fix: replace Markdown image links and raw <img> tags with alt text

src/markdown-helpers.ts

@@ -707,6 +707,11 @@ export const safelyJoinTokens = (tokens: Token[], options: JoinTokenOptions = {}
       joinedContent += safelyJoinTokens(tokenToCheck.children, options);
       continue;
     }
+    if (tokenToCheck.type === 'image') {
+      const altText = tokenToCheck.content || 'No Description';
+      joinedContent += `[Image: ${altText}]`;
+      continue;
+    }
     expect(tokenToCheck.children).to.equal(
       null,
       'There should be no nested children in the joinable tokens',
@@ -800,14 +805,26 @@ export const safelyJoinTokens = (tokens: Token[], options: JoinTokenOptions = {}
         break;
       case 'paragraph_open':
       case 'blockquote_close':
-      case 'html_block':
         break;
       case 'html_inline':
         // Replace <br> elements with a newline
         if (tokenToCheck.content.match(/<br\s*\/?>/)) {
           joinedContent += '\n';
         }
+      // Intentional fallthrough to next case
+      case 'html_block': {
+        // Replace <img> tags with [Image: <alt text>]
+        const imgMatch = tokenToCheck.content.match(/<img\b[^>]*>/i);
+        if (imgMatch) {
+          const altMatch = imgMatch[0].match(/alt=["']([^"']*)["']/i);
+          const altText = altMatch && altMatch[1] ? altMatch[1] : 'No Description';
+          joinedContent += `[Image: ${altText}]`;
+          if (tokenToCheck.type === 'html_block') {
+            joinedContent += '\n\n';
+          }
+        }
         break;
+      }
       case 'fence':
         if (options.parseCodeFences) {
           joinedContent += `\`\`\`\n${tokenToCheck.content}\`\`\`\n\n`;

tests/__snapshots__/markdown-helpers.spec.ts.snap

@@ -352,6 +352,10 @@ Process: Main
  _This class is not exported from the \`'electron'\` module. It is only available as a return value of other methods in the Electron API._"
 `;
 
+exports[`markdown-helpers > safelyJoinTokens > snapshots > should be correct for paragraph-with-image 1`] = `"This paragraph has an image [Image: screenshot] embedded in it."`;
+
+exports[`markdown-helpers > safelyJoinTokens > snapshots > should be correct for paragraph-with-image-no-alt 1`] = `"This paragraph has an image with no alt text [Image: No Description] in it."`;
+
 exports[`markdown-helpers > safelyJoinTokens > snapshots > should be correct for paragraph-with-inline-code 1`] = `"This is a \`inline code\` block that is \`code\` tagged."`;
 
 exports[`markdown-helpers > safelyJoinTokens > snapshots > should be correct for paragraph-with-links 1`] = `"This paragraph has a bunch of stuff.  Also some links Foo, these links should be stripped."`;

tests/fixtures/paragraph-with-image-no-alt.md

@@ -0,0 +1 @@
+This paragraph has an image with no alt text ![](https://example.com/screenshot.png) in it.

tests/fixtures/paragraph-with-image.md

@@ -0,0 +1 @@
+This paragraph has an image ![screenshot](https://example.com/screenshot.png) embedded in it.

tests/markdown-helpers.spec.ts

@@ -110,6 +110,46 @@ def fn():
         expect(safelyJoinTokens(tokens, { parseCodeFences: true })).toMatchSnapshot();
       });
     });
+
+    describe('images', () => {
+      it('should transform markdown image links with alt text', () => {
+        const tokens = getTokens('Hello ![screenshot](https://example.com/screenshot.png) world');
+        expect(safelyJoinTokens(tokens)).toBe('Hello [Image: screenshot] world');
+      });
+
+      it('should use "No Description" for markdown images without alt text', () => {
+        const tokens = getTokens('Hello ![](https://example.com/screenshot.png) world');
+        expect(safelyJoinTokens(tokens)).toBe('Hello [Image: No Description] world');
+      });
+
+      it('should handle raw <img> tags with alt text', () => {
+        const tokens = getTokens(
+          'Hello <img src="https://example.com/img.png" alt="my image"> world',
+        );
+        expect(safelyJoinTokens(tokens)).toBe('Hello [Image: my image] world');
+      });
+
+      it('should handle raw <img> tags without alt text', () => {
+        const tokens = getTokens('Hello <img src="https://example.com/img.png"> world');
+        expect(safelyJoinTokens(tokens)).toBe('Hello [Image: No Description] world');
+      });
+
+      it('should handle standalone <img> block tags with alt text', () => {
+        const tokens = getTokens(
+          'Before:\n\n<img width="487" alt="Image Before Adjustment" src="../images/before.png"/>\n\nAfter:\n\n<img width="487" alt="Image After Adjustment" src="../images/after.png"/>',
+        );
+        expect(safelyJoinTokens(tokens)).toBe(
+          'Before:\n\n[Image: Image Before Adjustment]\n\nAfter:\n\n[Image: Image After Adjustment]',
+        );
+      });
+
+      it('should handle standalone <img> block tags without alt text', () => {
+        const tokens = getTokens(
+          'Text\n\n<img width="487" src="../images/before.png"/>\n\nMore text',
+        );
+        expect(safelyJoinTokens(tokens)).toBe('Text\n\n[Image: No Description]\n\nMore text');
+      });
+    });
   });
 
   describe('extractStringEnum()', () => {