Bugfix to ensure image type is correctly extracted from content type (#…

…4062)

* fix: modified extension.py to split file extension by ';'

* test: added test that ensures that stripped file url is expected file type

* test: added pytest cases for test_get_extension_from_content_type()

* fix: added a . to one of the file extension parameters in pytest.mark.parametrize for test_get_extension_from_content_type()

* Apply linting

* Fix tests (initially)

* fix: added check for the case of a semicolon in the content_type and modified get_file_extension_from_content_type() to utilize mimetypes.guess_extension() to automatically find file extensions in strings

* fix: added check for test case for 'image/png;charset=UTF-8'

* fix: changed jpeg to jpg

* fix: changed test cases 'audio/midi' to compare to 'midi', and cases that return None when passed to mimetypes.guess_extension() to compare to None. Also, call mimetypes.guess_extension() in conditional header, assign it to variable, and then only strip it of '.' if it isn't None

* fix: removed test input '5' because not iterable and added test input 'foobar'

* Lint files

---------

Co-authored-by: Darien <[email protected]>
Co-authored-by: Madison Swain-Bowden <[email protected]>

api/api/utils/image_proxy/extension.py

@@ -1,4 +1,5 @@
 import logging
+import mimetypes
 from os.path import splitext
 from urllib.parse import urlparse
 
@@ -78,6 +79,10 @@ def _get_file_extension_from_content_type(content_type: str) -> str | None:
     Return the image extension if present in the Response's content type
     header.
     """
-    if content_type and "/" in content_type:
-        return content_type.split("/")[1]
+    if (
+        content_type
+        and "/" in content_type
+        and (ext := mimetypes.guess_extension(content_type.split(";")[0], strict=False))
+    ):
+        return ext.strip(".")
     return None

api/test/unit/utils/test_image_proxy.py

@@ -512,6 +512,28 @@ def test__get_extension_from_url(image_url, expected_ext):
     assert extension._get_file_extension_from_url(image_url) == expected_ext
 
 
+@pytest.mark.parametrize(
+    "content_type, expected_ext",
+    [
+        ("image/png;charset=UTF-8", "png"),
+        ("image/jpeg", "jpg"),
+        ("image/png", "png"),
+        ("image/gif", "gif"),
+        ("image/svg+xml", "svg"),
+        ("audio/midi", "midi"),
+        ("audio/mpeg", "mp3"),
+        ("audio/ogg", None),
+        ("audio/opus", "opus"),
+        ("audio/wav", None),
+        ("video/webm", "webm"),
+        (None, None),
+        ("foobar", None),
+    ],
+)
+def test_get_extension_from_content_type(content_type, expected_ext):
+    assert extension._get_file_extension_from_content_type(content_type) == expected_ext
+
+
 @pytest.mark.django_db
 @pytest.mark.parametrize("image_type", ["apng", "tiff", "bmp"])
 def test_photon_get_raises_by_not_allowed_types(photon_get, image_type):