@rmdes/indiekit-endpoint-rss 1.0.2 → 1.0.3

package/lib/rss-client.js

@@ -158,26 +158,35 @@ export class RssClient {
         !cat.startsWith("user/")
       );
 
+      // For aggregators like FreshRSS, use origin info for the real source
+      const originTitle = item.origin?.title || null;
+      const originUrl = item.origin?.htmlUrl || null;
+
+      // Extract a meaningful description/summary
+      const description = this.extractDescription(content, item.summary);
+
       return {
         guid: item["frss:id"] || item.id || item.guid || link,
         title: item.title || "Untitled",
         link,
-        description: item.summary?.content ||
-                    (typeof item.summary === "string" ? item.summary : "") ||
-                    "",
+        description,
         content,
-        author: item.author || item.origin?.title || null,
+        // For author, prefer the actual author field, fall back to origin title only if no author
+        author: item.author || null,
         pubDate,
-        imageUrl: this.extractJsonItemImage(item),
+        imageUrl: this.extractJsonItemImage(item, content),
         categories,
         enclosure: item.enclosure || null,
-        // Preserve FreshRSS-specific metadata
+        // Preserve FreshRSS-specific metadata - use origin for real source feed
         origin: item.origin ? {
           streamId: item.origin.streamId,
           title: item.origin.title,
           htmlUrl: item.origin.htmlUrl,
           feedUrl: item.origin.feedUrl,
         } : null,
+        // Denormalized fields for easier display
+        sourceTitle: originTitle,
+        sourceUrl: originUrl,
       };
     });
 
@@ -226,30 +235,71 @@ export class RssClient {
   /**
    * Extract image from JSON feed item
    * @param {Object} item - JSON item
+   * @param {string} content - Pre-extracted content string
    * @returns {string|null}
    */
-  extractJsonItemImage(item) {
+  extractJsonItemImage(item, content = "") {
     // Direct image property
     if (item.image) return item.image;
 
-    // Media content
+    // Media content (various formats)
     if (item.media?.$?.url) return item.media.$.url;
+    if (item["media:content"]?.["$"]?.url) return item["media:content"]["$"].url;
+    if (item["media:thumbnail"]?.["$"]?.url) return item["media:thumbnail"]["$"].url;
 
     // Enclosure
     if (item.enclosure?.href && this.isImageUrl(item.enclosure.href)) {
       return item.enclosure.href;
     }
+    if (item.enclosure?.url && this.isImageUrl(item.enclosure.url)) {
+      return item.enclosure.url;
+    }
 
-    // Extract from content
-    const content = item.content?.content || item.content || item.summary?.content || "";
-    if (typeof content === "string") {
+    // Extract from content HTML
+    if (typeof content === "string" && content.length > 0) {
       const imgMatch = content.match(/<img[^>]+src=["']([^"']+)["']/i);
       if (imgMatch) return imgMatch[1];
     }
 
+    // Try item's own content fields
+    const itemContent = item.content?.content || item.content || item.summary?.content || "";
+    if (typeof itemContent === "string" && itemContent.length > 0) {
+      const imgMatch = itemContent.match(/<img[^>]+src=["']([^"']+)["']/i);
+      if (imgMatch) return imgMatch[1];
+    }
+
     return null;
   }
 
+  /**
+   * Extract a clean description/summary from content
+   * @param {string} content - HTML content
+   * @param {Object|string} summary - Summary object or string
+   * @returns {string}
+   */
+  extractDescription(content, summary) {
+    // First try the summary
+    let desc = "";
+    if (typeof summary === "string") {
+      desc = summary;
+    } else if (summary?.content) {
+      desc = summary.content;
+    }
+
+    // If no summary, extract from content
+    if (!desc && typeof content === "string" && content.length > 0) {
+      // Strip HTML tags and get first 300 chars
+      desc = content
+        .replace(/<[^>]+>/g, " ")  // Remove HTML tags
+        .replace(/\s+/g, " ")       // Normalize whitespace
+        .trim()
+        .slice(0, 300);
+      if (content.length > 300) desc += "...";
+    }
+
+    return desc;
+  }
+
   /**
    * Extract feed metadata
    * @param {Object} parsed - Parsed feed object

package/lib/utils.js

@@ -99,6 +99,14 @@ function isImageType(type) {
 export function formatItem(item, options = {}) {
   const { includeContent = false, descriptionLength = 200 } = options;
 
+  // Handle description - use item.description if available, otherwise generate from content
+  let description = item.description;
+  if (!description && item.content) {
+    description = truncateText(item.content, descriptionLength);
+  } else if (description) {
+    description = truncateText(description, descriptionLength);
+  }
+
   const formatted = {
     id: item._id?.toString(),
     feedId: item.feedId?.toString(),
@@ -106,14 +114,22 @@ export function formatItem(item, options = {}) {
     guid: item.guid,
     title: item.title,
     link: item.link,
-    description: truncateText(item.description, descriptionLength),
+    description: description || "",
     author: item.author,
     pubDate: item.pubDate?.toISOString(),
     imageUrl: item.imageUrl,
     categories: item.categories || [],
     fetchedAt: item.fetchedAt?.toISOString(),
+    // Source info for aggregators (like FreshRSS) - represents the original feed
+    sourceTitle: item.sourceTitle || null,
+    sourceUrl: item.sourceUrl || null,
   };
 
+  // Include origin object if present (for aggregator metadata)
+  if (item.origin) {
+    formatted.origin = item.origin;
+  }
+
   if (includeContent) {
     formatted.content = sanitizeHtml(item.content);
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@rmdes/indiekit-endpoint-rss",
-  "version": "1.0.2",
+  "version": "1.0.3",
   "description": "RSS feed reader endpoint for Indiekit. Aggregates multiple feeds, caches in MongoDB, displays on frontend.",
   "keywords": [
     "indiekit",