Support images with non-base64 data URIs (#14006)

Co-authored-by: Adam Turner <9087854+aa-turner@users.noreply.github.com>

Author: SilverRainZ

CHANGES.rst

@@ -154,6 +154,8 @@ Bugs fixed
   Patch by Harmen Stoppels.
 * #13944: autodoc: show traceback during import in human readable representation.
   Patch by Florian Best.
+* #14006: Support images with data URIs that aren't base64-encoded.
+  Patch by Shengyu Zhang and Adam Turner.
 
 
 Testing

sphinx/util/images.py

@@ -5,6 +5,7 @@
 import base64
 from pathlib import Path
 from typing import TYPE_CHECKING, NamedTuple, overload
+from urllib.parse import unquote_to_bytes
 
 import imagesize
 
@@ -90,7 +91,6 @@ def get_image_extension(mimetype: str) -> str | None:
 def parse_data_uri(uri: str) -> DataURI | None:
     if not uri.startswith('data:'):
         return None
-    uri = uri[5:]
 
     if ',' not in uri:
         msg = 'malformed data URI'
@@ -100,16 +100,19 @@ def parse_data_uri(uri: str) -> DataURI | None:
     mimetype = 'text/plain'
     charset = 'US-ASCII'
 
+    uri = uri[5:]
     properties, _, data = uri.partition(',')
     for prop in properties.split(';'):
         if prop == 'base64':
             pass  # skip
-        elif prop.startswith('charset='):
+        elif prop.lower().startswith('charset='):
             charset = prop[8:]
         elif prop:
-            mimetype = prop
+            mimetype = prop.lower()
 
-    image_data = base64.b64decode(data)
+    image_data = unquote_to_bytes(data)  # data might be percent-encoded
+    if properties.endswith(';base64'):
+        image_data = base64.decodebytes(image_data)
     return DataURI(mimetype, charset, image_data)
 
 

tests/test_util/test_util_images.py

@@ -86,3 +86,15 @@ def test_parse_data_uri() -> None:
     )
     with pytest.raises(ValueError, match=r'malformed data URI'):
         parse_data_uri(uri)
+
+    # not base64
+    uri = (
+        'data:image/svg+xml,%3Csvg%20width%3D%22100%22%20height%3D%22100%22%20'
+        'xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%22%3E%3Ccircle%20cx'
+        '%3D%2250%22%20cy%3D%2250%22%20r%3D%2240%22%20fill%3D%22blue%22%2F%3E'
+        '%3C%2Fsvg%3E'
+    )
+    image = parse_data_uri(uri)
+    assert image is not None
+    assert image.mimetype == 'image/svg+xml'
+    assert b'%' not in image.data