraction.sourceFormat).toBe('xlsx');
-    expect(result.theme.seedColors?.primary).toBe('#4472c4');
-  });
-
-  it('ignores inferLayouts for non-PPTX sources with a warning', async () => {
-    const result = await inferThemeFromFile(await buildThemedDocx(), { inferLayouts: true });
-    expect(result.layouts).toBeUndefined();
-    expect(result.warnings.some((w) => w.includes('PowerPoint'))).toBe(true);
-  });
-});
-
-describe('inferThemeFromFile — rejection paths', () => {
-  it('rejects PDF bytes with unsupported-input', async () => {
-    const pdf = new TextEncoder().encode('%PDF-1.4 fake').buffer as ArrayBuffer;
-    await expect(inferThemeFromFile(pdf)).rejects.toMatchObject({
-      code: 'unsupported-input',
-    });
-  });
-
-  it('rejects non-zip bytes with invalid-input', async () => {
-    const junk = new TextEncoder().encode('definitely not a zip').buffer as ArrayBuffer;
-    const err = await inferThemeFromFile(junk).catch((e: unknown) => e);
-    expect(err).toBeInstanceOf(ConversionError);
-    expect((err as ConversionError).code).toBe('invalid-input');
-  });
-});
-
-describe('inferThemeFromFile — degraded themes', () => {
-  it('drops near-background accents from the scheme list with a warning', async () => {
-    // accent2 equals the background color — should be dropped from schemes.
-    const themeXml = buildThemeXml({ accents: ['4472c4', 'fdfdf8', 'a5a5a5'] });
-    const result = await inferThemeFromFile(await buildThemedPptx({ themeXml }));
-    expect(Object.keys(result.theme.colorSchemes)).toEqual(['accent1', 'accent3']);
-    expect(result.warnings.some((w) => w.includes('accent2'))).toBe(true);
-  });
-
-  it('keeps fonts inherited when the theme has no fontScheme', async () => {
-    const themeXml = buildThemeXml({ omitFontScheme: true });
-    const result = await inferThemeFromFile(await buildThemedPptx({ themeXml }));
-    // STARTER defaults survive.
-    expect(result.theme.typography.titleFont).toEqual({ stackId: 'system-serif' });
-    expect(result.warnings.some((w) => w.includes('fontScheme'))).toBe(true);
-  });
-});

package/src/__tests__/lossyWarnings.test.ts

@@ -1,146 +0,0 @@
-/**
- * Tests for the registry's consolidated lossy-path warnings, all surfaced
- * through `ConversionResult.warnings`:
- *  - xlsx export drops non-table content (tables-only fidelity)
- *  - pdf import under Node skips embedded images (no DOM canvas)
- *  - transform → markdown produces blocks that don't round-trip
- */
-
-import { describe, it, expect, afterEach, vi } from 'vitest';
-import { MemoryContentContainer } from '@bendyline/squisq/storage';
-import type { MarkdownDocument } from '@bendyline/squisq/markdown';
-import { convert, defaultRegistry } from '../registry/index';
-
-// ── shared Markdown exporter capability profiles ───────────────────
-
-describe('Markdown exporters warn about unsupported AST extensions', () => {
-  const extendedMarkdown: MarkdownDocument = {
-    type: 'document',
-    children: [
-      {
-        type: 'paragraph',
-        children: [
-          { type: 'text', value: 'Assigned to ' },
-          {
-            type: 'mention',
-            targetKind: 'user',
-            targetId: '42',
-            displayName: 'Ada',
-          },
-        ],
-      },
-      {
-        type: 'containerDirective',
-        name: 'note',
-        children: [{ type: 'paragraph', children: [{ type: 'text', value: 'Important' }] }],
-      },
-    ],
-  };
-
-  it('surfaces the omitted node types through ConversionResult', async () => {
-    const result = await convert({ kind: 'markdown', markdown: extendedMarkdown }, 'docx');
-    const warning = result.warnings.find((item) => /unsupported Markdown node/i.test(item));
-
-    expect(warning).toMatch(/mention \(1\)/);
-    expect(warning).toMatch(/containerDirective \(1\)/);
-  });
-
-  it('does not warn for the supported Markdown subset', async () => {
-    const result = await convert({ kind: 'markdown', markdown: '# Title\n\nBody.' }, 'docx');
-    expect(result.warnings.some((item) => /unsupported Markdown node/i.test(item))).toBe(false);
-  });
-});
-
-// ── xlsx: tables-only fidelity ──────────────────────────────────────
-
-describe('xlsx export warns about omitted non-table content', () => {
-  it('warns when prose / lists accompany the table(s)', async () => {
-    const md =
-      '# Report\n\nIntro prose paragraph.\n\n- a\n- b\n\n| A | B |\n| - | - |\n| 1 | 2 |\n';
-    const result = await convert({ kind: 'markdown', markdown: md }, 'xlsx');
-    expect(result.warnings.some((w) => /tables-only/i.test(w))).toBe(true);
-    // Two omitted blocks: the paragraph and the list (heading names the sheet).
-    expect(result.warnings.find((w) => /tables-only/i.test(w))).toMatch(/2 non-table block/);
-  });
-
-  it('does not warn for a heading + table only document', async () => {
-    const md = '# Sales\n\n| A |\n| - |\n| x |\n';
-    const result = await convert({ kind: 'markdown', markdown: md }, 'xlsx');
-    expect(result.warnings.some((w) => /tables-only/i.test(w))).toBe(false);
-  });
-});
-
-// ── pdf: Node has no DOM canvas → images skipped ────────────────────
-
-describe('pdf import warns when running without a DOM (Node)', () => {
-  afterEach(() => {
-    vi.unstubAllGlobals();
-  });
-
-  /** A pdf definition whose importContainer is stubbed (avoids pdfjs/canvas). */
-  function registryWithStubPdf() {
-    const registry = defaultRegistry();
-    registry.register({
-      ...registry.get('pdf')!,
-      async importContainer() {
-        const c = new MemoryContentContainer();
-        await c.writeDocument('# From PDF\n\nBody.');
-        return c;
-      },
-    });
-    return registry;
-  }
-
-  const pdfBytes = new TextEncoder().encode('%PDF-1.4\n%mock\n');
-
-  it('surfaces the skipped-images warning when document is undefined', async () => {
-    vi.stubGlobal('document', undefined);
-    const registry = registryWithStubPdf();
-    const result = await convert({ kind: 'bytes', data: pdfBytes, filename: 'x.pdf' }, 'md', {
-      from: 'pdf',
-      registry,
-    });
-    expect(result.warnings.some((w) => /embedded images were skipped/i.test(w))).toBe(true);
-  });
-
-  it('does not warn when a DOM is present (jsdom default)', async () => {
-    const registry = registryWithStubPdf();
-    const result = await convert({ kind: 'bytes', data: pdfBytes, filename: 'x.pdf' }, 'md', {
-      from: 'pdf',
-      registry,
-    });
-    expect(result.warnings.some((w) => /embedded images were skipped/i.test(w))).toBe(false);
-  });
-});
-
-// ── transform → markdown round-trip loss ────────────────────────────
-
-describe('transform-to-markdown export warns about non-round-tripping blocks', () => {
-  const CONTENT_MD = `# Big Title
-
-An intro paragraph with plenty of words so the transform has real content.
-
-## Numbers
-
-Revenue grew by 45% last quarter across every region and product line worldwide.
-
-- Point one with several words to analyze here
-- Point two with several words to analyze here
-
-| Metric | Value |
-| - | - |
-| Users | 1000 |
-`;
-
-  it('warns when a transform reshapes into template blocks that markdown can’t hold', async () => {
-    const result = await convert({ kind: 'markdown', markdown: CONTENT_MD }, 'md', {
-      transformStyle: 'magazine',
-    });
-    expect(result.warnings.some((w) => /round-trip to markdown/i.test(w))).toBe(true);
-  });
-
-  it('does not warn without a transform', async () => {
-    const result = await convert({ kind: 'markdown', markdown: CONTENT_MD }, 'md');
-    expect(result.warnings.some((w) => /round-trip to markdown/i.test(w))).toBe(false);
-  });
-});

package/src/__tests__/ooxml.test.ts

@@ -1,271 +0,0 @@
-/**
- * Tests for the shared OOXML layer: xmlUtils, writer, reader round-trips.
- */
-
-import { afterEach, describe, it, expect, vi } from 'vitest';
-import {
-  escapeXml,
-  xmlElement,
-  selfClosingElement,
-  textElement,
-  attrString,
-  xmlDeclaration,
-} from '../ooxml/xmlUtils';
-import { createPackage } from '../ooxml/writer';
-import { openPackage, getPartXml, getCoreProperties, getPartRelationships } from '../ooxml/reader';
-import { REL_OFFICE_DOCUMENT } from '../ooxml/namespaces';
-
-afterEach(() => {
-  vi.unstubAllGlobals();
-});
-
-// ============================================
-// XML Utilities
-// ============================================
-
-describe('escapeXml', () => {
-  it('escapes all five XML entities', () => {
-    expect(escapeXml('a & b < c > d " e \' f')).toBe('a &amp; b &lt; c &gt; d &quot; e &apos; f');
-  });
-
-  it('returns empty string unchanged', () => {
-    expect(escapeXml('')).toBe('');
-  });
-
-  it('returns plain text unchanged', () => {
-    expect(escapeXml('hello world')).toBe('hello world');
-  });
-});
-
-describe('attrString', () => {
-  it('builds attribute string from object', () => {
-    const result = attrString({ 'w:val': 'Heading1' });
-    expect(result).toBe(' w:val="Heading1"');
-  });
-
-  it('omits undefined values', () => {
-    const result = attrString({ 'w:val': 'test', 'w:other': undefined });
-    expect(result).toBe(' w:val="test"');
-  });
-
-  it('returns empty string for empty attrs', () => {
-    expect(attrString({})).toBe('');
-    expect(attrString(undefined)).toBe('');
-  });
-});
-
-describe('selfClosingElement', () => {
-  it('builds self-closing tag', () => {
-    expect(selfClosingElement('w:b')).toBe('<w:b/>');
-  });
-
-  it('includes attributes', () => {
-    expect(selfClosingElement('w:pStyle', { 'w:val': 'Heading1' })).toBe(
-      '<w:pStyle w:val="Heading1"/>',
-    );
-  });
-});
-
-describe('xmlElement', () => {
-  it('builds element with children', () => {
-    expect(xmlElement('w:p', {}, 'content')).toBe('<w:p>content</w:p>');
-  });
-
-  it('self-closes when no children', () => {
-    expect(xmlElement('w:p', {})).toBe('<w:p/>');
-  });
-
-  it('concatenates multiple children', () => {
-    expect(xmlElement('w:p', {}, 'a', 'b')).toBe('<w:p>ab</w:p>');
-  });
-});
-
-describe('textElement', () => {
-  it('escapes text content', () => {
-    expect(textElement('w:t', {}, 'a & b')).toBe('<w:t>a &amp; b</w:t>');
-  });
-
-  it('self-closes when no text', () => {
-    expect(textElement('w:t', {})).toBe('<w:t/>');
-  });
-});
-
-describe('xmlDeclaration', () => {
-  it('returns standard XML declaration', () => {
-    expect(xmlDeclaration()).toBe('<?xml version="1.0" encoding="UTF-8" standalone="yes"?>');
-  });
-});
-
-// ============================================
-// Package Writer + Reader Round-Trip
-// ============================================
-
-describe('createPackage / openPackage round-trip', () => {
-  it('rejects unsafe authored part paths', () => {
-    const pkg = createPackage();
-    expect(() => pkg.addPart('../outside.xml', '<doc/>', 'application/xml')).toThrow(
-      /path is unsafe/i,
-    );
-    expect(() =>
-      pkg.addRelationship('..\\outside.xml', {
-        id: 'rId1',
-        type: REL_OFFICE_DOCUMENT,
-        target: 'word/document.xml',
-      }),
-    ).toThrow(/path is unsafe/i);
-  });
-
-  it('parses packages when the browser DOMParser global is unavailable', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc><value>node</value></doc>', 'application/xml');
-    const buffer = await pkg.toArrayBuffer();
-
-    vi.stubGlobal('DOMParser', undefined);
-    const opened = await openPackage(buffer);
-    const doc = await getPartXml(opened, 'word/document.xml');
-    expect(doc?.getElementsByTagName('value')[0]?.textContent).toBe('node');
-  });
-
-  it('ignores an incomplete ambient DOMParser when running in Node', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc><value>node</value></doc>', 'application/xml');
-    const buffer = await pkg.toArrayBuffer();
-
-    class IncompleteDomParser {
-      parseFromString(): never {
-        throw new Error('ambient parser must not be used in Node');
-      }
-    }
-
-    vi.stubGlobal('DOMParser', IncompleteDomParser);
-    const opened = await openPackage(buffer);
-    const doc = await getPartXml(opened, 'word/document.xml');
-    expect(doc?.getElementsByTagName('value')[0]?.textContent).toBe('node');
-  });
-
-  it('applies archive resource limits before parsing OOXML parts', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc/>', 'application/xml');
-    const buffer = await pkg.toArrayBuffer();
-
-    await expect(openPackage(buffer, { maxEntries: 1 })).rejects.toMatchObject({
-      code: 'too-many-entries',
-      actual: 3,
-    });
-    await expect(openPackage(buffer, { maxUncompressedBytes: 1 })).rejects.toThrow(
-      /uncompressed content exceeds 1 byte limit/,
-    );
-  });
-
-  it('creates a valid ZIP with content types and parts', async () => {
-    const pkg = createPackage();
-    pkg.addPart(
-      'word/document.xml',
-      '<w:document><w:body/></w:document>',
-      'application/vnd.openxmlformats-officedocument.wordprocessingml.document.main+xml',
-    );
-    pkg.addRelationship('', {
-      id: 'rId1',
-      type: REL_OFFICE_DOCUMENT,
-      target: 'word/document.xml',
-    });
-
-    const arrayBuffer = await pkg.toArrayBuffer();
-    expect(arrayBuffer).toBeInstanceOf(ArrayBuffer);
-    expect(arrayBuffer.byteLength).toBeGreaterThan(0);
-
-    // Read it back
-    const opened = await openPackage(arrayBuffer);
-    expect(opened).not.toHaveProperty('zip');
-    expect(opened.contentTypes.overrides.has('word/document.xml')).toBe(true);
-    expect(opened.rootRelationships.length).toBeGreaterThanOrEqual(1);
-    expect(opened.rootRelationships[0].type).toBe(REL_OFFICE_DOCUMENT);
-  });
-
-  it('round-trips XML part content', async () => {
-    const pkg = createPackage();
-    const testXml =
-      '<?xml version="1.0" encoding="UTF-8" standalone="yes"?><root><child attr="value">text</child></root>';
-    pkg.addPart('test/data.xml', testXml, 'application/xml');
-
-    const buffer = await pkg.toArrayBuffer();
-    const opened = await openPackage(buffer);
-    const doc = await getPartXml(opened, 'test/data.xml');
-
-    expect(doc).not.toBeNull();
-    const child = doc!.getElementsByTagName('child')[0];
-    expect(child.getAttribute('attr')).toBe('value');
-    expect(child.textContent).toBe('text');
-  });
-
-  it('round-trips core properties', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc/>', 'application/xml');
-    pkg.setCoreProperties({
-      title: 'Test Title',
-      creator: 'Test Author',
-      description: 'Test description with <special> & chars',
-    });
-
-    const buffer = await pkg.toArrayBuffer();
-    const opened = await openPackage(buffer);
-    const props = await getCoreProperties(opened);
-
-    expect(props.title).toBe('Test Title');
-    expect(props.creator).toBe('Test Author');
-    expect(props.description).toBe('Test description with <special> & chars');
-  });
-
-  it('serializes repeatedly without accumulating generated relationships', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc/>', 'application/xml');
-    pkg.setCoreProperties({ title: 'Stable' });
-
-    await pkg.toArrayBuffer();
-    const second = await pkg.toArrayBuffer();
-    const opened = await openPackage(second);
-    const coreRelationships = opened.rootRelationships.filter((rel) =>
-      rel.type.endsWith('/metadata/core-properties'),
-    );
-    expect(coreRelationships).toHaveLength(1);
-  });
-
-  it('round-trips part relationships', async () => {
-    const pkg = createPackage();
-    pkg.addPart('word/document.xml', '<doc/>', 'application/xml');
-    pkg.addRelationship('word/document.xml', {
-      id: 'rId1',
-      type: 'http://example.com/styles',
-      target: 'styles.xml',
-    });
-    pkg.addRelationship('word/document.xml', {
-      id: 'rId2',
-      type: 'http://example.com/hyperlink',
-      target: 'https://example.com',
-      targetMode: 'External',
-    });
-
-    const buffer = await pkg.toArrayBuffer();
-    const opened = await openPackage(buffer);
-    const rels = await getPartRelationships(opened, 'word/document.xml');
-
-    expect(rels.length).toBe(2);
-    expect(rels[0].id).toBe('rId1');
-    expect(rels[0].target).toBe('styles.xml');
-    expect(rels[1].id).toBe('rId2');
-    expect(rels[1].target).toBe('https://example.com');
-    expect(rels[1].targetMode).toBe('External');
-  });
-
-  it('handles binary parts', async () => {
-    const pkg = createPackage();
-    const testData = new Uint8Array([0x89, 0x50, 0x4e, 0x47]); // PNG magic bytes
-    pkg.addBinaryPart('word/media/image1.png', testData, 'image/png');
-
-    const buffer = await pkg.toArrayBuffer();
-    const opened = await openPackage(buffer);
-
-    // Binary part should appear in content types via extension default
-    expect(opened.contentTypes.defaults.has('png')).toBe(true);
-  });
-});