Ignore data: URIs

Author: claudep

CHANGELOG

@@ -1,3 +1,7 @@
+Unreleased
+
+* Ignore `data:` URIs.
+
 2.4.0 (2025-09-28)
 
 * Add index to Link (David Venhoff, #202)

linkcheck/__init__.py

@@ -32,7 +32,7 @@ def handle_starttag(self, tag, attrs):
                 self.text += f' [image:{src[0]}] '
 
     def handle_endtag(self, tag):
-        if tag == 'a' and self.in_a:
+        if tag == 'a' and self.in_a and not self.url.startswith("data:"):
             self.urls.append((self.text[:256], self.url))
             self.in_a = False
             self.text = ''
@@ -48,7 +48,7 @@ class ImageLister(Lister):
     def handle_starttag(self, tag, attrs):
         if tag == 'img':
             src = [v for k, v in attrs if k == 'src']
-            if src:
+            if src and not src[0].startswith("data:"):
                 self.urls.append(('', src[0]))
 
 

linkcheck/tests/test_linkcheck.py

@@ -943,6 +943,19 @@ def test_urls_exceeding_max_length(self):
         )
         self.assertEqual(Url.objects.all().count(), 1)
 
+    def test_data_urls_ignored(self):
+        self.assertEqual(Url.objects.all().count(), 0)
+        Book.objects.create(
+            title="My Title",
+            description=(
+                'This is a normal link: <a href="https://www.example.org">Example</a>, '
+                'This is a data link: <a href="data:image/jpeg;base64,/9j/4AAQSkZJRgABAgAAZABkAAD">Example 2</a>, '
+                'This is a data img: <img src="data:image/jpeg;base64,/9j/4AAQGffsbRR4AAZABkAAD">'
+            )
+        )
+        # Only the normal link is extracted
+        self.assertEqual(Url.objects.all().count(), 1)
+
     def test_empty_url_field(self):
         """
         Test that URLField empty content is excluded depending on ignore_empty list.