[pytx] Implement FileContent class

python-threatexchange/threatexchange/content_type/content_base.py

@@ -10,7 +10,7 @@
 from enum import Enum, auto
 import typing as t
 
-from threatexchange import common
+import common
 
 
 class ContentType:
@@ -32,7 +32,7 @@ def extract_additional_content(
         * Photo => run OCR and extract text
         * Video => break out photo thumbnail, close caption text, audio
         """
-        return []
+        return {}
 
 
 class RotationType(Enum):

python-threatexchange/threatexchange/content_type/file_content.py

@@ -0,0 +1,30 @@
+import typing as t
+from content_base import ContentType
+from photo import PhotoContent
+from video import VideoContent
+from PIL import Image
+
+class FileContent(ContentType):
+    """
+    ContentType representing a generic file. 
+
+    Determines if a file is a photo or video based on file extension.
+    """
+
+    VALID_PHOTO_EXTENSIONS = {ext.lower() for ext in Image.registered_extensions()}
+    VALID_VIDEO_EXTENSIONS = {".mp4", ".avi", ".mov", ".mkv"}
+
+    @classmethod
+    def get_content_type_from_filename(cls, file_name: str) -> t.Type[ContentType]:
+        """
+        Determines content type based on file extension.
+        """
+        file_extension = file_name.lower().rsplit('.', 1)[-1]
+        file_extension = f".{file_extension}"
+
+        if file_extension in cls.VALID_PHOTO_EXTENSIONS:
+            return PhotoContent
+        elif file_extension in cls.VALID_VIDEO_EXTENSIONS:
+            return VideoContent
+        else:
+            return None

python-threatexchange/threatexchange/content_type/tests/test_file_content_type.py

@@ -0,0 +1,27 @@
+import pytest
+from threatexchange.content_type.photo import PhotoContent
+from threatexchange.content_type.video import VideoContent
+from threatexchange.content_type.file_content import FileContent
+
+@pytest.mark.parametrize("file_name,expected_content_type", [
+    ("file.jpg", PhotoContent),
+    ("file.JPG", PhotoContent),
+    ("file.mp4", VideoContent),
+    ("file.MP4", VideoContent),
+    ("archive.photo.png", PhotoContent),
+    ("movie.backup.mp4", VideoContent),
+])
+def test_file_content_detection(file_name, expected_content_type):
+    """
+    Tests that FileContent correctly identifies the content type
+    as either PhotoContent or VideoContent based on file extension.
+    """
+    content_type = FileContent.get_content_type_from_filename(file_name)
+    assert content_type == expected_content_type, f"Failed for {file_name}"
+
+def test_unknown_file_type():
+    """
+    Tests that an unknown file type returns None.
+    """
+    file_content = FileContent.get_content_type_from_filename("file.txt")
+    assert file_content is None