fix: Use urllib.request.DataHandler in parse_data_uri

CHANGES.rst

@@ -134,6 +134,9 @@ Bugs fixed
   Patch by Jeremy Maitin-Shepard.
 * #13939: LaTeX: page break can separate admonition title from contents.
   Patch by Jean-François B.
+* #14006: Use ``urllib.request.DataHandler`` in ``parse_data_uri`` to fix
+  parsing of non-base64 data URI.
+  Patch by Shengyu Zhang.
 
 
 Testing

sphinx/util/images.py

@@ -2,9 +2,9 @@
 
 from __future__ import annotations
 
-import base64
 from pathlib import Path
 from typing import TYPE_CHECKING, NamedTuple, overload
+from urllib.request import DataHandler, build_opener
 
 import imagesize
 
@@ -90,26 +90,16 @@ 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:
+    try:
+        response = build_opener(DataHandler).open(uri)
+    except ValueError as e:
         msg = 'malformed data URI'
-        raise ValueError(msg)
-
-    # data:[<MIME-type>][;charset=<encoding>][;base64],<data>
-    mimetype = 'text/plain'
-    charset = 'US-ASCII'
-
-    properties, _, data = uri.partition(',')
-    for prop in properties.split(';'):
-        if prop == 'base64':
-            pass  # skip
-        elif prop.startswith('charset='):
-            charset = prop[8:]
-        elif prop:
-            mimetype = prop
-
-    image_data = base64.b64decode(data)
+        raise ValueError(msg) from e
+    info = response.info()
+    mimetype = info.get_content_type() or 'text/plain'
+    charset = info.get_content_charset() or 'US-ASCII'
+    image_data = response.read()
+
     return DataURI(mimetype, charset, image_data)
 
 

tests/test_util/test_util_images.py

@@ -86,3 +86,14 @@ 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'