magpie-html 0.1.5 → 0.2.1

package/dist/index.d.cts

@@ -454,9 +454,10 @@ declare function isProbablyReaderable(doc: Document, options?: {
  * - `'rss'` - RSS 2.0, 0.9x, or RSS 1.0 (RDF)
  * - `'atom'` - Atom 1.0
  * - `'json-feed'` - JSON Feed 1.0 or 1.1
+ * - `'sitemap'` - XML Sitemap (urlset or sitemapindex)
  * - `'unknown'` - Format could not be determined
  */
-type FeedFormat = 'rss' | 'atom' | 'json-feed' | 'unknown';
+type FeedFormat = 'rss' | 'atom' | 'json-feed' | 'sitemap' | 'unknown';
 /**
  * Detect feed format from content string.
  *

package/dist/index.d.ts

@@ -454,9 +454,10 @@ declare function isProbablyReaderable(doc: Document, options?: {
  * - `'rss'` - RSS 2.0, 0.9x, or RSS 1.0 (RDF)
  * - `'atom'` - Atom 1.0
  * - `'json-feed'` - JSON Feed 1.0 or 1.1
+ * - `'sitemap'` - XML Sitemap (urlset or sitemapindex)
  * - `'unknown'` - Format could not be determined
  */
-type FeedFormat = 'rss' | 'atom' | 'json-feed' | 'unknown';
+type FeedFormat = 'rss' | 'atom' | 'json-feed' | 'sitemap' | 'unknown';
 /**
  * Detect feed format from content string.
  *

package/dist/index.js

@@ -1075,8 +1075,12 @@ function extractEntry(entryElement) {
 function removeComments(xml) {
   return xml.replace(/<!--[\s\S]*?-->/g, "");
 }
+function removeDoctype(xml) {
+  return xml.replace(/<!DOCTYPE[^>]*>/gi, "");
+}
 function parseAtomXML(xml) {
-  const withoutComments = removeComments(xml);
+  const withoutDoctype = removeDoctype(xml);
+  const withoutComments = removeComments(withoutDoctype);
   const { text: cleanedXML, cdataMap } = extractCDATA(withoutComments);
   const root = parseElement(cleanedXML, 0, null, cdataMap).element;
   return root;
@@ -1544,6 +1548,9 @@ function detectFormat(content) {
   if (cleaned.match(/<channel[\s>]/i)) {
     return "rss";
   }
+  if ((cleaned.match(/<urlset[\s>]/i) || cleaned.match(/<sitemapindex[\s>]/i)) && cleaned.includes("sitemaps.org")) {
+    return "sitemap";
+  }
   return "unknown";
 }
 function isFeed(content) {
@@ -1881,12 +1888,20 @@ function parseRSSDate(dateString) {
 // src/feed/rss/xml-parser.ts
 function parseRSSXML(xml) {
   const cleaned = cleanXMLDeclaration(xml);
-  const root = parseElement2(cleaned, 0).element;
+  const withoutDoctype = removeDoctype2(cleaned);
+  const withoutComments = removeComments2(withoutDoctype);
+  const root = parseElement2(withoutComments, 0).element;
   return root;
 }
 function cleanXMLDeclaration(xml) {
   return xml.replace(/<\?xml[^?]*\?>/g, "").trim();
 }
+function removeDoctype2(xml) {
+  return xml.replace(/<!DOCTYPE[^>]*>/gi, "");
+}
+function removeComments2(xml) {
+  return xml.replace(/<!--[\s\S]*?-->/g, "");
+}
 function extractCDATA2(text) {
   const cdataMap = /* @__PURE__ */ new Map();
   let counter = 0;
@@ -2324,6 +2339,11 @@ function parseFeedAs(content, format, baseUrl) {
         original: jsonFeed
       };
     }
+    case "sitemap": {
+      throw new Error(
+        "Sitemaps cannot be parsed with parseFeed(). Use parseSitemap() from the sitemap module instead."
+      );
+    }
   }
 }
 
@@ -2405,6 +2425,10 @@ function detectEncoding(buffer, contentType) {
   }
   const preview = new Uint8Array(buffer.slice(0, 1024));
   const previewText = new TextDecoder("utf-8", { fatal: false }).decode(preview);
+  const xmlEncoding = parseCharsetFromXml(previewText);
+  if (xmlEncoding) {
+    return xmlEncoding;
+  }
   const metaEncoding = parseCharsetFromHtml(previewText);
   if (metaEncoding) {
     return metaEncoding;
@@ -2445,6 +2469,13 @@ function parseCharsetFromHtml(html) {
   }
   return null;
 }
+function parseCharsetFromXml(xml) {
+  const xmlDeclMatch = /<\?xml[^?]*encoding\s*=\s*["']([^"']+)["'][^?]*\?>/i.exec(xml);
+  if (xmlDeclMatch) {
+    return normalizeEncoding(xmlDeclMatch[1]);
+  }
+  return null;
+}
 function normalizeEncoding(encoding) {
   const normalized = encoding.toLowerCase().trim();
   const aliases = {