✨(backend) allow uploading more types of attachments

We want to allow users to upload files to a document, not just images.
We try to enforce coherence between the file extension and the real
mime type of its content. If a file is deemed unsafe, it is still accepted
during upload and the information is stored as metadata on the object
for display to readers.

src/backend/core/api/serializers.py

@@ -6,6 +6,7 @@ from django.conf import settings
 from django.db.models import Q
 from django.utils.translation import gettext_lazy as _
 
+import magic
 from rest_framework import exceptions, serializers
 
 from core import models
@@ -219,16 +220,42 @@ class FileUploadSerializer(serializers.Serializer):
                 f"File size exceeds the maximum limit of {max_size:d} MB."
             )
 
-        # Validate file type
-        mime_type, _ = mimetypes.guess_type(file.name)
-        if mime_type not in settings.DOCUMENT_IMAGE_ALLOWED_MIME_TYPES:
-            mime_types = ", ".join(settings.DOCUMENT_IMAGE_ALLOWED_MIME_TYPES)
-            raise serializers.ValidationError(
-                f"File type '{mime_type:s}' is not allowed. Allowed types are: {mime_types:s}"
-            )
+        extension = file.name.rpartition(".")[-1] if "." in file.name else None
+
+        # Read the first few bytes to determine the MIME type accurately
+        mime = magic.Magic(mime=True)
+        magic_mime_type = mime.from_buffer(file.read(1024))
+        file.seek(0)  # Reset file pointer to the beginning after reading
+
+        self.context["is_unsafe"] = (
+            magic_mime_type in settings.DOCUMENT_UNSAFE_MIME_TYPES
+        )
+
+        extension_mime_type, _ = mimetypes.guess_type(file.name)
+
+        # Try guessing a coherent extension from the mimetype
+        if extension_mime_type != magic_mime_type:
+            self.context["is_unsafe"] = True
+
+        guessed_ext = mimetypes.guess_extension(magic_mime_type)
+        # Missing extensions or extensions longer than 5 characters (it's as long as an extension
+        # can be) are replaced by the extension we eventually guessed from mimetype.
+        if (extension is None or len(extension) > 5) and guessed_ext:
+            extension = guessed_ext[1:]
+
+        if extension is None:
+            raise serializers.ValidationError("Could not determine file extension.")
+
+        self.context["expected_extension"] = extension
 
         return file
 
+    def validate(self, attrs):
+        """Override validate to add the computed extension to validated_data."""
+        attrs["expected_extension"] = self.context["expected_extension"]
+        attrs["is_unsafe"] = self.context["is_unsafe"]
+        return attrs
+
 
 class TemplateSerializer(BaseResourceSerializer):
     """Serialize templates."""

src/backend/core/api/viewsets.py

@@ -1,6 +1,5 @@
 """API endpoints"""
 
-import os
 import re
 import uuid
 from urllib.parse import urlparse
@@ -476,15 +475,22 @@ class DocumentViewSet(
             return drf_response.Response(
                 serializer.errors, status=status.HTTP_400_BAD_REQUEST
             )
-        # Extract the file extension from the original filename
-        file = serializer.validated_data["file"]
-        extension = os.path.splitext(file.name)[1]
 
         # Generate a generic yet unique filename to store the image in object storage
         file_id = uuid.uuid4()
-        key = f"{document.key_base}/{ATTACHMENTS_FOLDER:s}/{file_id!s}{extension:s}"
+        extension = serializer.validated_data["expected_extension"]
+        key = f"{document.key_base}/{ATTACHMENTS_FOLDER:s}/{file_id!s}.{extension:s}"
+
+        # Prepare metadata for storage
+        metadata = {"Metadata": {"owner": str(request.user.id)}}
+        if serializer.validated_data["is_unsafe"]:
+            metadata["Metadata"]["is_unsafe"] = "true"
+
+        file = serializer.validated_data["file"]
+        default_storage.connection.meta.client.upload_fileobj(
+            file, default_storage.bucket_name, key, ExtraArgs=metadata
+        )
 
-        default_storage.save(key, file)
         return drf_response.Response(
             {"file": f"{settings.MEDIA_URL:s}{key:s}"}, status=status.HTTP_201_CREATED
         )