@rmdes/indiekit-endpoint-microsub 1.0.34 → 1.0.36

package/lib/feeds/normalizer.js

@@ -7,6 +7,28 @@ import crypto from "node:crypto";
 
 import sanitizeHtml from "sanitize-html";
 
+/**
+ * Extract image URLs from HTML content.
+ * Used as a fallback when no explicit photo/enclosure is provided.
+ * @param {string} html - HTML content (already sanitized)
+ * @returns {string[]} Array of image URLs
+ */
+function extractImagesFromHtml(html) {
+  if (!html) {
+    return [];
+  }
+  const urls = [];
+  const imgRegex = /<img[^>]+src=["']([^"']+)["'][^>]*>/gi;
+  let match;
+  while ((match = imgRegex.exec(html)) !== null) {
+    const src = match[1];
+    if (src && !urls.includes(src)) {
+      urls.push(src);
+    }
+  }
+  return urls;
+}
+
 /**
  * Parse a date string with fallback for non-standard formats
  * @param {string|Date} dateInput - Date string or Date object
@@ -232,6 +254,14 @@ export function normalizeItem(item, feedUrl, feedType) {
     }
   }
 
+  // Extract images from HTML content as fallback
+  if (!normalized.photo && normalized.content?.html) {
+    const extracted = extractImagesFromHtml(normalized.content.html);
+    if (extracted.length > 0) {
+      normalized.photo = extracted;
+    }
+  }
+
   return normalized;
 }
 
@@ -395,6 +425,14 @@ export function normalizeJsonFeedItem(item, feedUrl) {
     normalized["bookmark-of"] = [item.external_url];
   }
 
+  // Extract images from HTML content as fallback
+  if (!normalized.photo && normalized.content?.html) {
+    const extracted = extractImagesFromHtml(normalized.content.html);
+    if (extracted.length > 0) {
+      normalized.photo = extracted;
+    }
+  }
+
   return normalized;
 }
 
@@ -568,6 +606,14 @@ export function normalizeHfeedItem(entry, feedUrl) {
     normalized.syndication = properties.syndication;
   }
 
+  // Extract images from HTML content as fallback
+  if (!normalized.photo && normalized.content?.html) {
+    const extracted = extractImagesFromHtml(normalized.content.html);
+    if (extracted.length > 0) {
+      normalized.photo = extracted;
+    }
+  }
+
   return normalized;
 }
 

package/lib/storage/items.js

@@ -12,6 +12,27 @@ import {
   parseLimit,
 } from "../utils/pagination.js";
 
+/**
+ * Extract image URLs from HTML content (fallback for items without explicit photos)
+ * @param {string} html - HTML content
+ * @returns {string[]} Array of image URLs
+ */
+function extractImagesFromHtml(html) {
+  if (!html) {
+    return [];
+  }
+  const urls = [];
+  const imgRegex = /<img[^>]+src=["']([^"']+)["'][^>]*>/gi;
+  let match;
+  while ((match = imgRegex.exec(html)) !== null) {
+    const src = match[1];
+    if (src && !urls.includes(src)) {
+      urls.push(src);
+    }
+  }
+  return urls;
+}
+
 /**
  * Get items collection from application
  * @param {object} application - Indiekit application
@@ -201,6 +222,14 @@ function transformToJf2(item, userId) {
   const videos = normalizeMediaArray(item.video);
   const audios = normalizeMediaArray(item.audio);
 
+  // Fallback: extract images from HTML content if no explicit photos
+  if (photos.length === 0 && item.content?.html) {
+    const extracted = extractImagesFromHtml(item.content.html);
+    if (extracted.length > 0) {
+      photos.push(...extracted);
+    }
+  }
+
   if (photos.length > 0) jf2.photo = photos;
   if (videos.length > 0) jf2.video = videos;
   if (audios.length > 0) jf2.audio = audios;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@rmdes/indiekit-endpoint-microsub",
-  "version": "1.0.34",
+  "version": "1.0.36",
   "description": "Microsub endpoint for Indiekit. Enables subscribing to feeds and reading content using the Microsub protocol.",
   "keywords": [
     "indiekit",