core-api/src/config.py

"""
Global configuration for Core Code API
"""
from pydantic_settings import BaseSettings
from functools import lru_cache

# Import infrastructure credentials from gitignored module
try:
    from src.credentials import (
        PORTAINER_URL, PORTAINER_API_KEY,
        NPM_URL, NPM_EMAIL, NPM_PASSWORD,
        KUMA_URL, KUMA_USERNAME, KUMA_PASSWORD, KUMA_API_KEY,
        BRAVE_SEARCH_API_KEY,
        GOOGLE_SEARCH_API_KEY, GOOGLE_SEARCH_ENGINE_ID
    )
except ImportError:
    # Fallback to empty strings if credentials.py doesn't exist
    # (e.g., fresh clone before credentials setup)
    PORTAINER_URL = "http://localhost:8001"
    PORTAINER_API_KEY = ""
    NPM_URL = "http://localhost:81"
    NPM_EMAIL = ""
    NPM_PASSWORD = ""
    KUMA_URL = "http://localhost:3001"
    KUMA_USERNAME = ""
    KUMA_PASSWORD = ""
    KUMA_API_KEY = ""
    BRAVE_SEARCH_API_KEY = ""
    GOOGLE_SEARCH_API_KEY = ""
    GOOGLE_SEARCH_ENGINE_ID = ""


class Settings(BaseSettings):
    """Global application settings"""

    # Application
    app_name: str = "Core Code API"
    app_version: str = "1.0.0"
    debug: bool = False

    # Server
    host: str = "0.0.0.0"
    port: int = 8083

    # CORS
    cors_origins: list[str] = ["*"]
    cors_credentials: bool = True
    cors_methods: list[str] = ["*"]
    cors_headers: list[str] = ["*"]

    # Logging
    log_level: str = "DEBUG"

    # Ollama Configuration (for AI orchestration)
    ollama_base_url: str = "http://ollama:11434"
    ollama_timeout: int = 300  # 5 minutes

    # Model Configuration
    default_model: str = "mistral-tools:7b"
    agent_model: str = "gemma2:9b-instruct-q5_K_M"  # Must support tool calling with ADK (~4GB VRAM)
    lightweight_models: str = "gemma3-tools:1b,phi3:mini"
    heavy_models: str = "mistral:7b,gemma2:9b,gemma3:12b,mixtral:8x7b"
    code_models: str = "codestral:latest,codegemma:latest"
    # Previous config (gemma3:12b used ~10GB VRAM)
    # default_model: str = "gemma3:12b"
    # agent_model: str = "gemma3:12b"

    # System Prompt Variant (for A/B testing)
    # Options: v1_verbose, v2_concise, v3_imperative, v4_minimal, v4_gemini_suggestion, v5_adk_optimized, v7_adk_best_practice, v8_holistic
    system_prompt_variant: str = "v8_holistic"

    # Agent Configuration
    agent_fallback_enabled: bool = True

    # Model Aliases (OpenAI → Local)
    alias_gpt35: str = "gemma:7b"
    alias_gpt4: str = "mistral:7b"
    alias_gpt4_turbo: str = "mixtral:8x7b"
    alias_gpt4_code: str = "codestral:latest"

    # Memory Configuration
    memory_tier1_max_turns: int = 10
    memory_consolidation_threshold: int = 10

    # Qdrant Configuration
    qdrant_host: str = "qdrant"
    qdrant_port: int = 6333
    qdrant_collection_conversations: str = "core_api_conversations"
    qdrant_collection_documents: str = "core_api_documents"
    qdrant_collection_user_facts: str = "core_api_user_facts"

    # Embeddings (using Ollama - no local models needed)
    embedding_model: str = "nomic-embed-text"  # Ollama embedding model
    embedding_dimension: int = 768  # nomic-embed-text dimension
    embedding_batch_size: int = 32

    # Search Configuration
    search_provider: str = "google"  # Options: google, brave, searxng, duckduckgo
    searxng_url: str = "http://searxng:8080"  # For future self-hosted SearxNG

    # Search API Keys (from credentials.py)
    brave_search_api_key: str = BRAVE_SEARCH_API_KEY  # https://brave.com/search/api/
    google_search_api_key: str = GOOGLE_SEARCH_API_KEY  # https://console.cloud.google.com/
    google_search_engine_id: str = GOOGLE_SEARCH_ENGINE_ID  # Custom Search Engine ID

    # Infrastructure Management (from credentials.py)
    portainer_url: str = PORTAINER_URL
    portainer_api_key: str = PORTAINER_API_KEY

    npm_url: str = NPM_URL
    npm_email: str = NPM_EMAIL
    npm_password: str = NPM_PASSWORD

    kuma_url: str = KUMA_URL
    kuma_username: str = KUMA_USERNAME
    kuma_password: str = KUMA_PASSWORD
    kuma_api_key: str = KUMA_API_KEY

    # Core-AI Service (AI performance metrics)
    core_ai_base_url: str = "http://core-ai:8086"

    # OIDC Authentication (Authentik)
    oidc_enabled: bool = False  # Set to True to require authentication
    oidc_issuer: str = "https://auth.schweitz.net/application/o/core-api/"
    oidc_audience: str = "core-api"

    @property
    def model_aliases(self) -> dict:
        """Computed property for model aliases"""
        return {
            "gpt-3.5-turbo": self.alias_gpt35,
            "gpt-4": self.alias_gpt4,
            "gpt-4-turbo": self.alias_gpt4_turbo,
            "gpt-4-code": self.alias_gpt4_code,
        }

    def get_lightweight_models(self) -> list[str]:
        """Parse comma-separated lightweight models"""
        return [m.strip().strip('"').strip("'") for m in self.lightweight_models.split(",") if m.strip()]

    def get_heavy_models(self) -> list[str]:
        """Parse comma-separated heavy models"""
        return [m.strip().strip('"').strip("'") for m in self.heavy_models.split(",") if m.strip()]

    def get_code_models(self) -> list[str]:
        """Parse comma-separated code models"""
        return [m.strip().strip('"').strip("'") for m in self.code_models.split(",") if m.strip()]

    class Config:
        env_file = ".env"
        case_sensitive = False


@lru_cache()
def get_settings() -> Settings:
    """Cached settings instance"""
    return Settings()