Add 372 fix 489

application/single_app/config.py

@@ -88,7 +88,7 @@
 EXECUTOR_TYPE = 'thread'
 EXECUTOR_MAX_WORKERS = 30
 SESSION_TYPE = 'filesystem'
-VERSION = "0.229.063"
+VERSION = "0.229.098"
 
 
 SECRET_KEY = os.getenv('SECRET_KEY', 'dev-secret-key-change-in-production')
@@ -121,9 +121,9 @@
 CLIENTS_LOCK = threading.Lock()
 
 ALLOWED_EXTENSIONS = {
-    'txt', 'pdf', 'docx', 'xlsx', 'xls', 'csv', 'pptx', 'html', 'jpg', 'jpeg', 'png', 'bmp', 'tiff', 'tif', 'heif', 'md', 'json', 
+    'txt', 'pdf', 'doc', 'docm', 'docx', 'xlsx', 'xls', 'xlsm','csv', 'pptx', 'html', 'jpg', 'jpeg', 'png', 'bmp', 'tiff', 'tif', 'heif', 'md', 'json', 
     'mp4', 'mov', 'avi', 'mkv', 'flv', 'mxf', 'gxf', 'ts', 'ps', '3gp', '3gpp', 'mpg', 'wmv', 'asf', 'm4a', 'm4v', 'isma', 'ismv', 
-    'dvr-ms', 'wav'
+    'dvr-ms', 'wav', 'xml', 'yaml', 'yml', 'log'
 }
 ALLOWED_EXTENSIONS_IMG = {'png', 'jpg', 'jpeg'}
 MAX_CONTENT_LENGTH = 5000 * 1024 * 1024  # 5000 MB AKA 5 GB

application/single_app/functions_authentication.py

@@ -245,15 +245,34 @@ def get_valid_access_token_for_plugins(scopes=None):
 
 def get_video_indexer_account_token(settings, video_id=None):
     """
-    For ARM-based VideoIndexer accounts:
+    Get Video Indexer access token using managed identity authentication.
+
+    This function authenticates with Azure Video Indexer using the App Service's
+    managed identity. The managed identity must have Contributor role on the
+    Video Indexer resource.
+
+    Authentication flow:
+    1. Acquire ARM access token using DefaultAzureCredential (managed identity)
+    2. Call ARM generateAccessToken API to get Video Indexer access token
+    3. Use Video Indexer access token for all API operations
+    """
+    from functions_debug import debug_print
+
+    debug_print(f"[VIDEO INDEXER AUTH] Starting token acquisition using managed identity for video_id: {video_id}")
+    debug_print(f"[VIDEO INDEXER AUTH] Azure environment: {AZURE_ENVIRONMENT}")
+
+    return get_video_indexer_managed_identity_token(settings, video_id)
+
+def get_video_indexer_managed_identity_token(settings, video_id=None):
+    """
+    For ARM-based VideoIndexer accounts using managed identity:
     1) Acquire an ARM token with DefaultAzureCredential
     2) POST to the ARM generateAccessToken endpoint
     3) Return the account-level accessToken
     """
     from functions_debug import debug_print
 
-    debug_print(f"[VIDEO INDEXER AUTH] Starting token acquisition for video_id: {video_id}")
-    debug_print(f"[VIDEO INDEXER AUTH] Azure environment: {AZURE_ENVIRONMENT}")
+    debug_print(f"[VIDEO INDEXER AUTH] Using managed identity authentication")
 
     # 1) ARM token
     if AZURE_ENVIRONMENT == "usgovernment":

application/single_app/functions_content.py

@@ -172,7 +172,7 @@ def extract_table_file(file_path, file_ext):
     try:
         if file_ext == '.csv':
             df = pandas.read_csv(file_path)
-        elif file_ext in ['.xls', '.xlsx']:
+        elif file_ext in ['.xls', '.xlsx', '.xlsm']:
             df = pandas.read_excel(file_path)
         else:
             raise ValueError("Unsupported file extension for table extraction.")

application/single_app/functions_search.py

@@ -46,18 +46,22 @@ def hybrid_search(query, user_id, document_id=None, top_n=12, doc_scope="all", a
                 select=["id", "chunk_text", "chunk_id", "file_name", "user_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
             )
 
-            group_results = search_client_group.search(
-                search_text=query,
-                vector_queries=[vector_query],
-                filter=(
-                    f"(group_id eq '{active_group_id}' or shared_group_ids/any(g: g eq '{active_group_id},approved')) and document_id eq '{document_id}'"
-                ),
-                query_type="semantic",
-                semantic_configuration_name="nexus-group-index-semantic-configuration",
-                query_caption="extractive",
-                query_answer="extractive",
-                select=["id", "chunk_text", "chunk_id", "file_name", "group_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
-            )
+            # Only search group index if active_group_id is provided
+            if active_group_id:
+                group_results = search_client_group.search(
+                    search_text=query,
+                    vector_queries=[vector_query],
+                    filter=(
+                        f"(group_id eq '{active_group_id}' or shared_group_ids/any(g: g eq '{active_group_id},approved')) and document_id eq '{document_id}'"
+                    ),
+                    query_type="semantic",
+                    semantic_configuration_name="nexus-group-index-semantic-configuration",
+                    query_caption="extractive",
+                    query_answer="extractive",
+                    select=["id", "chunk_text", "chunk_id", "file_name", "group_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
+                )
+            else:
+                group_results = []
 
             # Get visible public workspace IDs from user settings
             visible_public_workspace_ids = get_user_visible_public_workspace_ids_from_settings(user_id)
@@ -97,18 +101,22 @@ def hybrid_search(query, user_id, document_id=None, top_n=12, doc_scope="all", a
                 select=["id", "chunk_text", "chunk_id", "file_name", "user_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
             )
 
-            group_results = search_client_group.search(
-                search_text=query,
-                vector_queries=[vector_query],
-                filter=(
-                    f"(group_id eq '{active_group_id}' or shared_group_ids/any(g: g eq '{active_group_id},approved'))"
-                ),
-                query_type="semantic",
-                semantic_configuration_name="nexus-group-index-semantic-configuration",
-                query_caption="extractive",
-                query_answer="extractive",
-                select=["id", "chunk_text", "chunk_id", "file_name", "group_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
-            )
+            # Only search group index if active_group_id is provided
+            if active_group_id:
+                group_results = search_client_group.search(
+                    search_text=query,
+                    vector_queries=[vector_query],
+                    filter=(
+                        f"(group_id eq '{active_group_id}' or shared_group_ids/any(g: g eq '{active_group_id},approved'))"
+                    ),
+                    query_type="semantic",
+                    semantic_configuration_name="nexus-group-index-semantic-configuration",
+                    query_caption="extractive",
+                    query_answer="extractive",
+                    select=["id", "chunk_text", "chunk_id", "file_name", "group_id", "version", "chunk_sequence", "upload_date", "document_classification", "page_number", "author", "chunk_keywords", "title", "chunk_summary"]
+                )
+            else:
+                group_results = []
 
             # Get visible public workspace IDs from user settings
             visible_public_workspace_ids = get_user_visible_public_workspace_ids_from_settings(user_id)

application/single_app/functions_settings.py

@@ -134,6 +134,10 @@ def get_settings():
         'number_of_historical_messages_to_summarize': 10,
         'enable_summarize_content_history_beyond_conversation_history_limit': False,
 
+        # Multi-Modal Vision Analysis
+        'enable_multimodal_vision': False,
+        'multimodal_vision_model': '',
+
         # Document Classification
         'enable_document_classification': False,
         'document_classification_categories': [
@@ -215,11 +219,10 @@ def get_settings():
         'video_indexer_endpoint': video_indexer_endpoint,
         'video_indexer_location': '',
         'video_indexer_account_id': '',
-        'video_indexer_api_key': '',
         'video_indexer_resource_group': '',
         'video_indexer_subscription_id': '',
         'video_indexer_account_name': '',
-        'video_indexer_arm_api_version': '2021-11-10-preview',
+        'video_indexer_arm_api_version': '2024-01-01',
         'video_index_timeout': 600,
 
         # Audio file settings with Azure speech service