Files
office_translator/services/storage_tracker.py
2026-03-07 11:42:58 +01:00

150 lines
4.2 KiB
Python

import os
import json
import logging
from datetime import datetime, timezone
from typing import Optional, Any, Dict
from config import config
try:
import structlog
logger = structlog.get_logger(__name__)
_HAS_STRUCTLOG = True
except ImportError:
logger = logging.getLogger(__name__)
_HAS_STRUCTLOG = False
def _log_info(event: str, **kwargs):
"""Log info with structlog or standard logging compatibility."""
if _HAS_STRUCTLOG:
logger.info(event, **kwargs)
else:
msg = f"{event} " + " ".join(f"{k}={v}" for k, v in kwargs.items())
logger.info(msg)
def _log_error(event: str, **kwargs):
"""Log error with structlog or standard logging compatibility."""
if _HAS_STRUCTLOG:
logger.error(event, **kwargs)
else:
msg = f"{event} " + " ".join(f"{k}={v}" for k, v in kwargs.items())
logger.error(msg)
# Key pattern: translation:file:{job_id}
KEY_PREFIX = "translation:file"
def _get_default_ttl() -> int:
"""Get TTL from config or default to 60 minutes."""
try:
from config import config
return config.FILE_TTL_MINUTES * 60
except Exception:
return 3600 # 60 minutes default
_async_redis = None
def _get_async_redis():
"""Return async Redis client or None. Uses REDIS_URL from env."""
global _async_redis
if _async_redis is not None:
return _async_redis if _async_redis is not False else None
# Try to get from environment first
url = os.getenv("REDIS_URL", "").strip()
if not url:
_async_redis = False
return None
try:
import redis.asyncio as redis
_async_redis = redis.Redis.from_url(url, decode_responses=True)
_log_info("redis_connected", service="storage_tracker")
return _async_redis
except Exception as e:
_log_error("redis_connection_failed", service="storage_tracker", error=str(e))
_async_redis = False
return None
class StorageTracker:
"""
Tracks file locations and metadata in Redis.
Pattern: translation:file:{job_id} -> JSON metadata
"""
def __init__(self):
self._redis = None
def _redis_client(self):
if self._redis is None:
self._redis = _get_async_redis()
return self._redis
async def track_file(
self, job_id: str, metadata: Dict[str, Any], ttl: Optional[int] = None
) -> bool:
"""
Store file metadata in Redis with TTL and log the upload.
"""
if ttl is None:
ttl = _get_default_ttl()
# Ensure timestamp is present
if "timestamp" not in metadata:
metadata["timestamp"] = datetime.now(timezone.utc).isoformat()
# Log metadata (no content)
_log_info(
"file_uploaded",
job_id=job_id,
original_filename=metadata.get("original_filename"),
file_size=metadata.get("file_size"),
file_hash=metadata.get("file_hash"),
user_id=metadata.get("user_id"),
timestamp=metadata.get("timestamp"),
)
redis_client = self._redis_client()
if not redis_client:
_log_error(
"redis_not_available", job_id=job_id, hint="File tracked in logs only"
)
return False
try:
key = f"{KEY_PREFIX}:{job_id}"
await redis_client.set(key, json.dumps(metadata), ex=ttl)
_log_info("file_tracked_in_redis", job_id=job_id, ttl_seconds=ttl)
return True
except Exception as e:
_log_error("redis_track_failed", job_id=job_id, error=str(e))
return False
async def get_file_metadata(self, job_id: str) -> Optional[Dict[str, Any]]:
"""
Retrieve file metadata from Redis.
"""
redis_client = self._redis_client()
if not redis_client:
return None
try:
key = f"{KEY_PREFIX}:{job_id}"
data = await redis_client.get(key)
return json.loads(data) if data else None
except Exception as e:
_log_error("redis_get_failed", job_id=job_id, error=str(e))
return None
# Singleton for app use
storage_tracker = StorageTracker()