🐛(config) restrict uploadable file types for RAG

This allows to let the frontend know whi file types can be
uploaded to do RAG.
By default it's the format accepted by markitdown + PDF.

src/backend/conversations/settings.py

@@ -500,6 +500,42 @@ class Base(Configuration):
         environ_prefix=None,
     )
 
+    # Uploaded files
+    RAG_FILES_ACCEPTED_FORMATS = values.ListValue(
+        default=[
+            # docx files
+            "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+            # pptx files
+            "application/vnd.openxmlformats-officedocument.presentationml",
+            # xlsx and xls files
+            "application/vnd.ms-excel",
+            "application/excel",
+            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+            # txt and csv files
+            "text/plain",
+            "text/csv",
+            "application/csv",
+            # pdf files
+            "application/pdf",
+            # html files
+            "text/html",
+            "application/xhtml+xml",
+            # markdown files
+            "text/markdown",
+            "application/markdown",
+            "application/x-markdown",
+            # outlook msg files
+            "application/vnd.ms-outlook",
+            # images
+            "image/jpeg",
+            "image/png",
+            "image/gif",
+            "image/webp",
+        ],
+        environ_name="RAG_FILES_ACCEPTED_FORMATS",
+        environ_prefix=None,
+    )
+
     # Documents
     ALBERT_API_PARSE_TIMEOUT = values.PositiveIntegerValue(
         default=120,  # seconds

src/backend/core/api/viewsets.py

@@ -216,6 +216,8 @@ class ConfigView(drf.views.APIView):
 
         dict_settings["theme_customization"] = self._load_theme_customization()
 
+        dict_settings["chat_upload_accept"] = ",".join(settings.RAG_FILES_ACCEPTED_FORMATS)
+
         return drf.response.Response(dict_settings)
 
     def _load_theme_customization(self):

src/backend/core/tests/test_api_config.py

@@ -25,6 +25,10 @@ pytestmark = pytest.mark.django_db
     POSTHOG_KEY={"id": "132456", "host": "https://eu.i.posthog-test.com"},
     SENTRY_DSN="https://sentry.test/123",
     THEME_CUSTOMIZATION_FILE_PATH="",
+    RAG_FILES_ACCEPTED_FORMATS=[
+        "application/pdf",
+        "text/plain",
+    ],
 )
 @pytest.mark.parametrize("is_authenticated", [False, True])
 def test_api_config(is_authenticated):
@@ -55,6 +59,7 @@ def test_api_config(is_authenticated):
         "POSTHOG_KEY": {"id": "132456", "host": "https://eu.i.posthog-test.com"},
         "SENTRY_DSN": "https://sentry.test/123",
         "theme_customization": {},
+        "chat_upload_accept": "application/pdf,text/plain",
     }
 
 

src/frontend/apps/conversations/src/core/config/api/useConfig.tsx

@@ -23,6 +23,7 @@ export interface ConfigResponse {
   POSTHOG_KEY?: PostHogConf;
   SENTRY_DSN?: string;
   theme_customization?: ThemeCustomization;
+  chat_upload_accept?: string;
 }
 
 const LOCAL_STORAGE_KEY = 'conversations_config';

src/frontend/apps/e2e/__tests__/app-conversations/common.ts

@@ -18,6 +18,26 @@ export const CONFIG = {
   POSTHOG_KEY: {},
   SENTRY_DSN: null,
   theme_customization: {},
+  chat_upload_accept:
+    'application/vnd.openxmlformats-officedocument.wordprocessingml.document,' +
+    'application/vnd.openxmlformats-officedocument.presentationml,' +
+    'application/vnd.ms-excel,' +
+    'application/excel,' +
+    'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet,' +
+    'text/plain,' +
+    'text/csv,' +
+    'application/csv,' +
+    'application/pdf,' +
+    'text/html,' +
+    'application/xhtml+xml,' +
+    'text/markdown,' +
+    'application/markdown,' +
+    'application/x-markdown' +
+    ',application/vnd.ms-outlook,' +
+    'image/jpeg,' +
+    'image/png,' +
+    'image/gif,' +
+    'image/webp',
 } as const;
 
 export const overrideConfig = async (
@@ -54,9 +74,7 @@ export const keyCloakSignIn = async (
   const password = `password-e2e-${browserName}`;
 
   await expect(page).toHaveURL(/http:\/\/localhost:8083\/.+/);
-  await expect(
-    page.getByText('conversations'),
-  ).toBeVisible();
+  await expect(page.getByText('conversations')).toBeVisible();
 
   if (await page.getByLabel('Restart login').isVisible()) {
     await page.getByLabel('Restart login').click();