magpie-html 0.2.1 → 0.2.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/index.cjs CHANGED
@@ -2353,6 +2353,378 @@ function parseFeedAs(content, format, baseUrl) {
2353
2353
  }
2354
2354
  }
2355
2355
 
2356
+ // src/feed/sitemap/xml-parser.ts
2357
+ function parseSitemapXML(xml) {
2358
+ const cleaned = cleanXMLDeclaration2(xml);
2359
+ const withoutDoctype = removeDoctype3(cleaned);
2360
+ const withoutComments = removeComments3(withoutDoctype);
2361
+ const root = parseElement3(withoutComments, 0).element;
2362
+ return root;
2363
+ }
2364
+ function cleanXMLDeclaration2(xml) {
2365
+ return xml.replace(/<\?xml[^?]*\?>/g, "").trim();
2366
+ }
2367
+ function removeDoctype3(xml) {
2368
+ return xml.replace(/<!DOCTYPE[^>]*>/gi, "");
2369
+ }
2370
+ function removeComments3(xml) {
2371
+ return xml.replace(/<!--[\s\S]*?-->/g, "");
2372
+ }
2373
+ function extractCDATA3(text) {
2374
+ const cdataMap = /* @__PURE__ */ new Map();
2375
+ let counter = 0;
2376
+ const processed = text.replace(/<!\[CDATA\[([\s\S]*?)\]\]>/g, (_match, content) => {
2377
+ const placeholder = `__CDATA_${counter}__`;
2378
+ cdataMap.set(placeholder, content);
2379
+ counter++;
2380
+ return placeholder;
2381
+ });
2382
+ return { text: processed, cdataMap };
2383
+ }
2384
+ function restoreCDATA3(text, cdataMap) {
2385
+ let result = text;
2386
+ for (const [placeholder, content] of cdataMap.entries()) {
2387
+ result = result.replace(placeholder, content);
2388
+ }
2389
+ return result;
2390
+ }
2391
+ function parseAttributes3(tagContent) {
2392
+ const attributes = {};
2393
+ const attrRegex = /(\S+)=["']([^"']*)["']/g;
2394
+ let match = attrRegex.exec(tagContent);
2395
+ while (match !== null) {
2396
+ attributes[match[1]] = match[2];
2397
+ match = attrRegex.exec(tagContent);
2398
+ }
2399
+ return attributes;
2400
+ }
2401
+ function findClosingTag3(xml, tagName, startPos) {
2402
+ const openTag = `<${tagName}`;
2403
+ const closeTag = `</${tagName}>`;
2404
+ let depth = 1;
2405
+ let pos = startPos;
2406
+ while (pos < xml.length && depth > 0) {
2407
+ const nextOpen = xml.indexOf(openTag, pos);
2408
+ const nextClose = xml.indexOf(closeTag, pos);
2409
+ if (nextClose === -1) {
2410
+ return -1;
2411
+ }
2412
+ if (nextOpen !== -1 && nextOpen < nextClose) {
2413
+ depth++;
2414
+ pos = nextOpen + openTag.length;
2415
+ } else {
2416
+ depth--;
2417
+ if (depth === 0) {
2418
+ return nextClose;
2419
+ }
2420
+ pos = nextClose + closeTag.length;
2421
+ }
2422
+ }
2423
+ return -1;
2424
+ }
2425
+ function parseElement3(xml, startPos, parent = null, cdataMap) {
2426
+ const extracted = cdataMap ? { text: xml, cdataMap } : extractCDATA3(xml);
2427
+ const cleanedXML = extracted.text;
2428
+ const currentCdataMap = extracted.cdataMap;
2429
+ const openTagStart = cleanedXML.indexOf("<", startPos);
2430
+ if (openTagStart === -1) {
2431
+ throw new Error("No opening tag found");
2432
+ }
2433
+ const openTagEnd = cleanedXML.indexOf(">", openTagStart);
2434
+ if (openTagEnd === -1) {
2435
+ throw new Error("Unclosed opening tag");
2436
+ }
2437
+ const openTagContent = cleanedXML.substring(openTagStart + 1, openTagEnd);
2438
+ const isSelfClosing = openTagContent.endsWith("/");
2439
+ const tagContent = isSelfClosing ? openTagContent.slice(0, -1).trim() : openTagContent;
2440
+ const spaceIndex = tagContent.indexOf(" ");
2441
+ const tagName = spaceIndex === -1 ? tagContent : tagContent.substring(0, spaceIndex);
2442
+ const attributes = spaceIndex === -1 ? {} : parseAttributes3(tagContent.substring(spaceIndex));
2443
+ const element = {
2444
+ tagName,
2445
+ attributes,
2446
+ text: "",
2447
+ children: [],
2448
+ parent
2449
+ };
2450
+ if (isSelfClosing) {
2451
+ return { element, endPos: openTagEnd + 1, cdataMap: currentCdataMap };
2452
+ }
2453
+ const closingTagPos = findClosingTag3(cleanedXML, tagName, openTagEnd + 1);
2454
+ if (closingTagPos === -1) {
2455
+ throw new Error(`No closing tag found for <${tagName}>`);
2456
+ }
2457
+ const content = cleanedXML.substring(openTagEnd + 1, closingTagPos);
2458
+ if (content.includes("<")) {
2459
+ let pos = 0;
2460
+ const trimmedContent = content.trim();
2461
+ while (pos < trimmedContent.length) {
2462
+ const nextTag = trimmedContent.indexOf("<", pos);
2463
+ if (nextTag === -1) break;
2464
+ if (trimmedContent[nextTag + 1] === "/" || trimmedContent[nextTag + 1] === "!") {
2465
+ pos = nextTag + 1;
2466
+ continue;
2467
+ }
2468
+ try {
2469
+ const { element: child, endPos } = parseElement3(
2470
+ trimmedContent,
2471
+ nextTag,
2472
+ element,
2473
+ currentCdataMap
2474
+ );
2475
+ element.children.push(child);
2476
+ pos = endPos;
2477
+ } catch {
2478
+ pos = nextTag + 1;
2479
+ }
2480
+ }
2481
+ let textContent = content.replace(/<[^>]+>/g, "").trim();
2482
+ textContent = restoreCDATA3(textContent, currentCdataMap);
2483
+ element.text = textContent;
2484
+ } else {
2485
+ let textContent = content.trim();
2486
+ textContent = restoreCDATA3(textContent, currentCdataMap);
2487
+ element.text = textContent;
2488
+ }
2489
+ const closingTagEnd = closingTagPos + `</${tagName}>`.length;
2490
+ return { element, endPos: closingTagEnd, cdataMap: currentCdataMap };
2491
+ }
2492
+ function querySelector3(element, selector, caseSensitive = false) {
2493
+ const tagName = caseSensitive ? selector : selector.toLowerCase();
2494
+ const elementTag = caseSensitive ? element.tagName : element.tagName.toLowerCase();
2495
+ if (elementTag === tagName) {
2496
+ return element;
2497
+ }
2498
+ for (const child of element.children) {
2499
+ const found = querySelector3(child, selector, caseSensitive);
2500
+ if (found) return found;
2501
+ }
2502
+ return null;
2503
+ }
2504
+ function querySelectorAll3(element, selector, caseSensitive = false) {
2505
+ const results = [];
2506
+ const tagName = caseSensitive ? selector : selector.toLowerCase();
2507
+ const elementTag = caseSensitive ? element.tagName : element.tagName.toLowerCase();
2508
+ if (elementTag === tagName) {
2509
+ results.push(element);
2510
+ }
2511
+ for (const child of element.children) {
2512
+ results.push(...querySelectorAll3(child, selector, caseSensitive));
2513
+ }
2514
+ return results;
2515
+ }
2516
+ function getText2(element) {
2517
+ return element?.text || "";
2518
+ }
2519
+ function getChild(element, tagName) {
2520
+ const lowerTag = tagName.toLowerCase();
2521
+ return element.children.find((c) => c.tagName.toLowerCase() === lowerTag) || null;
2522
+ }
2523
+ function getChildren(element, tagName) {
2524
+ const lowerTag = tagName.toLowerCase();
2525
+ return element.children.filter((c) => c.tagName.toLowerCase() === lowerTag);
2526
+ }
2527
+
2528
+ // src/feed/sitemap/parse.ts
2529
+ function parseSitemap(xml, baseUrl) {
2530
+ const doc = parseSitemapXML(xml);
2531
+ const sitemapIndex = querySelector3(doc, "sitemapindex");
2532
+ if (sitemapIndex) {
2533
+ return parseSitemapIndex(sitemapIndex, baseUrl);
2534
+ }
2535
+ const urlset = querySelector3(doc, "urlset");
2536
+ if (urlset) {
2537
+ return parseUrlset(urlset, baseUrl);
2538
+ }
2539
+ const urls = querySelectorAll3(doc, "url");
2540
+ if (urls.length > 0) {
2541
+ return {
2542
+ sitemap: {
2543
+ type: "urlset",
2544
+ urls: urls.map((url) => extractUrl(url, baseUrl)),
2545
+ sitemaps: []
2546
+ },
2547
+ isIndex: false
2548
+ };
2549
+ }
2550
+ return {
2551
+ sitemap: {
2552
+ type: "urlset",
2553
+ urls: [],
2554
+ sitemaps: []
2555
+ },
2556
+ isIndex: false
2557
+ };
2558
+ }
2559
+ function parseSitemapIndex(element, baseUrl) {
2560
+ const sitemapElements = getChildren(element, "sitemap");
2561
+ const sitemaps = sitemapElements.map((el) => {
2562
+ const loc = getText2(getChild(el, "loc"));
2563
+ const lastmod = getText2(getChild(el, "lastmod")) || void 0;
2564
+ return {
2565
+ loc: baseUrl ? normalizeUrlHttps(baseUrl, loc) : loc,
2566
+ lastmod
2567
+ };
2568
+ });
2569
+ return {
2570
+ sitemap: {
2571
+ type: "sitemapindex",
2572
+ urls: [],
2573
+ sitemaps
2574
+ },
2575
+ isIndex: true
2576
+ };
2577
+ }
2578
+ function parseUrlset(element, baseUrl) {
2579
+ const urlElements = getChildren(element, "url");
2580
+ const urls = urlElements.map((el) => extractUrl(el, baseUrl));
2581
+ return {
2582
+ sitemap: {
2583
+ type: "urlset",
2584
+ urls,
2585
+ sitemaps: []
2586
+ },
2587
+ isIndex: false
2588
+ };
2589
+ }
2590
+ function extractUrl(element, baseUrl) {
2591
+ const rawLoc = getText2(getChild(element, "loc"));
2592
+ const loc = decodeXmlEntities(rawLoc);
2593
+ const lastmod = getText2(getChild(element, "lastmod")) || void 0;
2594
+ const changefreq = getText2(getChild(element, "changefreq")) || void 0;
2595
+ const priorityText = getText2(getChild(element, "priority"));
2596
+ const priority = priorityText ? Number.parseFloat(priorityText) : void 0;
2597
+ const result = {
2598
+ loc: baseUrl ? normalizeUrlHttps(baseUrl, loc) : loc,
2599
+ lastmod,
2600
+ changefreq,
2601
+ priority: priority && !Number.isNaN(priority) ? priority : void 0
2602
+ };
2603
+ const news = extractNews(element);
2604
+ if (news) {
2605
+ result.news = news;
2606
+ }
2607
+ const images = extractImages(element, baseUrl);
2608
+ if (images.length > 0) {
2609
+ result.images = images;
2610
+ }
2611
+ const videos = extractVideos(element, baseUrl);
2612
+ if (videos.length > 0) {
2613
+ result.videos = videos;
2614
+ }
2615
+ return result;
2616
+ }
2617
+ function extractNews(urlElement) {
2618
+ const newsEl = getChild(urlElement, "news:news") || getChild(urlElement, "news") || urlElement.children.find((c) => c.tagName.toLowerCase().endsWith(":news"));
2619
+ if (!newsEl) {
2620
+ return void 0;
2621
+ }
2622
+ const news = {};
2623
+ const pubEl = getChild(newsEl, "news:publication") || getChild(newsEl, "publication") || newsEl.children.find((c) => c.tagName.toLowerCase().endsWith(":publication"));
2624
+ if (pubEl) {
2625
+ const name = getText2(getChild(pubEl, "news:name")) || getText2(getChild(pubEl, "name")) || getText2(pubEl.children.find((c) => c.tagName.toLowerCase().endsWith(":name")));
2626
+ const language = getText2(getChild(pubEl, "news:language")) || getText2(getChild(pubEl, "language")) || getText2(pubEl.children.find((c) => c.tagName.toLowerCase().endsWith(":language")));
2627
+ if (name || language) {
2628
+ news.publication = {
2629
+ name: name || void 0,
2630
+ language: language || void 0
2631
+ };
2632
+ }
2633
+ }
2634
+ const pubDate = getText2(getChild(newsEl, "news:publication_date")) || getText2(getChild(newsEl, "publication_date")) || getText2(newsEl.children.find((c) => c.tagName.toLowerCase().endsWith(":publication_date")));
2635
+ if (pubDate) {
2636
+ news.publicationDate = pubDate;
2637
+ }
2638
+ const title = getText2(getChild(newsEl, "news:title")) || getText2(getChild(newsEl, "title")) || getText2(newsEl.children.find((c) => c.tagName.toLowerCase().endsWith(":title")));
2639
+ if (title) {
2640
+ news.title = decodeXmlEntities(title);
2641
+ }
2642
+ const keywords = getText2(getChild(newsEl, "news:keywords")) || getText2(getChild(newsEl, "keywords")) || getText2(newsEl.children.find((c) => c.tagName.toLowerCase().endsWith(":keywords")));
2643
+ if (keywords) {
2644
+ news.keywords = keywords.split(",").map((k) => k.trim());
2645
+ }
2646
+ const stockTickers = getText2(getChild(newsEl, "news:stock_tickers")) || getText2(getChild(newsEl, "stock_tickers")) || getText2(newsEl.children.find((c) => c.tagName.toLowerCase().endsWith(":stock_tickers")));
2647
+ if (stockTickers) {
2648
+ news.stockTickers = stockTickers.split(",").map((t) => t.trim());
2649
+ }
2650
+ return Object.keys(news).length > 0 ? news : void 0;
2651
+ }
2652
+ function extractImages(urlElement, baseUrl) {
2653
+ const imageElements = urlElement.children.filter(
2654
+ (c) => c.tagName.toLowerCase() === "image:image" || c.tagName.toLowerCase() === "image" || c.tagName.toLowerCase().endsWith(":image")
2655
+ );
2656
+ return imageElements.map((imgEl) => {
2657
+ const loc = getText2(getChild(imgEl, "image:loc")) || getText2(getChild(imgEl, "loc")) || getText2(imgEl.children.find((c) => c.tagName.toLowerCase().endsWith(":loc")));
2658
+ if (!loc) return null;
2659
+ const image = {
2660
+ loc: baseUrl ? normalizeUrlHttps(baseUrl, loc) : loc
2661
+ };
2662
+ const caption = getText2(getChild(imgEl, "image:caption")) || getText2(getChild(imgEl, "caption")) || getText2(imgEl.children.find((c) => c.tagName.toLowerCase().endsWith(":caption")));
2663
+ if (caption) image.caption = decodeXmlEntities(caption);
2664
+ const geoLocation = getText2(getChild(imgEl, "image:geo_location")) || getText2(getChild(imgEl, "geo_location")) || getText2(imgEl.children.find((c) => c.tagName.toLowerCase().endsWith(":geo_location")));
2665
+ if (geoLocation) image.geoLocation = geoLocation;
2666
+ const title = getText2(getChild(imgEl, "image:title")) || getText2(getChild(imgEl, "title")) || getText2(imgEl.children.find((c) => c.tagName.toLowerCase().endsWith(":title")));
2667
+ if (title) image.title = decodeXmlEntities(title);
2668
+ const license = getText2(getChild(imgEl, "image:license")) || getText2(getChild(imgEl, "license")) || getText2(imgEl.children.find((c) => c.tagName.toLowerCase().endsWith(":license")));
2669
+ if (license) image.license = baseUrl ? normalizeUrlHttps(baseUrl, license) : license;
2670
+ return image;
2671
+ }).filter((img) => img !== null);
2672
+ }
2673
+ function extractVideos(urlElement, baseUrl) {
2674
+ const videoElements = urlElement.children.filter(
2675
+ (c) => c.tagName.toLowerCase() === "video:video" || c.tagName.toLowerCase() === "video" || c.tagName.toLowerCase().endsWith(":video")
2676
+ );
2677
+ return videoElements.map((vidEl) => {
2678
+ const thumbnailLoc = getText2(getChild(vidEl, "video:thumbnail_loc")) || getText2(getChild(vidEl, "thumbnail_loc")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":thumbnail_loc")));
2679
+ const title = getText2(getChild(vidEl, "video:title")) || getText2(getChild(vidEl, "title")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":title")));
2680
+ const description = getText2(getChild(vidEl, "video:description")) || getText2(getChild(vidEl, "description")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":description")));
2681
+ if (!thumbnailLoc || !title || !description) return null;
2682
+ const video = {
2683
+ thumbnailLoc: baseUrl ? normalizeUrlHttps(baseUrl, thumbnailLoc) : thumbnailLoc,
2684
+ title: decodeXmlEntities(title),
2685
+ description: decodeXmlEntities(description)
2686
+ };
2687
+ const contentLoc = getText2(getChild(vidEl, "video:content_loc")) || getText2(getChild(vidEl, "content_loc")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":content_loc")));
2688
+ if (contentLoc)
2689
+ video.contentLoc = baseUrl ? normalizeUrlHttps(baseUrl, contentLoc) : contentLoc;
2690
+ const playerLoc = getText2(getChild(vidEl, "video:player_loc")) || getText2(getChild(vidEl, "player_loc")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":player_loc")));
2691
+ if (playerLoc) video.playerLoc = baseUrl ? normalizeUrlHttps(baseUrl, playerLoc) : playerLoc;
2692
+ const duration = getText2(getChild(vidEl, "video:duration")) || getText2(getChild(vidEl, "duration")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":duration")));
2693
+ if (duration) {
2694
+ const dur = Number.parseInt(duration, 10);
2695
+ if (!Number.isNaN(dur)) video.duration = dur;
2696
+ }
2697
+ const rating = getText2(getChild(vidEl, "video:rating")) || getText2(getChild(vidEl, "rating")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":rating")));
2698
+ if (rating) {
2699
+ const r = Number.parseFloat(rating);
2700
+ if (!Number.isNaN(r)) video.rating = r;
2701
+ }
2702
+ const viewCount = getText2(getChild(vidEl, "video:view_count")) || getText2(getChild(vidEl, "view_count")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":view_count")));
2703
+ if (viewCount) {
2704
+ const vc = Number.parseInt(viewCount, 10);
2705
+ if (!Number.isNaN(vc)) video.viewCount = vc;
2706
+ }
2707
+ const publicationDate = getText2(getChild(vidEl, "video:publication_date")) || getText2(getChild(vidEl, "publication_date")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":publication_date")));
2708
+ if (publicationDate) video.publicationDate = publicationDate;
2709
+ const familyFriendly = getText2(getChild(vidEl, "video:family_friendly")) || getText2(getChild(vidEl, "family_friendly")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":family_friendly")));
2710
+ if (familyFriendly) {
2711
+ video.familyFriendly = familyFriendly.toLowerCase() === "yes";
2712
+ }
2713
+ const category = getText2(getChild(vidEl, "video:category")) || getText2(getChild(vidEl, "category")) || getText2(vidEl.children.find((c) => c.tagName.toLowerCase().endsWith(":category")));
2714
+ if (category) video.category = category;
2715
+ const tagElements = vidEl.children.filter(
2716
+ (c) => c.tagName.toLowerCase() === "video:tag" || c.tagName.toLowerCase() === "tag" || c.tagName.toLowerCase().endsWith(":tag")
2717
+ );
2718
+ if (tagElements.length > 0) {
2719
+ video.tags = tagElements.map((t) => getText2(t)).filter(Boolean);
2720
+ }
2721
+ return video;
2722
+ }).filter((vid) => vid !== null);
2723
+ }
2724
+ function decodeXmlEntities(text) {
2725
+ return text.replace(/&lt;/g, "<").replace(/&gt;/g, ">").replace(/&amp;/g, "&").replace(/&quot;/g, '"').replace(/&apos;/g, "'").replace(/&#(\d+);/g, (_, code) => String.fromCharCode(Number.parseInt(code, 10))).replace(/&#x([0-9a-fA-F]+);/g, (_, code) => String.fromCharCode(Number.parseInt(code, 16)));
2726
+ }
2727
+
2356
2728
  // src/pluck/types.ts
2357
2729
  var PluckError = class extends Error {
2358
2730
  constructor(message) {
@@ -2729,6 +3101,12 @@ function parseHTML(html, baseUrl) {
2729
3101
  });
2730
3102
  return document;
2731
3103
  }
3104
+ function ensureDocument(input, baseUrl) {
3105
+ if (typeof input === "string") {
3106
+ return parseHTML(input, baseUrl);
3107
+ }
3108
+ return input;
3109
+ }
2732
3110
 
2733
3111
  // src/utils/meta-helpers.ts
2734
3112
  function getMetaContent(doc, name) {
@@ -2756,7 +3134,8 @@ function getMetaHttpEquiv(doc, httpEquiv) {
2756
3134
  }
2757
3135
 
2758
3136
  // src/metadata/opengraph/extract.ts
2759
- function extractOpenGraph(doc) {
3137
+ function extractOpenGraph(input) {
3138
+ const doc = ensureDocument(input);
2760
3139
  const metadata = {};
2761
3140
  metadata.title = getMetaProperty(doc, "og:title");
2762
3141
  metadata.type = getMetaProperty(doc, "og:type");
@@ -2781,7 +3160,7 @@ function extractOpenGraph(doc) {
2781
3160
  if (Object.keys(audio).length > 0) {
2782
3161
  metadata.audio = audio;
2783
3162
  }
2784
- const images = extractImages(doc);
3163
+ const images = extractImages2(doc);
2785
3164
  if (images.length > 0) {
2786
3165
  metadata.images = images;
2787
3166
  }
@@ -2850,7 +3229,7 @@ function extractAudio(doc) {
2850
3229
  Object.entries(audio).filter(([_, value]) => value !== void 0)
2851
3230
  );
2852
3231
  }
2853
- function extractImages(doc) {
3232
+ function extractImages2(doc) {
2854
3233
  const images = [];
2855
3234
  const imageUrls = getAllMetaPropertyValues(doc, "og:image");
2856
3235
  const imageSecureUrls = getAllMetaPropertyValues(doc, "og:image:secure_url");
@@ -2969,7 +3348,8 @@ function matchesAnyType(obj, targetTypes) {
2969
3348
  }
2970
3349
 
2971
3350
  // src/metadata/schema-org/extract.ts
2972
- function extractSchemaOrg(doc) {
3351
+ function extractSchemaOrg(input) {
3352
+ const doc = ensureDocument(input);
2973
3353
  const metadata = {
2974
3354
  jsonLd: []
2975
3355
  };
@@ -3046,7 +3426,8 @@ function organizeByType(metadata) {
3046
3426
  }
3047
3427
 
3048
3428
  // src/metadata/seo/extract.ts
3049
- function extractSEO(doc) {
3429
+ function extractSEO(input) {
3430
+ const doc = ensureDocument(input);
3050
3431
  const metadata = {};
3051
3432
  const titleElement = doc.querySelector("title");
3052
3433
  if (titleElement?.textContent) {
@@ -3078,7 +3459,8 @@ function extractSEO(doc) {
3078
3459
  }
3079
3460
 
3080
3461
  // src/metadata/twitter-card/extract.ts
3081
- function extractTwitterCard(doc) {
3462
+ function extractTwitterCard(input) {
3463
+ const doc = ensureDocument(input);
3082
3464
  const metadata = {};
3083
3465
  metadata.card = getMetaContent(doc, "twitter:card");
3084
3466
  metadata.site = getMetaContent(doc, "twitter:site");
@@ -3235,7 +3617,8 @@ function getAllLinksByPrefix(doc, relPrefix) {
3235
3617
  }
3236
3618
 
3237
3619
  // src/metadata/icons/extract.ts
3238
- function extractIcons(doc) {
3620
+ function extractIcons(input) {
3621
+ const doc = ensureDocument(input);
3239
3622
  const metadata = {};
3240
3623
  const iconLinks = getAllLinksByRels(doc, ["icon", "shortcut icon"]);
3241
3624
  for (const link of iconLinks) {
@@ -3416,7 +3799,8 @@ function parseSizeString(sizeStr) {
3416
3799
  }
3417
3800
 
3418
3801
  // src/metadata/language/extract.ts
3419
- function extractLanguage(doc) {
3802
+ function extractLanguage(input) {
3803
+ const doc = ensureDocument(input);
3420
3804
  const metadata = {};
3421
3805
  const htmlElement = doc.querySelector("html");
3422
3806
  if (htmlElement) {
@@ -3468,7 +3852,8 @@ function extractBestLanguage(doc) {
3468
3852
  }
3469
3853
 
3470
3854
  // src/metadata/links/extract.ts
3471
- function extractLinks3(doc, baseUrl, options = {}) {
3855
+ function extractLinks3(input, baseUrl, options = {}) {
3856
+ const doc = ensureDocument(input);
3472
3857
  const opts = normalizeOptions3(options);
3473
3858
  const effectiveBaseUrl = getEffectiveBaseUrl(doc, baseUrl);
3474
3859
  const baseOrigin = effectiveBaseUrl ? getOrigin(effectiveBaseUrl) : null;
@@ -3799,7 +4184,8 @@ function getStringProperty3(obj, prop) {
3799
4184
  }
3800
4185
 
3801
4186
  // src/metadata/canonical/extract.ts
3802
- function extractCanonical(doc) {
4187
+ function extractCanonical(input) {
4188
+ const doc = ensureDocument(input);
3803
4189
  const metadata = {};
3804
4190
  metadata.canonical = getLinkHref(doc, "canonical");
3805
4191
  const alternateLinks = getAllLinks(doc, "alternate");
@@ -3944,9 +4330,63 @@ async function gatherFeed(url) {
3944
4330
  }
3945
4331
  const response = await pluck(feedUrl);
3946
4332
  const content = await response.textUtf8();
4333
+ const format = detectFormat(content);
4334
+ if (format === "sitemap") {
4335
+ return normalizeSitemapToFeed(content, response.finalUrl);
4336
+ }
3947
4337
  const result = parseFeed(content, response.finalUrl);
3948
4338
  return result.feed;
3949
4339
  }
4340
+ function normalizeSitemapToFeed(content, baseUrl) {
4341
+ const result = parseSitemap(content, baseUrl);
4342
+ if (result.isIndex) {
4343
+ const items2 = result.sitemap.sitemaps.map((sitemap, index) => ({
4344
+ id: sitemap.loc || `sitemap-${index}`,
4345
+ url: sitemap.loc,
4346
+ title: `Sitemap: ${sitemap.loc}`,
4347
+ modified: sitemap.lastmod
4348
+ }));
4349
+ return {
4350
+ format: "sitemap",
4351
+ title: "Sitemap Index",
4352
+ url: baseUrl,
4353
+ items: items2
4354
+ };
4355
+ }
4356
+ const items = result.sitemap.urls.map((url, index) => {
4357
+ const item = {
4358
+ id: url.loc || `url-${index}`,
4359
+ url: url.loc,
4360
+ modified: url.lastmod
4361
+ };
4362
+ if (url.news) {
4363
+ item.title = url.news.title;
4364
+ item.published = url.news.publicationDate;
4365
+ if (url.news.publication?.name) {
4366
+ item.authors = [{ name: url.news.publication.name }];
4367
+ }
4368
+ if (url.news.keywords) {
4369
+ item.tags = url.news.keywords;
4370
+ }
4371
+ }
4372
+ if (url.images && url.images.length > 0) {
4373
+ item.image = url.images[0].loc;
4374
+ }
4375
+ return item;
4376
+ });
4377
+ let title = "Sitemap";
4378
+ try {
4379
+ const urlObj = new URL(baseUrl);
4380
+ title = `${urlObj.hostname} Sitemap`;
4381
+ } catch {
4382
+ }
4383
+ return {
4384
+ format: "sitemap",
4385
+ title,
4386
+ url: baseUrl,
4387
+ items
4388
+ };
4389
+ }
3950
4390
 
3951
4391
  // src/metadata/feed-discovery/heuristics.ts
3952
4392
  var COMMON_FEED_PATHS = [
@@ -3981,7 +4421,8 @@ function generateFeedSuggestions(documentUrl) {
3981
4421
  }
3982
4422
 
3983
4423
  // src/metadata/feed-discovery/extract.ts
3984
- function extractFeedDiscovery(doc, documentUrl) {
4424
+ function extractFeedDiscovery(input, documentUrl) {
4425
+ const doc = ensureDocument(input);
3985
4426
  const metadata = {
3986
4427
  feeds: []
3987
4428
  };
@@ -4158,7 +4599,8 @@ async function gatherWebsite(url) {
4158
4599
  }
4159
4600
 
4160
4601
  // src/metadata/analytics/extract.ts
4161
- function extractAnalytics(doc) {
4602
+ function extractAnalytics(input) {
4603
+ const doc = ensureDocument(input);
4162
4604
  const metadata = {};
4163
4605
  const scripts = doc.querySelectorAll("script");
4164
4606
  const googleAnalytics = /* @__PURE__ */ new Set();
@@ -4250,10 +4692,11 @@ function extractAnalytics(doc) {
4250
4692
  }
4251
4693
 
4252
4694
  // src/metadata/assets/extract.ts
4253
- function extractAssets(doc, baseUrl) {
4695
+ function extractAssets(input, baseUrl) {
4696
+ const doc = ensureDocument(input);
4254
4697
  const metadata = {};
4255
4698
  const effectiveBaseUrl = getEffectiveBaseUrl2(doc, baseUrl);
4256
- const images = extractImages2(doc, effectiveBaseUrl);
4699
+ const images = extractImages3(doc, effectiveBaseUrl);
4257
4700
  if (images.length > 0) {
4258
4701
  metadata.images = images;
4259
4702
  }
@@ -4307,7 +4750,7 @@ function getEffectiveBaseUrl2(doc, baseUrl) {
4307
4750
  }
4308
4751
  return null;
4309
4752
  }
4310
- function extractImages2(doc, baseUrl) {
4753
+ function extractImages3(doc, baseUrl) {
4311
4754
  const urls = /* @__PURE__ */ new Set();
4312
4755
  const imgElements = doc.querySelectorAll("img[src]");
4313
4756
  for (const img of Array.from(imgElements)) {
@@ -4577,7 +5020,8 @@ function extractConnectionHints(doc, baseUrl) {
4577
5020
  }
4578
5021
 
4579
5022
  // src/metadata/copyright/extract.ts
4580
- function extractCopyright(doc) {
5023
+ function extractCopyright(input) {
5024
+ const doc = ensureDocument(input);
4581
5025
  const metadata = {};
4582
5026
  metadata.copyright = getMetaContent(doc, "copyright");
4583
5027
  metadata.license = getLinkHref(doc, "license");
@@ -4613,7 +5057,8 @@ function parseCopyright(copyrightString) {
4613
5057
  }
4614
5058
 
4615
5059
  // src/metadata/dublin-core/extract.ts
4616
- function extractDublinCore(doc) {
5060
+ function extractDublinCore(input) {
5061
+ const doc = ensureDocument(input);
4617
5062
  const metadata = {};
4618
5063
  metadata.title = getMetaContent(doc, "DC.title") || getMetaContent(doc, "dcterms.title");
4619
5064
  metadata.description = getMetaContent(doc, "DC.description") || getMetaContent(doc, "dcterms.description");
@@ -4654,7 +5099,8 @@ function extractMultiValue(doc, field) {
4654
5099
  }
4655
5100
 
4656
5101
  // src/metadata/geo/extract.ts
4657
- function extractGeo(doc) {
5102
+ function extractGeo(input) {
5103
+ const doc = ensureDocument(input);
4658
5104
  const metadata = {};
4659
5105
  const geoPosition = getMetaContent(doc, "geo.position");
4660
5106
  if (geoPosition) {
@@ -4711,7 +5157,8 @@ function parseICBM(icbm) {
4711
5157
  }
4712
5158
 
4713
5159
  // src/metadata/monetization/extract.ts
4714
- function extractMonetization(doc) {
5160
+ function extractMonetization(input) {
5161
+ const doc = ensureDocument(input);
4715
5162
  const metadata = {};
4716
5163
  metadata.webMonetization = getMetaContent(doc, "monetization");
4717
5164
  metadata.paypalVerification = getMetaContent(doc, "paypal-site-verification");
@@ -4725,7 +5172,8 @@ function extractMonetization(doc) {
4725
5172
  }
4726
5173
 
4727
5174
  // src/metadata/news/extract.ts
4728
- function extractNews(doc) {
5175
+ function extractNews2(input) {
5176
+ const doc = ensureDocument(input);
4729
5177
  const metadata = {};
4730
5178
  const newsKeywords = getMetaContent(doc, "news_keywords");
4731
5179
  if (newsKeywords) {
@@ -4743,7 +5191,8 @@ function extractNews(doc) {
4743
5191
  }
4744
5192
 
4745
5193
  // src/metadata/pagination/extract.ts
4746
- function extractPagination(doc) {
5194
+ function extractPagination(input) {
5195
+ const doc = ensureDocument(input);
4747
5196
  const metadata = {};
4748
5197
  metadata.prev = getLinkHref(doc, "prev") || getLinkHref(doc, "previous");
4749
5198
  metadata.next = getLinkHref(doc, "next");
@@ -4842,7 +5291,8 @@ function parseKeyValueDirective(key, value, result) {
4842
5291
  }
4843
5292
 
4844
5293
  // src/metadata/robots/extract.ts
4845
- function extractRobots(doc) {
5294
+ function extractRobots(input) {
5295
+ const doc = ensureDocument(input);
4846
5296
  const metadata = {};
4847
5297
  const robotsContent = getMetaContent(doc, "robots");
4848
5298
  if (robotsContent) {
@@ -4876,7 +5326,8 @@ function extractRobots(doc) {
4876
5326
  }
4877
5327
 
4878
5328
  // src/metadata/security/extract.ts
4879
- function extractSecurity(doc) {
5329
+ function extractSecurity(input) {
5330
+ const doc = ensureDocument(input);
4880
5331
  const metadata = {};
4881
5332
  metadata.referrerPolicy = getMetaContent(doc, "referrer");
4882
5333
  metadata.contentSecurityPolicy = getMetaHttpEquiv(doc, "Content-Security-Policy");
@@ -4927,7 +5378,8 @@ function generateSitemapSuggestions(documentUrl) {
4927
5378
  }
4928
5379
 
4929
5380
  // src/metadata/sitemap-discovery/extract.ts
4930
- function extractSitemapDiscovery(doc, documentUrl) {
5381
+ function extractSitemapDiscovery(input, documentUrl) {
5382
+ const doc = ensureDocument(input);
4931
5383
  const metadata = {
4932
5384
  sitemaps: []
4933
5385
  };
@@ -4940,7 +5392,8 @@ function extractSitemapDiscovery(doc, documentUrl) {
4940
5392
  }
4941
5393
 
4942
5394
  // src/metadata/social-profiles/extract.ts
4943
- function extractSocialProfiles(doc) {
5395
+ function extractSocialProfiles(input) {
5396
+ const doc = ensureDocument(input);
4944
5397
  const metadata = {};
4945
5398
  metadata.twitter = getMetaContent(doc, "twitter:site") || getMetaContent(doc, "twitter:creator") || extractFromProperty(doc, "twitter:site") || extractFromProperty(doc, "twitter:creator");
4946
5399
  if (metadata.twitter) {
@@ -5093,7 +5546,8 @@ function categorizeSchemaProfile(url, metadata) {
5093
5546
  }
5094
5547
 
5095
5548
  // src/metadata/verification/extract.ts
5096
- function extractVerification(doc) {
5549
+ function extractVerification(input) {
5550
+ const doc = ensureDocument(input);
5097
5551
  const metadata = {};
5098
5552
  metadata.googleSiteVerification = getMetaContent(doc, "google-site-verification");
5099
5553
  metadata.msvalidate = getMetaContent(doc, "msvalidate.01");
@@ -6969,7 +7423,7 @@ exports.extractIcons = extractIcons;
6969
7423
  exports.extractLanguage = extractLanguage;
6970
7424
  exports.extractLinks = extractLinks3;
6971
7425
  exports.extractMonetization = extractMonetization;
6972
- exports.extractNews = extractNews;
7426
+ exports.extractNews = extractNews2;
6973
7427
  exports.extractOpenGraph = extractOpenGraph;
6974
7428
  exports.extractPagination = extractPagination;
6975
7429
  exports.extractRobots = extractRobots;