refactor(backend): migrate print() to structured logging (#5067)

Author: beastoin

backend/database/action_items.py

@@ -4,6 +4,9 @@
 from google.cloud.firestore_v1 import FieldFilter
 
 from ._client import db
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 # Collection name
@@ -465,4 +468,4 @@ def unlock_all_action_items(uid: str):
             count = 0
     if count > 0:
         batch.commit()
-    print(f"Unlocked all action items for user {uid}")
+    logger.info(f"Unlocked all action items for user {uid}")

backend/database/apps.py

@@ -10,6 +10,9 @@
 from models.app import UsageHistoryType
 from ._client import db
 from .redis_db import get_app_reviews
+import logging
+
+logger = logging.getLogger(__name__)
 
 # *****************************
 # ********** CRUD *************
@@ -23,7 +26,7 @@
 def migrate_reviews_from_redis_to_firestore():
     apps_ref = db.collection(apps_collection).stream()
     for app in apps_ref:
-        print('migrating reviews for app:', app.id)
+        logger.info(f'migrating reviews for app: {app.id}')
         app_id = app.id
         reviews = get_app_reviews(app_id)
         for uid, review in reviews.items():

backend/database/auth.py

@@ -1,13 +1,16 @@
 from firebase_admin import auth
 
 from database.redis_db import cache_user_name, get_cached_user_name
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 def get_user_from_uid(uid: str):
     try:
         user = auth.get_user(uid) if uid else None
     except Exception as e:
-        print(e)
+        logger.error(e)
         user = None
     if not user:
         return None

backend/database/cache.py

@@ -29,13 +29,9 @@ def _ensure_initialized():
 
     # Register callbacks: when invalidation message received, clear memory cache
     _pubsub_manager.register_callback(
-        'get_public_approved_apps_data*',
-        lambda keys: [_memory_cache.delete(k) for k in keys]
-    )
-    _pubsub_manager.register_callback(
-        'get_popular_apps_data',
-        lambda keys: [_memory_cache.delete(k) for k in keys]
+        'get_public_approved_apps_data*', lambda keys: [_memory_cache.delete(k) for k in keys]
     )
+    _pubsub_manager.register_callback('get_popular_apps_data', lambda keys: [_memory_cache.delete(k) for k in keys])
 
     # Start pub/sub subscription
     _pubsub_manager.start()

backend/database/cache_manager.py

@@ -20,6 +20,7 @@
 @dataclass
 class CacheEntry:
     """Represents a cache entry with metadata."""
+
     data: Any
     timestamp: float
     size_bytes: int
@@ -149,12 +150,7 @@ def set(self, key: str, data: Any, ttl: int = 30):
             self._evict_if_needed(size_bytes)
 
             # Add new entry
-            entry = CacheEntry(
-                data=data,
-                timestamp=time.time(),
-                size_bytes=size_bytes,
-                ttl=ttl
-            )
+            entry = CacheEntry(data=data, timestamp=time.time(), size_bytes=size_bytes, ttl=ttl)
             self.cache[key] = entry
             self.current_size += size_bytes
 
@@ -195,8 +191,7 @@ def _evict_if_needed(self, required_bytes: int):
         Args:
             required_bytes: Bytes needed for new entry
         """
-        while (self.current_size + required_bytes > self.max_memory_bytes
-               and len(self.cache) > 0):
+        while self.current_size + required_bytes > self.max_memory_bytes and len(self.cache) > 0:
             # Remove oldest (first item in OrderedDict)
             key, entry = self.cache.popitem(last=False)
             self.current_size -= entry.size_bytes
@@ -237,5 +232,5 @@ def get_stats(self) -> dict:
                 'hits': self.hits,
                 'misses': self.misses,
                 'hit_rate': round(hit_rate, 2),
-                'evictions': self.evictions
+                'evictions': self.evictions,
             }

backend/database/chat.py

@@ -12,6 +12,9 @@
 from utils.other.endpoints import timeit
 from ._client import db
 from .helpers import set_data_protection_level, prepare_for_write, prepare_for_read
+import logging
+
+logger = logging.getLogger(__name__)
 
 # *********************************
 # ******* ENCRYPTION HELPERS ******
@@ -176,7 +179,7 @@ def get_messages(
     chat_session_id: Optional[str] = None,
     # include_plugin_id_filter: bool = True,
 ):
-    print('get_messages', uid, limit, offset, app_id, include_conversations)
+    logger.info(f'get_messages {uid} {limit} {offset} {app_id} {include_conversations}')
     user_ref = db.collection('users').document(uid)
     messages_ref = user_ref.collection('messages')
     # if include_plugin_id_filter:
@@ -261,7 +264,7 @@ def report_message(uid: str, msg_doc_id: str):
         message_ref.update({'reported': True})
         return {"message": "Message reported"}
     except Exception as e:
-        print("Update failed:", e)
+        logger.error(f"Update failed: {e}")
         return {"message": f"Update failed: {e}"}
 
 
@@ -278,15 +281,15 @@ def update_message_rating(uid: str, message_id: str, rating: int | None):
     message_ref = user_ref.collection('messages').where('id', '==', message_id).limit(1).stream()
     message_doc = next(message_ref, None)
     if not message_doc:
-        print(f"⚠️ Message {message_id} not found for user {uid}")
+        logger.warning(f"⚠️ Message {message_id} not found for user {uid}")
         return False
 
     try:
         user_ref.collection('messages').document(message_doc.id).update({'rating': rating})
-        print(f"✅ Updated message {message_id} rating to {rating}")
+        logger.info(f"✅ Updated message {message_id} rating to {rating}")
         return True
     except Exception as e:
-        print(f"❌ Failed to update message rating: {e}")
+        logger.error(f"❌ Failed to update message rating: {e}")
         return False
 
 
@@ -297,32 +300,32 @@ def batch_delete_messages(
     messages_ref = messages_ref.where(filter=FieldFilter('plugin_id', '==', app_id))
     if chat_session_id:
         messages_ref = messages_ref.where(filter=FieldFilter('chat_session_id', '==', chat_session_id))
-    print('batch_delete_messages', app_id)
+    logger.info(f'batch_delete_messages {app_id}')
 
     while True:
         docs_stream = messages_ref.limit(batch_size).stream()
         docs_list = list(docs_stream)
 
         if not docs_list:
-            print("No more messages to delete")
+            logger.info("No more messages to delete")
             break
 
         batch = db.batch()
         for doc in docs_list:
             batch.delete(doc.reference)
         batch.commit()
 
-        print(f'Deleted {len(docs_list)} messages')
+        logger.info(f'Deleted {len(docs_list)} messages')
 
         if len(docs_list) < batch_size:
-            print("Processed all messages")
+            logger.info("Processed all messages")
             break
 
 
 def clear_chat(uid: str, app_id: Optional[str] = None, chat_session_id: Optional[str] = None):
     try:
         user_ref = db.collection('users').document(uid)
-        print(f"Deleting messages for user: {uid}")
+        logger.info(f"Deleting messages for user: {uid}")
         if not user_ref.get().exists:
             return {"message": "User not found"}
         batch_delete_messages(user_ref, app_id=app_id, chat_session_id=chat_session_id)
@@ -474,7 +477,7 @@ def update_chat_session_openai_ids(uid: str, chat_session_id: str, thread_id: st
 
     if update_data:
         session_ref.update(update_data)
-        print(f"Updated session {chat_session_id} with thread {thread_id} and assistant {assistant_id}")
+        logger.info(f"Updated session {chat_session_id} with thread {thread_id} and assistant {assistant_id}")
 
 
 # **************************************
@@ -512,7 +515,7 @@ def migrate_chats_level_batch(uid: str, message_doc_ids: List[str], target_level
 
     for doc_snapshot in doc_snapshots:
         if not doc_snapshot.exists:
-            print(f"Message {doc_snapshot.id} not found, skipping.")
+            logger.warning(f"Message {doc_snapshot.id} not found, skipping.")
             continue
 
         message_data = doc_snapshot.to_dict()

backend/database/conversations.py

@@ -22,6 +22,9 @@
 from ._client import db
 from .helpers import set_data_protection_level, prepare_for_write, prepare_for_read, with_photos
 from utils.other.storage import list_audio_chunks
+import logging
+
+logger = logging.getLogger(__name__)
 
 conversations_collection = 'conversations'
 
@@ -58,7 +61,7 @@ def _decrypt_conversation_data(conversation_data: Dict[str, Any], uid: str) -> D
             else:
                 data['transcript_segments'] = json.loads(decrypted_payload)
         except (json.JSONDecodeError, TypeError, zlib.error, ValueError) as e:
-            print(e, uid)
+            logger.error(f"{e} {uid}")
             data['transcript_segments'] = []
     # backward compatibility, will be removed soon
     elif isinstance(data['transcript_segments'], bytes):
@@ -68,7 +71,7 @@ def _decrypt_conversation_data(conversation_data: Dict[str, Any], uid: str) -> D
                 decompressed_json = zlib.decompress(compressed_bytes).decode('utf-8')
                 data['transcript_segments'] = json.loads(decompressed_json)
         except (json.JSONDecodeError, TypeError, zlib.error, ValueError) as e:
-            print(e, uid)
+            logger.error(f"{e} {uid}")
             data['transcript_segments'] = []
 
     return data
@@ -106,7 +109,7 @@ def _prepare_conversation_for_read(conversation_data: Optional[Dict[str, Any]],
                 decompressed_json = zlib.decompress(data['transcript_segments']).decode('utf-8')
                 data['transcript_segments'] = json.loads(decompressed_json)
             except (json.JSONDecodeError, TypeError, zlib.error) as e:
-                print(e)
+                logger.error(e)
                 pass
 
     return data
@@ -600,7 +603,7 @@ def migrate_conversations_level_batch(uid: str, conversation_ids: List[str], tar
 
     for doc_snapshot in doc_snapshots:
         if not doc_snapshot.exists:
-            print(f"Conversation {doc_snapshot.id} not found, skipping.")
+            logger.warning(f"Conversation {doc_snapshot.id} not found, skipping.")
             continue
 
         conversation_data = doc_snapshot.to_dict()
@@ -908,7 +911,7 @@ def unlock_all_conversations(uid: str):
             count = 0
     if count > 0:
         batch.commit()
-    print(f"Unlocked all conversations for user {uid}")
+    logger.info(f"Unlocked all conversations for user {uid}")
 
 
 def get_public_conversations(data: List[Tuple[str, str]]):
@@ -1050,13 +1053,13 @@ def get_closest_conversation_to_timestamps(uid: str, start_timestamp: int, end_t
     )
 
     conversations = [doc.to_dict() for doc in query.stream()]
-    print('get_closest_conversation_to_timestamps len(conversations)', len(conversations))
+    logger.info(f'get_closest_conversation_to_timestamps len(conversations) {len(conversations)}')
     if not conversations:
         return None
 
-    print('get_closest_conversation_to_timestamps found:')
+    logger.info('get_closest_conversation_to_timestamps found:')
     for conversation in conversations:
-        print('-', conversation['id'], conversation['started_at'], conversation['finished_at'])
+        logger.info(f'- {conversation['id']} {conversation['started_at']} {conversation['finished_at']}')
 
     # get the conversation that has the closest start timestamp or end timestamp
     closest_conversation = None
@@ -1070,7 +1073,7 @@ def get_closest_conversation_to_timestamps(uid: str, start_timestamp: int, end_t
             min_diff = min(diff1, diff2)
             closest_conversation = conversation
 
-    print('get_closest_conversation_to_timestamps closest_conversation:', closest_conversation['id'])
+    logger.info(f'get_closest_conversation_to_timestamps closest_conversation: {closest_conversation['id']}')
     return closest_conversation
 
 

backend/database/folders.py

@@ -8,7 +8,6 @@
 from ._client import db
 from models.folder import Folder
 
-
 # System folders that are created for new users
 SYSTEM_FOLDERS = [
     {

backend/database/helpers.py

@@ -6,6 +6,9 @@
 
 from database import users as users_db, redis_db
 from ._client import db
+import logging
+
+logger = logging.getLogger(__name__)
 
 
 def set_data_protection_level(data_arg_name: str):
@@ -66,7 +69,7 @@ def wrapper(*args, **kwargs):
                     level = user_profile.get('data_protection_level', 'enhanced') if user_profile else 'enhanced'
                     redis_db.set_user_data_protection_level(uid, level)
                 except Exception as e:
-                    print(f"Failed to get user profile for {uid}: {e}")
+                    logger.error(f"Failed to get user profile for {uid}: {e}")
                     level = 'enhanced'
 
             if not level: