refactor: replace standard logging with a centralized loguru configuration and interception handler

asr/batch_workers.py

@@ -15,14 +15,12 @@ Architecture (mirrors Raij/src/smart_search/batch_workers.py, audio workers only
 import asyncio
 import time
 import uuid
-import logging
+from loguru import logger
 from typing import Any
 
 from .constants import ASR_BATCH_MAX, ASR_BATCH_WINDOW_S, WARMUP_INTERVAL_S, PARAKEET_WARMUP_EVERY
 from .schemas import AudioJob
 
-logger = logging.getLogger(__name__)
-
 
 # ═══════════════════════ Job Store ════════════════════════
 

asr/models.py

@@ -16,9 +16,7 @@ Strategy (ported from Raij/src/models.py, hotword biasing removed):
 
 import os
 import threading
-import logging
-
-logger = logging.getLogger(__name__)
+from loguru import logger
 
 # Force PyTorch path, no TensorFlow
 os.environ.setdefault("USE_TF", "0")

asr/routes.py

@@ -2,7 +2,7 @@ import asyncio
 import os
 import tempfile
 import uuid
-import logging
+from loguru import logger
 from fastapi import APIRouter, UploadFile, File, Form, HTTPException, Depends
 from fastapi.responses import JSONResponse
 
@@ -10,8 +10,6 @@ from src.dependencies import get_current_user_id
 from src.asr.schemas import AudioJob
 from src.asr.batch_workers import audio_en_queue, audio_ar_queue, job_store
 
-logger = logging.getLogger(__name__)
-
 router = APIRouter(prefix="/api/asr", tags=["ASR"])
 
 # Allowed audio MIME types from browsers (MediaRecorder output)

auth/refresh_token_store.py

@@ -34,7 +34,7 @@ Table DDL (run once in the Supabase SQL editor):
     CREATE INDEX ON refresh_tokens(user_id);
 """
 
-import logging
+from loguru import logger
 from datetime import datetime, timedelta, timezone
 from typing import Optional
 
@@ -47,8 +47,6 @@ from src.auth.constants import (
 )
 from src.redis_client import get_redis
 
-logger = logging.getLogger(__name__)
-
 
 def _rt_key(token_hash: str) -> str:
     return f"{REFRESH_TOKEN_KEY_PREFIX}{token_hash}"

auth/routes.py

@@ -1,5 +1,5 @@
 import uuid
-import logging
+from loguru import logger
 import cloudinary
 import cloudinary.uploader
 from datetime import timezone, datetime
@@ -7,8 +7,6 @@ from fastapi import APIRouter, HTTPException, UploadFile, File, Response, Reques
 from fastapi import Depends
 from typing import Optional
 
-logger = logging.getLogger(__name__)
-
 from src.config import settings
 from src.database import get_auth_supabase, get_supabase
 from src.store import create_user, get_user_by_email, delete_user_data, update_user_profile, get_user_by_id

database.py

@@ -1,10 +1,8 @@
-import logging
 from typing import Optional
+from loguru import logger
 from supabase import Client, create_client
 from src.config import settings
 
-logger = logging.getLogger(__name__)
-
 # Singletons — created once, reused on every request
 _supabase_client: Optional[Client] = None
 _auth_supabase_client: Optional[Client] = None

logger.py

@@ -0,0 +1,83 @@
+import os
+import sys
+import logging
+from loguru import logger
+
+class InterceptHandler(logging.Handler):
+    """
+    Standard logging handler to intercept standard logging messages 
+    and redirect them to Loguru.
+    """
+    def emit(self, record: logging.LogRecord) -> None:
+        # Get corresponding Loguru level if it exists
+        try:
+            level: str | int = logger.level(record.levelname).name
+        except ValueError:
+            level = record.levelno
+
+        # Find caller from where originated the logged message
+        frame = logging.currentframe()
+        depth = 2
+        while frame and frame.f_code.co_filename == logging.__file__:
+            frame = frame.f_back
+            depth += 1
+
+        logger.opt(depth=depth, exception=record.exc_info).log(
+            level, record.getMessage()
+        )
+
+def setup_logging():
+    """
+    Configures Loguru for console and file logging, and intercepts
+    standard library logging (uvicorn, httpx, etc.).
+    """
+    os.makedirs("logs", exist_ok=True)
+
+    # Remove default handlers
+    logger.remove()
+
+    # Log format specifications
+    console_format = (
+        "<green>{time:YYYY-MM-DD HH:mm:ss}</green> | "
+        "<level>{level: <8}</level> | "
+        "<cyan>{name}</cyan>:<cyan>{function}</cyan>:<cyan>{line}</cyan> - "
+        "<level>{message}</level>"
+    )
+
+    file_format = "{time:YYYY-MM-DD HH:mm:ss} | {level: <8} | {name}:{function}:{line} - {message}"
+
+    # Console Handler (colored)
+    logger.add(
+        sys.stdout,
+        level="INFO",
+        format=console_format,
+        colorize=True,
+        enqueue=True,
+        backtrace=True,
+        diagnose=True,
+    )
+
+    # Rotating File Handler (5 MB per file, max 3 backups)
+    logger.add(
+        "logs/app.log",
+        level="INFO",
+        format=file_format,
+        rotation="5 MB",
+        retention=3,
+        encoding="utf-8",
+        enqueue=True,
+    )
+
+    # Intercept standard Python logging
+    logging.basicConfig(handlers=[InterceptHandler()], level=0, force=True)
+
+    # Ensure uvicorn, fastapi, watchfiles, and httpx use the intercept handler
+    for log_name in ("uvicorn", "uvicorn.access", "uvicorn.error", "fastapi", "watchfiles", "httpx"):
+        mod_logger = logging.getLogger(log_name)
+        mod_logger.handlers = [InterceptHandler()]
+        if "watchfiles" in log_name:
+            mod_logger.setLevel(logging.WARNING)
+
+    return logger
+
+__all__ = ["logger", "setup_logging", "InterceptHandler"]

main.py

@@ -1,5 +1,3 @@
-import logging
-import logging.handlers
 import os
 import sys
 
@@ -8,6 +6,9 @@ parent_dir = os.path.dirname(current_dir)
 if parent_dir not in sys.path:
     sys.path.insert(0, parent_dir)
 
+from src.logger import setup_logging, logger
+setup_logging()
+
 from contextlib import asynccontextmanager
 from fastapi import FastAPI, Depends
 from fastapi.middleware.cors import CORSMiddleware
@@ -21,38 +22,7 @@ from src.auth.routes import router as auth_router
 from src.asr.routes import router as asr_router
 from src.store import get_usage
 from src.dependencies import get_current_user_id
-from src.config import settings 
-
-# ── Logging Setup ──────────────────────────────────────
-os.makedirs("logs", exist_ok=True)
-
-log_formatter = logging.Formatter(
-    "%(asctime)s | %(levelname)-8s | %(name)s | %(message)s",
-    datefmt="%Y-%m-%d %H:%M:%S",
-)
-
-# File handler — rotates at 5MB, keeps 3 backups
-file_handler = logging.handlers.RotatingFileHandler(
-    "logs/app.log", maxBytes=5 * 1024 * 1024, backupCount=3, encoding="utf-8"
-)
-file_handler.setFormatter(log_formatter)
-
-# Console handler
-console_handler = logging.StreamHandler()
-console_handler.setFormatter(log_formatter)
-
-# Apply to root logger so all src.* modules inherit it
-root_logger = logging.getLogger()
-root_logger.setLevel(logging.INFO)
-root_logger.addHandler(file_handler)
-root_logger.addHandler(console_handler)
-
-# Suppress noisy watchfiles reload logs
-logging.getLogger("watchfiles").setLevel(logging.WARNING)
-logging.getLogger("watchfiles.main").setLevel(logging.WARNING)
-
-logger = logging.getLogger(__name__)
-
+from src.config import settings
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
@@ -79,24 +49,24 @@ async def lifespan(app: FastAPI):
     except Exception as e:
         logger.warning(f"Embedder failed to load: {e}")
 
-    # # Eagerly load ASR models so warmup runs at startup, not on first request
-    # try:
-    #     from src.asr.models import get_audio_model_en
-    #     get_audio_model_en()
-    # except Exception as e:
-    #     logger.warning(f"English ASR model failed to load: {e}")
+    # Eagerly load ASR models so warmup runs at startup, not on first request
+    try:
+        from src.asr.models import get_audio_model_en
+        get_audio_model_en()
+    except Exception as e:
+        logger.warning(f"English ASR model failed to load: {e}")
 
-    # try:
-    #     from src.asr.models import get_audio_model_ar
-    #     get_audio_model_ar()
-    # except Exception as e:
-    #     logger.warning(f"Arabic ASR model failed to load: {e}")
+    try:
+        from src.asr.models import get_audio_model_ar
+        get_audio_model_ar()
+    except Exception as e:
+        logger.warning(f"Arabic ASR model failed to load: {e}")
 
-    # from src.rag.batch_workers import start_workers
-    # start_workers()
+    from src.rag.batch_workers import start_workers
+    start_workers()
 
-    # from src.asr.batch_workers import start_asr_workers
-    # start_asr_workers()
+    from src.asr.batch_workers import start_asr_workers
+    start_asr_workers()
 
     yield
 
@@ -125,12 +95,28 @@ _raw_origins = settings.cors_allowed_origins if settings.cors_allowed_origins el
 _cors_origins = [o.strip() for o in _raw_origins if o.strip() and o.strip() != "*"] or _DEFAULT_ORIGINS
 
 @app.middleware("http")
-async def normalize_path(request, call_next):
+async def log_request_timing(request, call_next):
+    import time
+    start = time.perf_counter()
     # Fix double slashes in paths (e.g., //api/usage -> /api/usage)
     path = request.scope.get("path")
     if path and "//" in path:
         request.scope["path"] = path.replace("//", "/")
-    return await call_next(request)
+    response = await call_next(request)
+    duration_sec = time.perf_counter() - start
+
+    status = response.status_code
+    if 200 <= status < 300:
+        status_str = f"<green>{status}</green>"
+    elif 300 <= status < 400:
+        status_str = f"<cyan>{status}</cyan>"
+    elif 400 <= status < 500:
+        status_str = f"<red>{status}</red>"
+    else:
+        status_str = f"<bold><red>{status}</red></bold>"
+
+    logger.opt(colors=True).info(f"{request.method} {request.url.path} - {status_str} ({duration_sec:.2f}s)")
+    return response
 
 app.add_middleware(ProxyHeadersMiddleware, trusted_hosts="*")
 

materials/routes.py

@@ -1,6 +1,6 @@
 import time
 import asyncio
-import logging
+from loguru import logger
 import validators
 from fastapi import APIRouter, UploadFile, File, HTTPException, Depends, BackgroundTasks, Header, Request
 from postgrest.exceptions import APIError
@@ -13,8 +13,6 @@ from src.database import get_supabase, get_auth_supabase
 from .constants import ALLOWED_TYPES, MAX_SIZE_MB, MAX_SIZE_BYTES
 from .schemas import URLInput, RenameMaterialRequest, BulkDeleteRequest, TopicRequest, SearchRequest
 
-logger = logging.getLogger(__name__)
-
 router = APIRouter(prefix="/api/materials", tags=["Materials"])
 
 

materials/validator.py

@@ -1,10 +1,8 @@
 import re
-import logging
+from loguru import logger
 from src.rag.rag import get_llm
 from src.config import settings
 
-logger = logging.getLogger(__name__)
-
 # Load English NSFW word list from config settings (kept out of committed code)
 LOCAL_NSFW_WORDS = set(settings.local_nsfw_words)
 

quiz_generator/quiz.py

@@ -1,7 +1,7 @@
 import json
 import random
 import re
-import logging
+from loguru import logger
 from typing import Optional
 from langchain.agents import create_tool_calling_agent, AgentExecutor
 from langchain_core.tools import create_retriever_tool
@@ -18,8 +18,6 @@ from .constants import (
     ARXIV_DOC_CONTENT_CHARS_MAX,
 )
 
-logger = logging.getLogger(__name__)
-
 
 def _quiz_prompt():
     return QUIZ_PROMPT_TEMPLATE

quiz_generator/routes.py

@@ -1,5 +1,5 @@
 import asyncio
-import logging
+from loguru import logger
 from fastapi import APIRouter, HTTPException, Depends
 from typing import Optional
 from src.quiz_generator.quiz import smart_quiz_generator
@@ -14,8 +14,6 @@ from .constants import (
     MAX_TF_COUNT,
 )
 
-logger = logging.getLogger(__name__)
-
 router = APIRouter(prefix="/api/quiz", tags=["Quiz"])
 
 

rag/batch_workers.py

@@ -12,14 +12,12 @@ Architecture:
 import asyncio
 import time
 import uuid
-import logging
+from loguru import logger
 from typing import Any
 
 from .constants import BATCH_MAX_SIZE, BATCH_WINDOW_S, WARMUP_INTERVAL_S
 from .schemas import EmbeddingJob
 
-logger = logging.getLogger(__name__)
-
 
 # ═══════════════════════ Job Store ════════════════════════
 

rag/rag.py

@@ -1,7 +1,7 @@
 import os
 import asyncio
 import uuid
-import logging
+from loguru import logger
 from functools import lru_cache
 from typing import Optional
 from langchain_community.tools import DuckDuckGoSearchResults
@@ -31,8 +31,6 @@ from .constants import (
 )
 from .schemas import EmbeddingJob
 
-logger = logging.getLogger(__name__)
-
 
 # ── Embeddings ─────────────────────────────────────────
 

rag/routes.py

@@ -1,6 +1,6 @@
 import time
 import asyncio
-import logging
+from loguru import logger
 from fastapi import APIRouter, HTTPException, Depends
 from typing import Optional, Any
 
@@ -20,8 +20,6 @@ from src.store import (
 from src.summary_generator.summary import clean_summary
 from .schemas import TutorQuery, TutorResponse, SessionRequest, RenameSessionRequest, ExtractTitleRequest, SaveChatRequest
 
-logger = logging.getLogger(__name__)
-
 router = APIRouter(prefix="/api/tutor", tags=["Tutor"])
 
 

redis_client.py

@@ -8,13 +8,11 @@ do a simple `r = get_redis(); if r is None: <fallback>` without crashing.
 Connection is created once and reused for the lifetime of the process.
 """
 
-import logging
 from typing import Optional
+from loguru import logger
 
 import redis
 
-logger = logging.getLogger(__name__)
-
 _redis_client: Optional[redis.Redis] = None
 _redis_available: bool = False
 
@@ -54,6 +52,6 @@ def get_redis() -> Optional[redis.Redis]:
         logger.info("Redis connected successfully.")
         return _redis_client
     except Exception as e:
-        logger.warning("Redis connection failed (%s) — falling back to Supabase/in-memory.", e)
+        logger.warning(f"Redis connection failed ({e}) — falling back to Supabase/in-memory.")
         _redis_available = False
         return None

store.py

@@ -3,7 +3,7 @@ import time
 import json
 from httpx import RemoteProtocolError
 from typing import Optional
-import logging
+from loguru import logger
 from datetime import datetime, timezone, date, timedelta
 from langchain.memory import ConversationBufferMemory, ConversationBufferWindowMemory
 from src.database import get_supabase
@@ -38,9 +38,6 @@ ADMIN_EMAILS = set(
 )
 
 
-logger = logging.getLogger(__name__)
-
-
 def _get_next_id() -> str:
     _in_memory["next_id"] += 1
     return str(_in_memory["next_id"])

summary_generator/routes.py

@@ -1,6 +1,6 @@
 import asyncio
 import time
-import logging 
+from loguru import logger
 from fastapi import APIRouter, HTTPException, Depends
 
 from src.summary_generator.summary import summarizer, web_summarizer, fetch_web_content
@@ -17,7 +17,6 @@ from .schemas import SummarizeRequest, SummarizeResponse
 from .constants import MAX_COMBINED_TEXT_LEN
 
 router = APIRouter(prefix="/api/materials", tags=["Summarizer"])
-logger = logging.getLogger(__name__)
 
 
 @router.post("/summarize", response_model=SummarizeResponse)

summary_generator/summary.py

@@ -1,5 +1,5 @@
 import re
-import logging
+from loguru import logger
 import concurrent.futures
 from langchain_community.utilities import WikipediaAPIWrapper, DuckDuckGoSearchAPIWrapper
 from src.rag.rag import get_summary_llm, get_summary_fallback_llm
@@ -11,8 +11,6 @@ from .constants import (
     WIKI_DOC_CONTENT_CHARS_MAX,
 )
 
-logger = logging.getLogger(__name__)
-
 
 def summarizer_prompt():
     return SUMMARIZER_PROMPT_TEMPLATE