recipe-manager/test_facebook_scraper.js

const assert = require('assert');

// 1. URL detection helper
function isFacebookUrl(url) {
  try {
    const parsed = new URL(url);
    const host = parsed.hostname.toLowerCase();
    const isFb = host.includes('facebook.com') || host.includes('fb.watch') || host.includes('fb.com');
    if (!isFb) return false;
    return host.includes('fb.watch') ||
           parsed.pathname.includes('/reel') ||
           parsed.pathname.includes('/watch') ||
           parsed.pathname.includes('/videos') ||
           parsed.pathname.includes('/share/');
  } catch (e) {
    return false;
  }
}

function cleanSocialCaption(raw) {
  if (!raw) return '';
  let text = raw.replace(/(?:\s*#[^\s#]+)+\s*$/g, '').trim();
  text = text.replace(/[\uFE00-\uFE0F\uD83C-\uDBFF\uDC00-\uDFFF\u2600-\u26FF\u2700-\u27BF]+$/, '').trim();
  return text;
}

function cleanFacebookCaption(raw) {
  if (!raw) return '';
  let text = raw.trim();
  const quotedMatch = text.match(/(?:on Facebook:\s*["“]|:\s*["“])([\s\S]*?)["”]?$/i);
  if (quotedMatch && quotedMatch[1]) {
    text = quotedMatch[1];
  } else {
    const colonQuote = text.indexOf(': "');
    if (colonQuote !== -1) {
      text = text.slice(colonQuote + 3).replace(/["”]$/, '');
    }
  }
  text = text.replace(/\s*\|\s*By\s+[^|]+\|\s*Facebook.*$/i, '');
  text = text.replace(/\s*\|\s*Facebook.*$/i, '');
  return cleanSocialCaption(text);
}

function isCaptionTruncated(text) {
  if (!text || typeof text !== 'string') return false;
  const trimmed = text.trim();
  if (/(?:\.{3,}|…)\s*(?:See\s+more)?$/i.test(trimmed)) return true;
  if (/\bSee\s+more$/i.test(trimmed)) return true;
  if (trimmed.length >= 180 && trimmed.length <= 320) {
    const endsWithPunctuation = /[.!?\n\r"']$/.test(trimmed);
    if (!endsWithPunctuation) return true;
  }
  return false;
}

function extractFullFacebookMessageFromHtml(html) {
  if (!html || typeof html !== 'string') return '';
  let longestText = '';

  const isBoilerplate = (str) => /log in|watch popular videos|sign up for facebook|connect with friends/i.test(str);

  const messageRegex = /"(?:message|savable_description)"\s*:\s*\{\s*"text"\s*:\s*"((?:[^"\\]|\\.)*)"/g;
  let match;
  while ((match = messageRegex.exec(html)) !== null) {
    try {
      const decoded = JSON.parse(`"${match[1]}"`);
      if (decoded && decoded.length > longestText.length && !isBoilerplate(decoded)) {
        longestText = decoded;
      }
    } catch (e) {}
  }

  const storyRegex = /"creation_story"\s*:\s*\{[\s\S]*?"message"\s*:\s*\{\s*"text"\s*:\s*"((?:[^"\\]|\\.)*)"/g;
  while ((match = storyRegex.exec(html)) !== null) {
    try {
      const decoded = JSON.parse(`"${match[1]}"`);
      if (decoded && decoded.length > longestText.length && !isBoilerplate(decoded)) {
        longestText = decoded;
      }
    } catch (e) {}
  }

  const ldJsonRegex = /<script\s+type="application\/ld\+json"[^>]*>([\s\S]*?)<\/script>/gi;
  while ((match = ldJsonRegex.exec(html)) !== null) {
    try {
      const data = JSON.parse(match[1]);
      const body = data.articleBody || data.description || '';
      if (body && body.length > longestText.length && !isBoilerplate(body)) {
        longestText = body;
      }
    } catch (e) {}
  }

  return cleanFacebookCaption(longestText);
}

console.log('Testing isFacebookUrl...');
assert.strictEqual(isFacebookUrl('https://www.facebook.com/reel/123456789'), true);
assert.strictEqual(isFacebookUrl('https://fb.watch/abcdef123/'), true);
assert.strictEqual(isFacebookUrl('https://m.facebook.com/reel/987654321/'), true);
assert.strictEqual(isFacebookUrl('https://www.facebook.com/watch/?v=11223344'), true);
assert.strictEqual(isFacebookUrl('https://www.facebook.com/share/r/abc123xyz/'), true);
assert.strictEqual(isFacebookUrl('https://www.instagram.com/reel/xyz123'), false);
console.log('✔ isFacebookUrl tests passed!');

console.log('\nTesting isCaptionTruncated...');
assert.strictEqual(isCaptionTruncated('1 cup flour, 2 eggs, 1 tbsp butter...'), true);
assert.strictEqual(isCaptionTruncated('1 cup flour, 2 eggs, 1 tbsp butter…'), true);
assert.strictEqual(isCaptionTruncated('1 cup flour, 2 eggs, 1 tbsp butter... See more'), true);
assert.strictEqual(isCaptionTruncated('1 cup flour, 2 eggs, 1 tbsp butter… See More'), true);
assert.strictEqual(isCaptionTruncated('This is a truncated facebook description that hits around two hundred characters and stops abruptly in the middle of a sentence without any final punctuation mark like a period so it indicates cut off'), true);
assert.strictEqual(isCaptionTruncated('1 cup flour, 2 eggs. Mix thoroughly and bake for 25 minutes at 350F.'), false);
console.log('✔ isCaptionTruncated tests passed!');

console.log('\nTesting extractFullFacebookMessageFromHtml...');
const sampleHtml = `
<html>
<head>
  <meta property="og:description" content="Chef Gordon: Crispy Garlic Butter Steak Bites. 1 lb sirloin steak, 2 tbsp butter...">
</head>
<body>
  <script type="application/json" data-sjs>
    {"require":[["RelayPrefetchedStreamCache","next",[],[{"message":{"text":"Crispy Garlic Butter Steak Bites.\\n\\nIngredients:\\n- 1 lb sirloin steak\\n- 2 tbsp butter\\n- 4 cloves garlic, minced\\n\\nDirections:\\n1. Sear steak in hot pan for 3 mins.\\n2. Add butter and garlic.\\n3. Serve warm!"}}]]]}
  </script>
</body>
</html>
`;

const extracted = extractFullFacebookMessageFromHtml(sampleHtml);
assert.ok(extracted.includes('4 cloves garlic, minced'));
assert.ok(extracted.includes('Directions:'));
assert.strictEqual(isCaptionTruncated(extracted), false);
console.log('✔ extractFullFacebookMessageFromHtml successfully retrieved complete message!');

console.log('\nAll Facebook Reel scraper & truncation tests passed successfully!');