feat: Add SaaS robustness middleware - Rate limiting with token bucket and sliding window algorithms - Input validation (file, language, provider) - Security headers middleware (CSP, XSS protection, etc.) - Automatic file cleanup with TTL tracking - Memory and disk monitoring - Enhanced health check and metrics endpoints - Request logging with unique IDs

config.py

@@ -1,5 +1,6 @@
 """
 Configuration module for the Document Translation API
+SaaS-ready with comprehensive settings for production deployment
 """
 import os
 from pathlib import Path
@@ -8,7 +9,7 @@ from dotenv import load_dotenv
 load_dotenv()
 
 class Config:
-    # Translation Service
+    # ============== Translation Service ==============
     TRANSLATION_SERVICE = os.getenv("TRANSLATION_SERVICE", "google")
     DEEPL_API_KEY = os.getenv("DEEPL_API_KEY", "")
     
@@ -17,20 +18,51 @@ class Config:
     OLLAMA_MODEL = os.getenv("OLLAMA_MODEL", "llama3")
     OLLAMA_VISION_MODEL = os.getenv("OLLAMA_VISION_MODEL", "llava")
     
-    # File Upload Configuration
+    # ============== File Upload Configuration ==============
     MAX_FILE_SIZE_MB = int(os.getenv("MAX_FILE_SIZE_MB", "50"))
     MAX_FILE_SIZE_BYTES = MAX_FILE_SIZE_MB * 1024 * 1024
     
     # Directories
-    BASE_DIR = Path(__file__).parent.parent
+    BASE_DIR = Path(__file__).parent
     UPLOAD_DIR = BASE_DIR / "uploads"
     OUTPUT_DIR = BASE_DIR / "outputs"
     TEMP_DIR = BASE_DIR / "temp"
+    LOGS_DIR = BASE_DIR / "logs"
     
     # Supported file types
     SUPPORTED_EXTENSIONS = {".xlsx", ".docx", ".pptx"}
     
-    # API Configuration
+    # ============== Rate Limiting (SaaS) ==============
+    RATE_LIMIT_ENABLED = os.getenv("RATE_LIMIT_ENABLED", "true").lower() == "true"
+    RATE_LIMIT_PER_MINUTE = int(os.getenv("RATE_LIMIT_PER_MINUTE", "30"))
+    RATE_LIMIT_PER_HOUR = int(os.getenv("RATE_LIMIT_PER_HOUR", "200"))
+    TRANSLATIONS_PER_MINUTE = int(os.getenv("TRANSLATIONS_PER_MINUTE", "10"))
+    TRANSLATIONS_PER_HOUR = int(os.getenv("TRANSLATIONS_PER_HOUR", "50"))
+    MAX_CONCURRENT_TRANSLATIONS = int(os.getenv("MAX_CONCURRENT_TRANSLATIONS", "5"))
+    
+    # ============== Cleanup Service ==============
+    CLEANUP_ENABLED = os.getenv("CLEANUP_ENABLED", "true").lower() == "true"
+    CLEANUP_INTERVAL_MINUTES = int(os.getenv("CLEANUP_INTERVAL_MINUTES", "15"))
+    FILE_TTL_MINUTES = int(os.getenv("FILE_TTL_MINUTES", "60"))
+    INPUT_FILE_TTL_MINUTES = int(os.getenv("INPUT_FILE_TTL_MINUTES", "30"))
+    OUTPUT_FILE_TTL_MINUTES = int(os.getenv("OUTPUT_FILE_TTL_MINUTES", "120"))
+    
+    # Disk space thresholds
+    DISK_WARNING_THRESHOLD_GB = float(os.getenv("DISK_WARNING_THRESHOLD_GB", "5.0"))
+    DISK_CRITICAL_THRESHOLD_GB = float(os.getenv("DISK_CRITICAL_THRESHOLD_GB", "1.0"))
+    
+    # ============== Security ==============
+    ENABLE_HSTS = os.getenv("ENABLE_HSTS", "false").lower() == "true"
+    CORS_ORIGINS = os.getenv("CORS_ORIGINS", "*").split(",")
+    MAX_REQUEST_SIZE_MB = int(os.getenv("MAX_REQUEST_SIZE_MB", "100"))
+    REQUEST_TIMEOUT_SECONDS = int(os.getenv("REQUEST_TIMEOUT_SECONDS", "300"))
+    
+    # ============== Monitoring ==============
+    LOG_LEVEL = os.getenv("LOG_LEVEL", "INFO")
+    ENABLE_REQUEST_LOGGING = os.getenv("ENABLE_REQUEST_LOGGING", "true").lower() == "true"
+    MAX_MEMORY_PERCENT = float(os.getenv("MAX_MEMORY_PERCENT", "80"))
+    
+    # ============== API Configuration ==============
     API_TITLE = "Document Translation API"
     API_VERSION = "1.0.0"
     API_DESCRIPTION = """
@@ -40,6 +72,12 @@ class Config:
     - Excel (.xlsx) - Preserves cell formatting, formulas, merged cells, images
     - Word (.docx) - Preserves styles, tables, images, headers/footers
     - PowerPoint (.pptx) - Preserves layouts, animations, embedded media
+    
+    SaaS Features:
+    - Rate limiting per client IP
+    - Automatic file cleanup
+    - Health monitoring
+    - Request logging
     """
     
     @classmethod
@@ -48,5 +86,7 @@ class Config:
         cls.UPLOAD_DIR.mkdir(exist_ok=True, parents=True)
         cls.OUTPUT_DIR.mkdir(exist_ok=True, parents=True)
         cls.TEMP_DIR.mkdir(exist_ok=True, parents=True)
+        cls.LOGS_DIR.mkdir(exist_ok=True, parents=True)
+
 
 config = Config()