Extract render + factory from Conversation god-object (#6423)

backend/models/conversation.py

@@ -114,78 +114,6 @@ def __init__(self, **data):
         self.plugins_results = [PluginResult(plugin_id=app.app_id, content=app.content) for app in self.apps_results]
         self.processing_memory_id = self.processing_conversation_id
 
-    @staticmethod
-    def conversations_to_string(
-        conversations: List['Conversation'],
-        use_transcript: bool = False,
-        include_timestamps: bool = False,
-        people: List[Person] = None,
-        user_name: str = None,
-    ) -> str:
-        result = []
-        people_map = {p.id: p for p in people} if people else {}
-        for i, conversation in enumerate(conversations):
-            if isinstance(conversation, dict):
-                conversation = Conversation(**conversation)
-            formatted_date = conversation.created_at.astimezone(timezone.utc).strftime("%d %b %Y at %H:%M") + " UTC"
-            conversation_str = (
-                f"Conversation #{i + 1}\n"
-                f"{formatted_date} ({str(conversation.structured.category.value).capitalize()})\n"
-            )
-
-            # Add started_at and finished_at if available
-            if conversation.started_at:
-                formatted_started = (
-                    conversation.started_at.astimezone(timezone.utc).strftime("%d %b %Y at %H:%M") + " UTC"
-                )
-                conversation_str += f"Started: {formatted_started}\n"
-            if conversation.finished_at:
-                formatted_finished = (
-                    conversation.finished_at.astimezone(timezone.utc).strftime("%d %b %Y at %H:%M") + " UTC"
-                )
-                conversation_str += f"Finished: {formatted_finished}\n"
-
-            conversation_str += f"{str(conversation.structured.title).capitalize()}\n"
-
-            if (
-                conversation.apps_results
-                and len(conversation.apps_results) > 0
-                and conversation.apps_results[0].content.strip()
-            ):
-                conversation_str += f"{conversation.apps_results[0].content}\n"
-            else:
-                conversation_str += f"{str(conversation.structured.overview).capitalize()}\n"
-
-            # attendees
-            if people_map:
-                conv_person_ids = set(conversation.get_person_ids())
-                if conv_person_ids:
-                    attendees_names = [people_map[pid].name for pid in conv_person_ids if pid in people_map]
-                    if attendees_names:
-                        attendees = ", ".join(attendees_names)
-                        conversation_str += f"Attendees: {attendees}\n"
-
-            if conversation.structured.action_items:
-                conversation_str += "Action Items:\n"
-                for item in conversation.structured.action_items:
-                    conversation_str += f"- {item.description}\n"
-
-            if conversation.structured.events:
-                conversation_str += "Events:\n"
-                for event in conversation.structured.events:
-                    conversation_str += f"- {event.title} ({event.start} - {event.duration} minutes)\n"
-
-            if use_transcript:
-                conversation_str += f"\nTranscript:\n{conversation.get_transcript(include_timestamps=include_timestamps, people=people, user_name=user_name)}\n"
-                # photos
-                photo_descriptions = conversation.get_photos_descriptions(include_timestamps=include_timestamps)
-                if photo_descriptions != 'None':
-                    conversation_str += f"Photo Descriptions from a wearable camera:\n{photo_descriptions}\n"
-
-            result.append(conversation_str.strip())
-
-        return "\n\n---------------------\n\n".join(result).strip()
-
     def get_transcript(self, include_timestamps: bool, people: List[Person] = None, user_name: str = None) -> str:
         # Warn: missing transcript for workflow source, external integration source
         return TranscriptSegment.segments_as_string(

backend/routers/conversations.py

@@ -24,6 +24,9 @@
     UpdateActionItemDescriptionRequest,
     UpdateSegmentTextRequest,
 )
+from utils.conversations.factory import deserialize_conversation
+from utils.conversations.render import redact_conversations_for_list
+from utils.conversations.render import conversation_to_dict
 from models.conversation_enums import ConversationStatus, ConversationVisibility
 from models.conversation_photo import ConversationPhoto
 from models.geolocation import Geolocation
@@ -71,7 +74,7 @@ def process_in_progress_conversation(
         raise HTTPException(status_code=404, detail="Conversation in progress not found")
     redis_db.remove_in_progress_conversation_id(uid)
 
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
 
     # Inject calendar context if provided
     if request and request.calendar_meeting_context:
@@ -107,7 +110,7 @@ def reprocess_conversation(
     :return: The updated conversation after reprocessing.
     """
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     if not language_code:
         language_code = conversation.language or 'en'
 
@@ -147,15 +150,7 @@ def get_conversations(
         starred=starred,
     )
 
-    for conv in conversations:
-        if conv.get('is_locked', False):
-            if 'structured' in conv:
-                conv['structured']['action_items'] = []
-                conv['structured']['events'] = []
-            conv['apps_results'] = []
-            conv['plugins_results'] = []
-            conv['suggested_summarization_apps'] = []
-            conv['transcript_segments'] = []
+    redact_conversations_for_list(conversations)
     return conversations
 
 
@@ -253,7 +248,7 @@ def set_conversation_events_state(
     conversation_id: str, data: SetConversationEventsStateRequest, uid: str = Depends(auth.get_current_user_uid)
 ):
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     events = conversation.structured.events
     for i, event_idx in enumerate(data.events_idx):
         if event_idx >= len(events):
@@ -269,7 +264,7 @@ def set_action_item_status(
     data: SetConversationActionItemsStateRequest, conversation_id: str, uid=Depends(auth.get_current_user_uid)
 ):
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     action_items = conversation.structured.action_items
     for i, action_item_idx in enumerate(data.items_idx):
         if action_item_idx >= len(action_items):
@@ -330,7 +325,7 @@ def update_action_item_description(
     conversation_id: str, data: UpdateActionItemDescriptionRequest, uid=Depends(auth.get_current_user_uid)
 ):
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     action_items = conversation.structured.action_items
 
     found_item = False
@@ -361,7 +356,7 @@ def update_action_item_description(
 @router.delete("/v1/conversations/{conversation_id}/action-items", response_model=dict, tags=['conversations'])
 def delete_action_item(data: DeleteActionItemRequest, conversation_id: str, uid=Depends(auth.get_current_user_uid)):
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     action_items = conversation.structured.action_items
     updated_action_items = [item for item in action_items if not (item.description == data.description)]
     conversations_db.update_conversation_action_items(
@@ -414,7 +409,7 @@ def set_assignee_conversation_segment(
         f'set_assignee_conversation_segment {conversation_id} {segment_idx} {assign_type} {value} {use_for_speech_training} {uid}'
     )
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
 
     if value == 'null':
         value = None
@@ -483,7 +478,7 @@ def set_assignee_conversation_segment(
         f'set_assignee_conversation_segment {conversation_id} {speaker_id} {assign_type} {value} {use_for_speech_training} {uid}'
     )
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
 
     if value == 'null':
         value = None
@@ -542,7 +537,7 @@ def assign_segments_bulk(
     uid: str = Depends(auth.get_current_user_uid),
 ):
     conversation = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
 
     value = data.value
     if value == 'null':
@@ -619,7 +614,7 @@ def get_shared_conversation_by_id(conversation_id: str):
     visibility = conversation.get('visibility', ConversationVisibility.private)
     if not visibility or visibility == ConversationVisibility.private:
         raise HTTPException(status_code=404, detail="Conversation is private")
-    conversation = Conversation(**conversation)
+    conversation = deserialize_conversation(conversation)
     conversation.geolocation = None
 
     # Fetch people data for speaker names
@@ -630,7 +625,7 @@ def get_shared_conversation_by_id(conversation_id: str):
         people = [Person(**p) for p in people_data]
 
     # Return conversation with people data
-    response_dict = conversation.as_dict_cleaned_dates()
+    response_dict = conversation_to_dict(conversation)
     response_dict['people'] = [p.dict() for p in people]
     return response_dict
 
@@ -684,7 +679,7 @@ def get_conversation_suggested_apps(conversation_id: str, uid: str = Depends(aut
     from models.app import App
 
     conversation_data = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation_data)
+    conversation = deserialize_conversation(conversation_data)
 
     # Get suggested app models with full data (similar to /v1/apps endpoint)
     suggested_apps = []
@@ -719,7 +714,7 @@ def test_prompt(
     uid: str = Depends(auth.with_rate_limit(auth.get_current_user_uid, "test:prompt")),
 ):
     conversation_data = _get_valid_conversation_by_id(uid, conversation_id)
-    conversation = Conversation(**conversation_data)
+    conversation = deserialize_conversation(conversation_data)
 
     full_transcript = "\n".join([seg.text for seg in conversation.transcript_segments if seg.text])
 

backend/routers/developer.py

@@ -13,7 +13,6 @@
 import database.action_items as action_items_db
 import database.goals as goals_db
 import database.users as users_db
-import database.folders as folders_db
 from database.vector_db import upsert_memory_vectors_batch
 
 from models.memories import MemoryCategory, Memory, MemoryDB
@@ -24,6 +23,7 @@
     ExternalIntegrationConversationSource,
 )
 from models.geolocation import Geolocation
+from utils.conversations.render import populate_speaker_names, populate_folder_names
 from models.conversation import Conversation as ConversationModel
 from models.transcript_segment import TranscriptSegment
 from dependencies import (
@@ -732,52 +732,6 @@ class CreateConversationFromTranscriptRequest(BaseModel):
     geolocation: Optional[Geolocation] = Field(default=None, description="Geolocation where conversation occurred")
 
 
-def _add_speaker_names_to_segments(uid, conversations: list):
-    """Add speaker_name to transcript segments based on person_id mappings."""
-    user_profile = users_db.get_user_profile(uid)
-    user_name = user_profile.get('name') or 'User'
-
-    all_person_ids = set()
-    for conv in conversations:
-        for seg in conv.get('transcript_segments', []):
-            if seg.get('person_id'):
-                all_person_ids.add(seg['person_id'])
-
-    people_map = {}
-    if all_person_ids:
-        people_data = users_db.get_people_by_ids(uid, list(all_person_ids))
-        people_map = {p['id']: p['name'] for p in people_data}
-
-    for conv in conversations:
-        for seg in conv.get('transcript_segments', []):
-            if seg.get('is_user'):
-                seg['speaker_name'] = user_name
-            elif seg.get('person_id') and seg['person_id'] in people_map:
-                seg['speaker_name'] = people_map[seg['person_id']]
-            else:
-                seg['speaker_name'] = f"Speaker {seg.get('speaker_id', 0)}"
-
-
-def _add_folder_names_to_conversations(uid, conversations: list):
-    """Add folder_name to conversations based on folder_id mappings."""
-    folder_ids = set()
-    for conv in conversations:
-        if conv.get('folder_id'):
-            folder_ids.add(conv['folder_id'])
-
-    if not folder_ids:
-        for conv in conversations:
-            conv['folder_name'] = None
-        return
-
-    all_folders = folders_db.get_folders(uid)
-    folder_map = {f['id']: f['name'] for f in all_folders}
-
-    for conv in conversations:
-        folder_id = conv.get('folder_id')
-        conv['folder_name'] = folder_map.get(folder_id) if folder_id else None
-
-
 @router.get("/v1/dev/user/conversations", response_model=List[Conversation], tags=["developer"])
 def get_conversations(
     start_date: Optional[datetime] = None,
@@ -817,9 +771,9 @@ def get_conversations(
         for conv in unlocked_conversations:
             conv.pop('transcript_segments', None)
     else:
-        _add_speaker_names_to_segments(uid, unlocked_conversations)
+        populate_speaker_names(uid, unlocked_conversations)
 
-    _add_folder_names_to_conversations(uid, unlocked_conversations)
+    populate_folder_names(uid, unlocked_conversations)
 
     return unlocked_conversations
 
@@ -926,9 +880,9 @@ def get_conversation_endpoint(
     if not include_transcript:
         conversation.pop('transcript_segments', None)
     else:
-        _add_speaker_names_to_segments(uid, [conversation])
+        populate_speaker_names(uid, [conversation])
 
-    _add_folder_names_to_conversations(uid, [conversation])
+    populate_folder_names(uid, [conversation])
 
     return conversation
 
@@ -1122,7 +1076,7 @@ def update_conversation_endpoint(
 
     conversation = conversations_db.get_conversation(uid, conversation_id)
     if conversation:
-        _add_folder_names_to_conversations(uid, [conversation])
+        populate_folder_names(uid, [conversation])
     return conversation
 
 

backend/routers/folders.py

@@ -12,6 +12,7 @@
     ReorderFoldersRequest,
 )
 from models.conversation import Conversation
+from utils.conversations.render import redact_conversations_for_list
 from utils.other import endpoints as auth
 
 router = APIRouter()
@@ -111,15 +112,7 @@ def get_folder_conversations(
     conversations = folders_db.get_conversations_in_folder(
         uid, folder_id, limit=limit, offset=offset, include_discarded=include_discarded
     )
-    for conv in conversations:
-        if conv.get('is_locked', False):
-            if 'structured' in conv:
-                conv['structured']['action_items'] = []
-                conv['structured']['events'] = []
-            conv['apps_results'] = []
-            conv['plugins_results'] = []
-            conv['suggested_summarization_apps'] = []
-            conv['transcript_segments'] = []
+    redact_conversations_for_list(conversations)
     return conversations
 
 

backend/routers/integration.py

@@ -21,6 +21,7 @@
 from models.app import App
 from routers.conversations import process_conversation, trigger_external_integrations
 from utils.conversations.location import get_google_maps_location
+from utils.conversations.render import redact_conversation_for_integration
 from utils.conversations.memories import process_external_integration_memory
 from utils.conversations.search import search_conversations
 from utils.app_integrations import send_app_notification
@@ -347,16 +348,7 @@ async def get_conversations_via_integration(
     conversation_items = []
     for conv in conversations_data:
         try:
-            if conv.get('is_locked', False):
-                conv['structured']['title'] = ''
-                conv['structured']['overview'] = ''
-                conv['structured']['action_items'] = []
-                conv['structured']['events'] = []
-                conv['transcript_segments'] = []
-                conv['apps_results'] = []
-                conv['plugins_results'] = []
-                conv['suggested_summarization_apps'] = []
-
+            redact_conversation_for_integration(conv)
             item = integration_models.ConversationItem.parse_obj(conv)
 
             # Limit transcript segments
@@ -480,16 +472,7 @@ async def search_conversations_via_integration(
     conversation_items = []
     for conv in full_conversations:
         try:
-            if conv.get('is_locked', False):
-                conv['structured']['title'] = ''
-                conv['structured']['overview'] = ''
-                conv['structured']['action_items'] = []
-                conv['structured']['events'] = []
-                conv['transcript_segments'] = []
-                conv['apps_results'] = []
-                conv['plugins_results'] = []
-                conv['suggested_summarization_apps'] = []
-
+            redact_conversation_for_integration(conv)
             item = integration_models.ConversationItem.parse_obj(conv)
 
             # Limit transcript segments