"""
Optimized configuration for ALL RAG systems - BACKWARD COMPATIBLE.
"""
import os
from pathlib import Path

# Base paths
BASE_DIR = Path(__file__).parent
DATA_DIR = BASE_DIR / "data"
MODELS_DIR = BASE_DIR / "models"
CACHE_DIR = BASE_DIR / ".cache"

# Ensure directories exist
for directory in [DATA_DIR, MODELS_DIR, CACHE_DIR]:
    directory.mkdir(exist_ok=True)

# Model Configuration
EMBEDDING_MODEL = "all-MiniLM-L6-v2"
LLM_MODEL = "microsoft/phi-2"

# ===== BACKWARD COMPATIBLE CONFIGS =====
# For Naive RAG and Optimized RAG
CHUNK_SIZE = 500
CHUNK_OVERLAP = 50
TOP_K = 5  # For backward compatibility

# For Optimized RAG
TOP_K_DYNAMIC_OPTIMIZED = {
    "short": 2,     # < 10 tokens
    "medium": 3,    # 10-30 tokens  
    "long": 4       # > 30 tokens
}

# For Hyper RAG (more aggressive)
TOP_K_DYNAMIC_HYPER = {
    "short": 3,     # < 5 words
    "medium": 4,    # 5-15 words
    "long": 5       # > 15 words
}

# Alias for backward compatibility
TOP_K_DYNAMIC = TOP_K_DYNAMIC_OPTIMIZED

# FAISS Configuration
FAISS_INDEX_PATH = DATA_DIR / "faiss_index.bin"
DOCSTORE_PATH = DATA_DIR / "docstore.db"

# Cache Configuration
EMBEDDING_CACHE_PATH = DATA_DIR / "embedding_cache.db"
QUERY_CACHE_TTL = 3600

# LLM Inference Configuration
MAX_TOKENS = 1024
TEMPERATURE = 0.1
CONTEXT_SIZE = 2048

# Performance Settings
ENABLE_EMBEDDING_CACHE = True
ENABLE_QUERY_CACHE = True
USE_QUANTIZED_LLM = False
BATCH_SIZE = 1

# FILTERING SETTINGS
ENABLE_PRE_FILTER = True
ENABLE_PROMPT_COMPRESSION = True
MIN_FILTER_MATCHES = 1
FILTER_EXPANSION_FACTOR = 2.0

# Dataset Configuration
SAMPLE_DOCUMENTS = 1000

# Monitoring
ENABLE_METRICS = True
METRICS_FILE = DATA_DIR / "metrics.csv"

# HYPER RAG SPECIFIC OPTIMIZATIONS
HYPER_CACHE_SIZE = 1000
HYPER_THREAD_WORKERS = 4
HYPER_MIN_CHUNKS = 1

# ===== CONFIG VALIDATION =====
def validate_config():
    """Validate configuration settings."""
    errors = []
    
    # Check required directories
    for dir_name, dir_path in [("DATA", DATA_DIR), ("MODELS", MODELS_DIR)]:
        if not dir_path.exists():
            errors.append(f"{dir_name} directory does not exist: {dir_path}")
    
    # Check FAISS index
    if not FAISS_INDEX_PATH.exists():
        print(f"⚠ WARNING: FAISS index not found at {FAISS_INDEX_PATH}")
        print("   Run: python scripts/initialize_rag.py")
    
    # Check embedding cache
    if ENABLE_EMBEDDING_CACHE and not EMBEDDING_CACHE_PATH.exists():
        print(f"⚠ WARNING: Embedding cache not found at {EMBEDDING_CACHE_PATH}")
        print("   It will be created automatically on first use.")
    
    if errors:
        print("\n❌ CONFIGURATION ERRORS:")
        for error in errors:
            print(f"   - {error}")
        return False
    
    print("✅ Configuration validated successfully")
    return True

# Auto-validate on import
if __name__ != "__main__":
    validate_config()