ep_images_extended 1.0.61 → 1.0.62

package/README.md

@@ -4,7 +4,7 @@
 `ep_images_extended` builds on `ep_image_insert` and other image upload plugins.  
 The main difference is that images are built as custom span structures using the CSS background image attribute. This bypasses the Content Collector which always requires images to be block-level styles (so they couldn't share the line with text). As a result, we can now type around images, which allows the creation of more interactive pad content. The plugin includes some other features like click + drag resize, image float, and cut/copy/delete through a context menu. It was designed for compatibility with my forthcoming tables plugin. It's a pretty heavyweight plugin (some would say overengineered), because I was prioritizing meeting functional requirements for my project. Etherpad wizards might have tips for optimization, it would surely be appreciated.
 
-![Demo](https://i.imgur.com/JdyA50U.png)
+![Demo](https://i.imgur.com/qGOBRep.png)
 
 ## Installation
 

package/index.js

@@ -22,8 +22,14 @@ try {
   console.warn('[ep_images_extended] AWS SDK not installed; s3_presigned storage will not work.');
 }
 
-// Removed log4js; use built-in console for logging
-const logger = console;
+// Replaced log4js with lightweight console wrapper to avoid external dependency
+// while preserving same API surface used below.
+const logger = {
+  debug: console.debug.bind(console),
+  info: console.info.bind(console),
+  warn: console.warn.bind(console),
+  error: console.error.bind(console),
+};
 
 // Simple in-memory IP rate limiter
 const _presignRateStore = new Map();

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "ep_images_extended",
   "description": "Insert images inline with text, float them, resize them, and more.",
-  "version": "1.0.61",
+  "version": "1.0.62",
   "author": {
     "name": "DCastelone",
     "url": "https://github.com/dcastelone"

package/static/js/clientHooks.js

@@ -1,6 +1,18 @@
 'use strict';
 // Modified from ep_image_insert 1.0.7 
 
+// Optional helper (shared with ep_docx_html_customizer) that provides a CORS fetch with
+// automatic same-origin proxy fallback.  If the plugin is not present we simply fall back
+// to the native fetch logic.
+let fetchWithCorsProxy;
+try {
+  ({fetchWithCorsProxy} = require('../../../ep_docx_html_customizer/transform_common'));
+} catch (_) { /* helper not available – fallback to plain fetch */ }
+
+if (!fetchWithCorsProxy && typeof window !== 'undefined') {
+  fetchWithCorsProxy = window.fetchWithCorsProxy;
+}
+
 // Simple UUID generator
 function generateUUID() {
   return 'xxxx-xxxx-4xxx-yxxx-xxxx'.replace(/[xy]/g, function(c) {
@@ -1146,9 +1158,19 @@ exports.postAceInit = function (hook, context) {
                 // For HTTP URLs, fetch and convert to PNG
                 else if (imageSrc.startsWith('http')) {
                     // First try to fetch the image
-                    const response = await fetch(imageSrc, { mode: 'cors' });
-                    if (!response.ok) {
-                        throw new Error(`HTTP ${response.status}`);
+                    let response;
+                    try {
+                        if (typeof fetchWithCorsProxy === 'function') {
+                            response = await fetchWithCorsProxy(imageSrc);
+                        } else {
+                            response = await fetch(imageSrc, {mode: 'cors'});
+                            if (!response.ok) throw new Error(`status ${response.status}`);
+                        }
+                    } catch (fetchErr) {
+                        // Remote fetch failed – fall back to copying the URL as plain text
+                        await navigator.clipboard.writeText(imageSrc);
+                        showCopyFeedback(shouldCut ? 'Image URL cut to clipboard (fallback)' : 'Image URL copied to clipboard (fallback)');
+                        return; // Abort PNG-blob path
                     }
                     
                     // Convert the fetched image to a data URL, then to PNG