🏷️(backend) add content-type to uploaded files

All the uploaded files had the content-type set
to `application/octet-stream`. It create issues
when the file is downloaded from the frontend
because the browser doesn't know how to handle
the file.
We now determine the content-type of the file
and set it to the file object.

.github/workflows/impress.yml

@@ -206,10 +206,11 @@ jobs:
       - name: Install development dependencies
         run: pip install --user .[dev]
 
-      - name: Install gettext (required to compile messages)
+      - name: Install gettext (required to compile messages) and MIME support
         run: |
           sudo apt-get update
-          sudo apt-get install -y gettext pandoc
+          sudo apt-get install -y gettext pandoc shared-mime-info
+          sudo wget https://svn.apache.org/repos/asf/httpd/httpd/trunk/docs/conf/mime.types -O /etc/mime.types
 
       - name: Generate a MO file from strings extracted from the project
         run: python manage.py compilemessages

CHANGELOG.md

@@ -13,7 +13,7 @@ and this project adheres to
 
 - github actions to managed Crowdin workflow
 - 📈Integrate Posthog #540
-
+- 🏷️(backend) add content-type to uploaded files #552
 
 ## Changed
 

Dockerfile

@@ -76,6 +76,8 @@ RUN apk add \
   pango \
   shared-mime-info
 
+RUN wget https://svn.apache.org/repos/asf/httpd/httpd/trunk/docs/conf/mime.types -O /etc/mime.types
+
 # Copy entrypoint
 COPY ./docker/files/usr/local/bin/entrypoint /usr/local/bin/entrypoint
 

src/backend/core/api/serializers.py

@@ -388,6 +388,7 @@ class FileUploadSerializer(serializers.Serializer):
             raise serializers.ValidationError("Could not determine file extension.")
 
         self.context["expected_extension"] = extension
+        self.context["content_type"] = magic_mime_type
 
         return file
 
@@ -395,6 +396,7 @@ class FileUploadSerializer(serializers.Serializer):
         """Override validate to add the computed extension to validated_data."""
         attrs["expected_extension"] = self.context["expected_extension"]
         attrs["is_unsafe"] = self.context["is_unsafe"]
+        attrs["content_type"] = self.context["content_type"]
         return attrs
 
 

src/backend/core/api/viewsets.py

@@ -605,7 +605,10 @@ class DocumentViewSet(
         key = f"{document.key_base}/{ATTACHMENTS_FOLDER:s}/{file_id!s}.{extension:s}"
 
         # Prepare metadata for storage
-        extra_args = {"Metadata": {"owner": str(request.user.id)}}
+        extra_args = {
+            "Metadata": {"owner": str(request.user.id)},
+            "ContentType": serializer.validated_data["content_type"],
+        }
         if serializer.validated_data["is_unsafe"]:
             extra_args["Metadata"]["is_unsafe"] = "true"
 

src/backend/core/tests/documents/test_api_documents_attachment_upload.py

@@ -64,12 +64,22 @@ def test_api_documents_attachment_upload_anonymous_success():
     assert response.status_code == 201
 
     pattern = re.compile(rf"^/media/{document.id!s}/attachments/(.*)\.png")
-    match = pattern.search(response.json()["file"])
+    file_path = response.json()["file"]
+    match = pattern.search(file_path)
     file_id = match.group(1)
 
     # Validate that file_id is a valid UUID
     uuid.UUID(file_id)
 
+    # Now, check the metadata of the uploaded file
+    key = file_path.replace("/media", "")
+    file_head = default_storage.connection.meta.client.head_object(
+        Bucket=default_storage.bucket_name, Key=key
+    )
+
+    assert file_head["Metadata"] == {"owner": "None"}
+    assert file_head["ContentType"] == "image/png"
+
 
 @pytest.mark.parametrize(
     "reach, role",
@@ -206,6 +216,7 @@ def test_api_documents_attachment_upload_success(via, role, mock_user_teams):
         Bucket=default_storage.bucket_name, Key=key
     )
     assert file_head["Metadata"] == {"owner": str(user.id)}
+    assert file_head["ContentType"] == "image/png"
 
 
 def test_api_documents_attachment_upload_invalid(client):
@@ -247,16 +258,18 @@ def test_api_documents_attachment_upload_size_limit_exceeded(settings):
 
 
 @pytest.mark.parametrize(
-    "name,content,extension",
+    "name,content,extension,content_type",
     [
-        ("test.exe", b"text", "exe"),
+        ("test.exe", b"text", "exe", "text/plain"),
-        ("test", b"text", "txt"),
+        ("test", b"text", "txt", "text/plain"),
-        ("test.aaaaaa", b"test", "txt"),
+        ("test.aaaaaa", b"test", "txt", "text/plain"),
-        ("test.txt", PIXEL, "txt"),
+        ("test.txt", PIXEL, "txt", "image/png"),
-        ("test.py", b"#!/usr/bin/python", "py"),
+        ("test.py", b"#!/usr/bin/python", "py", "text/plain"),
     ],
 )
-def test_api_documents_attachment_upload_fix_extension(name, content, extension):
+def test_api_documents_attachment_upload_fix_extension(
+    name, content, extension, content_type
+):
     """
     A file with no extension or a wrong extension is accepted and the extension
     is corrected in storage.
@@ -287,6 +300,7 @@ def test_api_documents_attachment_upload_fix_extension(name, content, extension)
         Bucket=default_storage.bucket_name, Key=key
     )
     assert file_head["Metadata"] == {"owner": str(user.id), "is_unsafe": "true"}
+    assert file_head["ContentType"] == content_type
 
 
 def test_api_documents_attachment_upload_empty_file():
@@ -335,3 +349,4 @@ def test_api_documents_attachment_upload_unsafe():
         Bucket=default_storage.bucket_name, Key=key
     )
     assert file_head["Metadata"] == {"owner": str(user.id), "is_unsafe": "true"}
+    assert file_head["ContentType"] == "application/octet-stream"