"""
Configuration module for Log Analyzer MCP Server
FAISS-based vector search with local embeddings
"""
import os
from pathlib import Path
from dotenv import load_dotenv
# Load environment variables from .env file
load_dotenv()
class Config:
"""Configuration management with environment variables"""
# ========================================================================
# EMBEDDING MODEL CONFIGURATION
# ========================================================================
# Local embedding model (sentence-transformers)
EMBED_MODEL_NAME = os.environ.get('EMBED_MODEL', 'all-MiniLM-L6-v2')
EMBED_DIMENSION = int(os.environ.get('EMBED_DIMENSION', '384')) # Dimension for all-MiniLM-L6-v2
# ========================================================================
# LOG ANALYZER CONFIGURATION
# ========================================================================
# Get base directory (project root)
_BASE_DIR = Path(__file__).parent.absolute()
# File paths - use relative paths by default
LOG_FOLDER = os.environ.get('LOG_FOLDER', str(_BASE_DIR / 'logs'))
# Log file extensions to process
LOG_EXTENSIONS = ('.log', '.txt', '.out', '.err')
# Ensure folders exist
@classmethod
def ensure_folders(cls):
"""Create required folders if they don't exist"""
Path(cls.LOG_FOLDER).mkdir(parents=True, exist_ok=True)
# Chunking parameters
MAX_CHARS_PER_CHUNK = int(os.environ.get('MAX_CHARS_PER_CHUNK', '5000'))
DEFAULT_CHUNK_SIZE = int(os.environ.get('DEFAULT_CHUNK_SIZE', '4096'))
DEFAULT_OVERLAP = int(os.environ.get('DEFAULT_OVERLAP', '1024'))
# ========================================================================
# FAISS CONFIGURATION
# ========================================================================
# FAISS index type: 'Flat' (exact), 'IVFFlat' (fast), 'IVFPQ' (memory efficient), 'HNSW' (graph-based)
FAISS_INDEX_TYPE = os.environ.get('FAISS_INDEX_TYPE', 'IVFFlat')
# IVF parameters
FAISS_NLIST = int(os.environ.get('FAISS_NLIST', '100')) # Number of clusters for IVF
FAISS_NPROBE = int(os.environ.get('FAISS_NPROBE', '10')) # Clusters to search
# Search parameters
FAISS_TOP_K = int(os.environ.get('FAISS_TOP_K', '150')) # Default top-k results
# Index persistence
FAISS_INDEX_PATH = os.environ.get(
'FAISS_INDEX_PATH',
str(_BASE_DIR / 'logs' / 'faiss_index')
)
# ========================================================================
# HELPER METHODS
# ========================================================================
@classmethod
def validate(cls) -> bool:
"""Validate required configuration"""
# All configuration is optional with sensible defaults
return True
