Draft: ✨ Import of documents

CHANGELOG.md

@@ -6,6 +6,10 @@ and this project adheres to
 
 ## [Unreleased]
 
+### Added
+
+- ✨ Import of documents #7765
+
 ### Changed
 
 - ♿(frontend) improve accessibility:

Makefile

@@ -213,6 +213,7 @@ logs: ## display app-dev logs (follow mode)
 .PHONY: logs
 
 run-backend: ## Start only the backend application and all needed services
+	@$(COMPOSE) up --force-recreate -d docspec
 	@$(COMPOSE) up --force-recreate -d celery-dev
 	@$(COMPOSE) up --force-recreate -d y-provider-development
 	@$(COMPOSE) up --force-recreate -d nginx

compose.yml

@@ -217,3 +217,8 @@ services:
       kc_postgresql:
         condition: service_healthy
         restart: true
+
+  docspec:
+    image: ghcr.io/docspecio/api:2.4.4
+    ports:
+      - "4000:4000"

docs/env.md

@@ -103,6 +103,7 @@ These are the environment variables you can set for the `impress-backend` contai
 | USER_OIDC_ESSENTIAL_CLAIMS                      | Essential claims in OIDC token                                                                                              | []                                                                      |
 | Y_PROVIDER_API_BASE_URL                         | Y Provider url                                                                                                              |                                                                         |
 | Y_PROVIDER_API_KEY                              | Y provider API key                                                                                                          |                                                                         |
+| DOCSPEC_API_URL                                 | URL to endpoint of DocSpec conversion API                                                                                   |                                                                         |
 
 
 ## impress-frontend image

env.d/development/common

@@ -67,5 +67,7 @@ DJANGO_SERVER_TO_SERVER_API_TOKENS=server-api-token
 Y_PROVIDER_API_BASE_URL=http://y-provider-development:4444/api/
 Y_PROVIDER_API_KEY=yprovider-api-key
 
+DOCSPEC_API_URL=http://docspec:4000/conversion
+
 # Theme customization
-THEME_CUSTOMIZATION_CACHE_TIMEOUT=15
+THEME_CUSTOMIZATION_CACHE_TIMEOUT=15

env.d/development/common.e2e

@@ -6,4 +6,4 @@ Y_PROVIDER_API_BASE_URL=http://y-provider:4444/api/
 
 # Throttle
 API_DOCUMENT_THROTTLE_RATE=1000/min
-API_CONFIG_THROTTLE_RATE=1000/min
+API_CONFIG_THROTTLE_RATE=1000/min

src/backend/core/api/serializers.py

@@ -15,10 +15,11 @@
 from rest_framework import serializers
 
 from core import choices, enums, models, utils, validators
+from core.services import mime_types
 from core.services.ai_services import AI_ACTIONS
 from core.services.converter_services import (
     ConversionError,
-    YdocConverter,
+    Converter,
 )
 
 
@@ -188,6 +189,7 @@ class DocumentSerializer(ListDocumentSerializer):
 
     content = serializers.CharField(required=False)
     websocket = serializers.BooleanField(required=False, write_only=True)
+    file = serializers.FileField(required=False, write_only=True, allow_null=True)
 
     class Meta:
         model = models.Document
@@ -204,6 +206,7 @@ class Meta:
             "deleted_at",
             "depth",
             "excerpt",
+            "file",
             "is_favorite",
             "link_role",
             "link_reach",
@@ -461,7 +464,9 @@ def create(self, validated_data):
             language = user.language or language
 
         try:
-            document_content = YdocConverter().convert(validated_data["content"])
+            document_content = Converter().convert(
+                validated_data["content"], mime_types.MARKDOWN, mime_types.YJS
+            )
         except ConversionError as err:
             raise serializers.ValidationError(
                 {"content": ["Could not convert content"]}

src/backend/core/api/viewsets.py

@@ -37,16 +37,18 @@
 from rest_framework.permissions import AllowAny
 
 from core import authentication, choices, enums, models
+from core.services import mime_types
 from core.services.ai_services import AIService
 from core.services.collaboration_services import CollaborationService
 from core.services.converter_services import (
-    ServiceUnavailableError as YProviderServiceUnavailableError,
+    ConversionError,
+    Converter,
 )
 from core.services.converter_services import (
-    ValidationError as YProviderValidationError,
+    ServiceUnavailableError as YProviderServiceUnavailableError,
 )
 from core.services.converter_services import (
-    YdocConverter,
+    ValidationError as YProviderValidationError,
 )
 from core.tasks.mail import send_ask_for_access_mail
 from core.utils import extract_attachments, filter_descendants
@@ -504,6 +506,28 @@ def perform_create(self, serializer):
                 "IN SHARE ROW EXCLUSIVE MODE;"
             )
 
+        # Remove file from validated_data as it's not a model field
+        # Process it if present
+        uploaded_file = serializer.validated_data.pop("file", None)
+
+        # If a file is uploaded, convert it to Yjs format and set as content
+        if uploaded_file:
+            try:
+                file_content = uploaded_file.read()
+
+                converter = Converter()
+                converted_content = converter.convert(
+                    file_content,
+                    content_type=uploaded_file.content_type,
+                    accept=mime_types.YJS,
+                )
+                serializer.validated_data["content"] = converted_content
+                serializer.validated_data["title"] = uploaded_file.name
+            except ConversionError as err:
+                raise drf.exceptions.ValidationError(
+                    {"file": ["Could not convert file content"]}
+                ) from err
+
         obj = models.Document.add_root(
             creator=self.request.user,
             **serializer.validated_data,
@@ -1603,14 +1627,14 @@ def content(self, request, pk=None):
         if base64_content is not None:
             # Convert using the y-provider service
             try:
-                yprovider = YdocConverter()
+                yprovider = Converter()
                 result = yprovider.convert(
                     base64.b64decode(base64_content),
-                    "application/vnd.yjs.doc",
+                    mime_types.YJS,
                     {
-                        "markdown": "text/markdown",
-                        "html": "text/html",
-                        "json": "application/json",
+                        "markdown": mime_types.MARKDOWN,
+                        "html": mime_types.HTML,
+                        "json": mime_types.JSON,
                     }[content_format],
                 )
                 content = result

src/backend/core/services/converter_services.py

@@ -1,11 +1,14 @@
 """Y-Provider API services."""
 
+import typing
 from base64 import b64encode
 
 from django.conf import settings
 
 import requests
 
+from core.services import mime_types
+
 
 class ConversionError(Exception):
     """Base exception for conversion-related errors."""
@@ -19,8 +22,72 @@ class ServiceUnavailableError(ConversionError):
     """Raised when the conversion service is unavailable."""
 
 
+class ConverterProtocol(typing.Protocol):
+    """Protocol for converter classes."""
+
+    def convert(self, text, content_type, accept):
+        """Convert content from one format to another."""
+
+
+class Converter:
+    """Orchestrates conversion between different formats using specialized converters."""
+
+    docspec: ConverterProtocol
+    ydoc: ConverterProtocol
+
+    def __init__(self):
+        self.docspec = DocSpecConverter()
+        self.ydoc = YdocConverter()
+
+    def convert(self, data, content_type, accept):
+        """Convert input into other formats using external microservices."""
+
+        if content_type == mime_types.DOCX and accept == mime_types.YJS:
+            return self.convert(
+                self.docspec.convert(data, mime_types.DOCX, mime_types.BLOCKNOTE),
+                mime_types.BLOCKNOTE,
+                mime_types.YJS,
+            )
+
+        return self.ydoc.convert(data, content_type, accept)
+
+
+class DocSpecConverter:
+    """Service class for DocSpec conversion-related operations."""
+
+    def _request(self, url, data, content_type):
+        """Make a request to the DocSpec API."""
+
+        response = requests.post(
+            url,
+            headers={"Accept": mime_types.BLOCKNOTE},
+            files={"file": ("document.docx", data, content_type)},
+            timeout=settings.CONVERSION_API_TIMEOUT,
+            verify=settings.CONVERSION_API_SECURE,
+        )
+        response.raise_for_status()
+        return response
+
+    def convert(self, data, content_type, accept):
+        """Convert a Document to BlockNote."""
+        if not data:
+            raise ValidationError("Input data cannot be empty")
+
+        if content_type != mime_types.DOCX or accept != mime_types.BLOCKNOTE:
+            raise ValidationError(
+                f"Conversion from {content_type} to {accept} is not supported."
+            )
+
+        try:
+            return self._request(settings.DOCSPEC_API_URL, data, content_type).content
+        except requests.RequestException as err:
+            raise ServiceUnavailableError(
+                "Failed to connect to DocSpec conversion service",
+            ) from err
+
+
 class YdocConverter:
-    """Service class for conversion-related operations."""
+    """Service class for YDoc conversion-related operations."""
 
     @property
     def auth_header(self):
@@ -44,9 +111,7 @@ def _request(self, url, data, content_type, accept):
         response.raise_for_status()
         return response
 
-    def convert(
-        self, text, content_type="text/markdown", accept="application/vnd.yjs.doc"
-    ):
+    def convert(self, text, content_type=mime_types.MARKDOWN, accept=mime_types.YJS):
         """Convert a Markdown text into our internal format using an external microservice."""
 
         if not text:
@@ -59,14 +124,14 @@ def convert(
                 content_type,
                 accept,
             )
-            if accept == "application/vnd.yjs.doc":
+            if accept == mime_types.YJS:
                 return b64encode(response.content).decode("utf-8")
-            if accept in {"text/markdown", "text/html"}:
+            if accept in {mime_types.MARKDOWN, "text/html"}:
                 return response.text
-            if accept == "application/json":
+            if accept == mime_types.JSON:
                 return response.json()
             raise ValidationError("Unsupported format")
         except requests.RequestException as err:
             raise ServiceUnavailableError(
-                "Failed to connect to conversion service",
+                f"Failed to connect to YDoc conversion service {content_type}, {accept}",
             ) from err

src/backend/core/services/mime_types.py

@@ -0,0 +1,8 @@
+"""MIME type constants for document conversion."""
+
+BLOCKNOTE = "application/vnd.blocknote+json"
+YJS = "application/vnd.yjs.doc"
+MARKDOWN = "text/markdown"
+JSON = "application/json"
+DOCX = "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
+HTML = "text/html"

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

@@ -16,6 +16,7 @@
 from core import factories
 from core.api.serializers import ServerCreateDocumentSerializer
 from core.models import Document, Invitation, User
+from core.services import mime_types
 from core.services.converter_services import ConversionError, YdocConverter
 
 pytestmark = pytest.mark.django_db
@@ -191,7 +192,9 @@ def test_api_documents_create_for_owner_existing(mock_convert_md):
 
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     document = Document.objects.get()
     assert response.json() == {"id": str(document.id)}
@@ -236,7 +239,9 @@ def test_api_documents_create_for_owner_new_user(mock_convert_md):
 
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     document = Document.objects.get()
     assert response.json() == {"id": str(document.id)}
@@ -297,7 +302,9 @@ def test_api_documents_create_for_owner_existing_user_email_no_sub_with_fallback
 
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     document = Document.objects.get()
     assert response.json() == {"id": str(document.id)}
@@ -393,7 +400,9 @@ def test_api_documents_create_for_owner_new_user_no_sub_no_fallback_allow_duplic
         HTTP_AUTHORIZATION="Bearer DummyToken",
     )
     assert response.status_code == 201
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     document = Document.objects.get()
     assert response.json() == {"id": str(document.id)}
@@ -474,7 +483,9 @@ def test_api_documents_create_for_owner_with_default_language(
     )
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
     assert mock_send.call_args[0][3] == "de-de"
 
 
@@ -501,7 +512,9 @@ def test_api_documents_create_for_owner_with_custom_language(mock_convert_md):
 
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     assert len(mail.outbox) == 1
     email = mail.outbox[0]
@@ -537,7 +550,9 @@ def test_api_documents_create_for_owner_with_custom_subject_and_message(
 
     assert response.status_code == 201
 
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     assert len(mail.outbox) == 1
     email = mail.outbox[0]
@@ -571,7 +586,9 @@ def test_api_documents_create_for_owner_with_converter_exception(
         format="json",
         HTTP_AUTHORIZATION="Bearer DummyToken",
     )
-    mock_convert_md.assert_called_once_with("Document content")
+    mock_convert_md.assert_called_once_with(
+        "Document content", mime_types.MARKDOWN, mime_types.YJS
+    )
 
     assert response.status_code == 400
     assert response.json() == {"content": ["Could not convert content"]}