@link-assistant/web-capture 2.0.0 → 2.0.1

package/tests/unit/image-mode-defaults.test.js

@@ -1,54 +0,0 @@
-import { applyImageMode } from '../../src/extract-images.js';
-import fs from 'fs';
-import path from 'path';
-import os from 'os';
-
-const TINY_PNG =
-  'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8/5+hHgAHggJ/PchI7wAAAABJRU5ErkJggg==';
-const REMOTE = `Hi.\n\n![](https://example.invalid/foo.png)\n\nBye.\n`;
-const BASE64 = `Hi.\n\n![](data:image/png;base64,${TINY_PNG})\n\nBye.\n`;
-
-describe('image-mode contract', () => {
-  let tmp;
-  beforeEach(() => {
-    tmp = fs.mkdtempSync(path.join(os.tmpdir(), 'imode-'));
-  });
-  afterEach(() => fs.rmSync(tmp, { recursive: true, force: true }));
-
-  it('default mode keeps remote URLs', async () => {
-    const r = await applyImageMode(REMOTE, { mode: 'default' });
-    expect(r.markdown).toContain('https://example.invalid/foo.png');
-    expect(r.markdown).not.toMatch(/images\//);
-    expect(r.markdown).not.toMatch(/data:image/);
-  });
-
-  it('--extract-images writes files', async () => {
-    const r = await applyImageMode(BASE64, {
-      mode: 'extract',
-      dir: tmp,
-      subdir: 'images',
-    });
-    expect(r.markdown).toMatch(/images\/image-/);
-    expect(fs.readdirSync(path.join(tmp, 'images'))).toHaveLength(1);
-  });
-
-  it('--embed-images keeps base64', async () => {
-    const r = await applyImageMode(BASE64, { mode: 'embed' });
-    expect(r.markdown).toContain('data:image');
-  });
-
-  it('flags have observable effect on every input', async () => {
-    for (const src of [REMOTE, BASE64]) {
-      const a = await applyImageMode(src, { mode: 'default' });
-      const b = await applyImageMode(src, { mode: 'embed' });
-      const c = await applyImageMode(src, {
-        mode: 'extract',
-        dir: tmp,
-        subdir: 'images',
-      });
-      expect(
-        new Set([a.markdown, b.markdown, c.markdown]).size
-      ).toBeGreaterThan(1);
-    }
-  });
-});

package/tests/unit/kreuzberg.test.js

@@ -1,249 +0,0 @@
-import {
-  convertWithKreuzberg,
-  isKreuzbergAvailable,
-  normalizeStructuredKeys,
-} from '../../src/kreuzberg.js';
-
-let available;
-
-beforeAll(async () => {
-  available = await isKreuzbergAvailable();
-});
-
-function skipIfUnavailable() {
-  if (!available) {
-    return 'kreuzberg native binding not installed';
-  }
-  return false;
-}
-
-describe('kreuzberg html-to-markdown integration', () => {
-  it('reports availability correctly', async () => {
-    expect(typeof available).toBe('boolean');
-  });
-
-  it('converts basic HTML to Markdown', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const result = await convertWithKreuzberg(
-      '<h1>Hello World</h1><p>This is a test.</p>'
-    );
-    expect(result.content).toContain('# Hello World');
-    expect(result.content).toContain('This is a test.');
-  });
-
-  it('returns structured result with all fields', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const result = await convertWithKreuzberg('<p>Test</p>');
-    expect(result).toHaveProperty('content');
-    expect(result).toHaveProperty('metadata');
-    expect(result).toHaveProperty('tables');
-    expect(result).toHaveProperty('images');
-    expect(result).toHaveProperty('warnings');
-  });
-
-  it('extracts metadata from HTML', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = `<html>
-      <head>
-        <title>Test Page</title>
-        <meta property="og:description" content="A test page">
-      </head>
-      <body><h1>Hello</h1></body>
-    </html>`;
-    const result = await convertWithKreuzberg(html);
-    expect(result.metadata).toBeTruthy();
-    expect(result.metadata.document.title).toBe('Test Page');
-    expect(result.metadata.document.open_graph.description).toBe('A test page');
-  });
-
-  it('extracts headings from metadata', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<h1>First</h1><h2>Second</h2><h3>Third</h3>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.metadata).toBeTruthy();
-    expect(result.metadata.headers.length).toBe(3);
-    expect(result.metadata.headers[0].text).toBe('First');
-    expect(result.metadata.headers[0].level).toBe(1);
-  });
-
-  it('extracts links from metadata', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html =
-      '<a href="https://example.com">Example</a><a href="mailto:test@test.com">Email</a>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.metadata).toBeTruthy();
-    expect(result.metadata.links.length).toBeGreaterThanOrEqual(1);
-    const externalLink = result.metadata.links.find(
-      (l) => l.href === 'https://example.com'
-    );
-    expect(externalLink).toBeTruthy();
-    expect(externalLink.text).toBe('Example');
-  });
-
-  it('converts tables to Markdown', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = `<table>
-      <thead><tr><th>Name</th><th>Value</th></tr></thead>
-      <tbody><tr><td>A</td><td>1</td></tr><tr><td>B</td><td>2</td></tr></tbody>
-    </table>`;
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('| Name | Value |');
-    expect(result.content).toMatch(/\| A\s+\| 1\s+\|/);
-    expect(result.content).toMatch(/\| B\s+\| 2\s+\|/);
-  });
-
-  it('converts links correctly', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<a href="https://example.com">Click here</a>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('[Click here](https://example.com)');
-  });
-
-  it('resolves relative links with baseUrl', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<a href="/about">About</a>';
-    const result = await convertWithKreuzberg(html, {
-      baseUrl: 'https://example.com/docs/page',
-    });
-    expect(result.content).toContain('[About](https://example.com/about)');
-  });
-
-  it('converts bold and italic correctly', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<p>This is <strong>bold</strong> and <em>italic</em></p>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('**bold**');
-    expect(result.content).toContain('*italic*');
-  });
-
-  it('converts code blocks correctly', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<pre><code class="language-js">const x = 1;</code></pre>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('```');
-    expect(result.content).toContain('const x = 1;');
-  });
-
-  it('handles empty HTML gracefully', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const result = await convertWithKreuzberg('');
-    expect(result.content).toBeDefined();
-  });
-
-  it('removes script tags from output', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<p>Content</p><script>alert("xss")</script>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('Content');
-    expect(result.content).not.toContain('alert');
-    expect(result.content).not.toContain('script');
-  });
-
-  it('removes style tags from output', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<style>body { color: red; }</style><p>Content</p>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('Content');
-    expect(result.content).not.toContain('color: red');
-  });
-
-  it('handles GFM strikethrough', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<p>This is <del>deleted</del> text</p>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('~~deleted~~');
-  });
-
-  it('handles lists correctly', async () => {
-    const skip = skipIfUnavailable();
-    if (skip) {
-      return;
-    }
-    const html = '<ul><li>Item 1</li><li>Item 2</li><li>Item 3</li></ul>';
-    const result = await convertWithKreuzberg(html);
-    expect(result.content).toContain('Item 1');
-    expect(result.content).toContain('Item 2');
-    expect(result.content).toContain('Item 3');
-  });
-});
-
-describe('kreuzberg structured key normalization (issue #137 parity)', () => {
-  // Mirrors the Rust `inline_image_to_json` regression tests: after the
-  // html-to-markdown 3.6 change, inline image dimensions are a structured
-  // `{ width, height }` value rather than a tuple. The normalization pipeline
-  // that backs `result.images` must preserve those `width`/`height` keys.
-  it('preserves width/height keys on inline image dimensions', () => {
-    const images = normalizeStructuredKeys([
-      {
-        format: 'png',
-        filename: 'pixel.png',
-        description: 'one pixel',
-        dimensions: { width: 800, height: 600 },
-        source: 'ImgDataUri',
-      },
-    ]);
-
-    expect(images).toHaveLength(1);
-    expect(images[0].dimensions).toEqual({ width: 800, height: 600 });
-    expect(images[0].format).toBe('png');
-    expect(images[0].filename).toBe('pixel.png');
-  });
-
-  it('leaves images without dimensions untouched', () => {
-    const images = normalizeStructuredKeys([
-      { format: 'svg', source: 'SvgElement', dimensions: null },
-    ]);
-
-    expect(images[0].dimensions).toBeNull();
-  });
-
-  it('snake_cases nested structured keys', () => {
-    const normalized = normalizeStructuredKeys({
-      openGraph: { siteName: 'Example' },
-    });
-
-    expect(normalized).toEqual({ open_graph: { site_name: 'Example' } });
-  });
-});

package/tests/unit/latex.test.js

@@ -1,123 +0,0 @@
-import * as cheerio from 'cheerio';
-import {
-  extractHabrFormula,
-  extractKatexFormula,
-  extractMathJaxFormula,
-  isFormulaImage,
-  isMathElement,
-  extractFormula,
-} from '../../src/latex.js';
-
-describe('latex module', () => {
-  describe('extractHabrFormula', () => {
-    it('extracts LaTeX from Habr formula img source attribute', () => {
-      const $ = cheerio.load(
-        '<img class="formula" source=" L \\to L^2 " alt="formula">'
-      );
-      const el = $('img')[0];
-      expect(extractHabrFormula($, el)).toBe('L \\to L^2');
-    });
-
-    it('falls back to alt text when source is missing', () => {
-      const $ = cheerio.load('<img class="formula" alt=" x + y ">');
-      const el = $('img')[0];
-      expect(extractHabrFormula($, el)).toBe('x + y');
-    });
-
-    it('returns null when no source or alt', () => {
-      const $ = cheerio.load('<img class="formula">');
-      const el = $('img')[0];
-      expect(extractHabrFormula($, el)).toBeNull();
-    });
-  });
-
-  describe('extractKatexFormula', () => {
-    it('extracts from annotation element', () => {
-      const $ = cheerio.load(
-        '<span class="katex"><annotation encoding="application/x-tex">E = mc^2</annotation></span>'
-      );
-      const el = $('span.katex')[0];
-      expect(extractKatexFormula($, el)).toBe('E = mc^2');
-    });
-
-    it('extracts from data-tex attribute', () => {
-      const $ = cheerio.load(
-        '<span class="math" data-tex="\\alpha + \\beta"></span>'
-      );
-      const el = $('span.math')[0];
-      expect(extractKatexFormula($, el)).toBe('\\alpha + \\beta');
-    });
-
-    it('returns null when no TeX source found', () => {
-      const $ = cheerio.load('<span class="katex">rendered</span>');
-      const el = $('span.katex')[0];
-      expect(extractKatexFormula($, el)).toBeNull();
-    });
-  });
-
-  describe('extractMathJaxFormula', () => {
-    it('extracts from data-tex attribute', () => {
-      const $ = cheerio.load(
-        '<mjx-container data-tex="\\sum_{i=1}^{n}"></mjx-container>'
-      );
-      const el = $('mjx-container')[0];
-      expect(extractMathJaxFormula($, el)).toBe('\\sum_{i=1}^{n}');
-    });
-  });
-
-  describe('isFormulaImage', () => {
-    it('identifies img.formula elements', () => {
-      const $ = cheerio.load('<img class="formula" source="x^2">');
-      expect(isFormulaImage($, $('img')[0])).toBe(true);
-    });
-
-    it('identifies img with source attribute', () => {
-      const $ = cheerio.load('<img source="x^2">');
-      expect(isFormulaImage($, $('img')[0])).toBe(true);
-    });
-
-    it('rejects regular images', () => {
-      const $ = cheerio.load('<img src="photo.jpg" alt="photo">');
-      expect(isFormulaImage($, $('img')[0])).toBe(false);
-    });
-  });
-
-  describe('isMathElement', () => {
-    it('identifies katex elements', () => {
-      const $ = cheerio.load('<span class="katex">x</span>');
-      expect(isMathElement($, $('span')[0])).toBe(true);
-    });
-
-    it('identifies math elements', () => {
-      const $ = cheerio.load('<span class="math">x</span>');
-      expect(isMathElement($, $('span')[0])).toBe(true);
-    });
-
-    it('identifies mjx-container', () => {
-      const $ = cheerio.load('<mjx-container>x</mjx-container>');
-      expect(isMathElement($, $('mjx-container')[0])).toBe(true);
-    });
-
-    it('rejects regular elements', () => {
-      const $ = cheerio.load('<span class="text">x</span>');
-      expect(isMathElement($, $('span')[0])).toBe(false);
-    });
-  });
-
-  describe('extractFormula', () => {
-    it('extracts from Habr formula image', () => {
-      const $ = cheerio.load('<img class="formula" source="L \\to L^2">');
-      expect(extractFormula($, $('img')[0])).toBe('L \\to L^2');
-    });
-
-    it('extracts from KaTeX element', () => {
-      const $ = cheerio.load('<span class="katex" data-tex="\\alpha"></span>');
-      expect(extractFormula($, $('span')[0])).toBe('\\alpha');
-    });
-
-    it('returns null for regular elements', () => {
-      const $ = cheerio.load('<p>text</p>');
-      expect(extractFormula($, $('p')[0])).toBeNull();
-    });
-  });
-});

package/tests/unit/localize-images.test.js

@@ -1,114 +0,0 @@
-import {
-  extractImageReferences,
-  getExtensionFromUrl,
-  generateLocalFilename,
-  generateFigureFilename,
-  localizeImages,
-} from '../../src/localize-images.js';
-
-describe('localize-images module', () => {
-  describe('extractImageReferences', () => {
-    it('extracts markdown image references', () => {
-      const md =
-        '![alt text](https://example.com/img.png)\n![photo](https://cdn.example.com/photo.jpg)';
-      const images = extractImageReferences(md);
-      expect(images).toHaveLength(2);
-      expect(images[0].altText).toBe('alt text');
-      expect(images[0].url).toBe('https://example.com/img.png');
-      expect(images[1].altText).toBe('photo');
-      expect(images[1].url).toBe('https://cdn.example.com/photo.jpg');
-    });
-
-    it('ignores local image references', () => {
-      const md = '![local](images/figure-1.png)';
-      const images = extractImageReferences(md);
-      expect(images).toHaveLength(0);
-    });
-
-    it('ignores non-image links', () => {
-      const md = '[link text](https://example.com)';
-      const images = extractImageReferences(md);
-      expect(images).toHaveLength(0);
-    });
-
-    it('returns empty array for no images', () => {
-      const images = extractImageReferences('plain text');
-      expect(images).toHaveLength(0);
-    });
-  });
-
-  describe('getExtensionFromUrl', () => {
-    it('detects .png extension', () => {
-      expect(getExtensionFromUrl('https://example.com/image.png')).toBe('.png');
-    });
-
-    it('detects .jpg extension', () => {
-      expect(getExtensionFromUrl('https://example.com/photo.jpg')).toBe('.jpg');
-    });
-
-    it('detects .gif extension', () => {
-      expect(getExtensionFromUrl('https://example.com/anim.gif')).toBe('.gif');
-    });
-
-    it('handles query parameters', () => {
-      expect(
-        getExtensionFromUrl('https://example.com/image.png?w=800&h=600')
-      ).toBe('.png');
-    });
-
-    it('defaults to .png for unknown extension', () => {
-      expect(getExtensionFromUrl('https://example.com/image')).toBe('.png');
-    });
-  });
-
-  describe('generateLocalFilename', () => {
-    it('generates zero-padded filename', () => {
-      expect(generateLocalFilename('https://example.com/img.png', 0)).toBe(
-        'image-01.png'
-      );
-      expect(generateLocalFilename('https://example.com/img.jpg', 9)).toBe(
-        'image-10.jpg'
-      );
-    });
-
-    it('preserves file extension from URL', () => {
-      expect(generateLocalFilename('https://example.com/photo.gif', 2)).toBe(
-        'image-03.gif'
-      );
-    });
-  });
-
-  describe('generateFigureFilename', () => {
-    it('generates figure-numbered filenames while preserving extensions', () => {
-      expect(
-        generateFigureFilename('https://habrastorage.org/getpro/habr/a.png', 0)
-      ).toBe('figure-1.png');
-      expect(
-        generateFigureFilename('https://habrastorage.org/getpro/habr/a.jpg', 9)
-      ).toBe('figure-10.jpg');
-    });
-  });
-
-  describe('localizeImages', () => {
-    it('accepts a custom filename generator for archive-compatible paths', async () => {
-      const markdown =
-        '![Figure 1](https://habrastorage.org/r/w1560/getpro/habr/figure.png "Figure 1")';
-
-      const result = await localizeImages(markdown, {
-        dryRun: true,
-        filenameGenerator: generateFigureFilename,
-      });
-
-      expect(result.markdown).toBe('![Figure 1](images/figure-1.png)');
-      expect(result.metadata).toEqual([
-        {
-          index: 1,
-          originalUrl:
-            'https://habrastorage.org/r/w1560/getpro/habr/figure.png',
-          altText: 'Figure 1',
-          localPath: 'images/figure-1.png',
-        },
-      ]);
-    });
-  });
-});

package/tests/unit/markdown.test.js

@@ -1,36 +0,0 @@
-import { shouldRenderHtmlWithBrowser } from '../../src/markdown.js';
-
-describe('markdown rendering decisions', () => {
-  it('uses browser rendering for JavaScript-heavy HTML shells', () => {
-    const html = `<!doctype html>
-      <html>
-        <head><script src="/_next/static/app.js"></script></head>
-        <body><div id="__next"></div></body>
-      </html>`;
-
-    expect(shouldRenderHtmlWithBrowser(html)).toBe(true);
-  });
-
-  it('keeps plain HTML on the direct fetch path', () => {
-    const html = `<!doctype html>
-      <html>
-        <head><title>Static page</title></head>
-        <body><main><h1>Static content</h1></main></body>
-      </html>`;
-
-    expect(shouldRenderHtmlWithBrowser(html)).toBe(false);
-  });
-
-  it('keeps static pages with incidental scripts on the direct fetch path', () => {
-    const html = `<!doctype html>
-      <html>
-        <head>
-          <title>Static page</title>
-          <script src="/analytics.js"></script>
-        </head>
-        <body><main><h1>Static content</h1></main></body>
-      </html>`;
-
-    expect(shouldRenderHtmlWithBrowser(html)).toBe(false);
-  });
-});