villsim/backend/core/ai.py

"""AI decision system for agents in the Village Simulation.

This module provides shared AI classes and utilities used by the GOAP
(Goal-Oriented Action Planning) system.

GOAP Benefits:
- Agents plan multi-step sequences to achieve goals
- Goals are dynamically prioritized based on state
- More emergent and adaptive behavior
- Easier to extend with new goals and actions

Major features:
- Each agent has unique personality traits affecting all decisions
- Emergent professions: Hunters, Gatherers, Traders, Generalists
- Class inequality through varied strategies and skills
- Traders focus on arbitrage (buy low, sell high)
- Personality affects: risk tolerance, hoarding, market participation
"""

from dataclasses import dataclass, field
from typing import Optional, TYPE_CHECKING

from backend.domain.agent import Agent
from backend.domain.action import ActionType, ACTION_CONFIG
from backend.domain.resources import ResourceType

if TYPE_CHECKING:
    from backend.core.market import OrderBook


@dataclass
class TradeItem:
    """A single item to buy/sell in a trade."""
    order_id: str
    resource_type: ResourceType
    quantity: int
    price_per_unit: int


@dataclass
class AIDecision:
    """A decision made by the AI for an agent."""
    action: ActionType
    target_resource: Optional[ResourceType] = None
    order_id: Optional[str] = None
    quantity: int = 1
    price: int = 0
    reason: str = ""
    # For multi-item trades
    trade_items: list[TradeItem] = field(default_factory=list)
    # For price adjustments
    adjust_order_id: Optional[str] = None
    new_price: Optional[int] = None

    def to_dict(self) -> dict:
        return {
            "action": self.action.value,
            "target_resource": self.target_resource.value if self.target_resource else None,
            "order_id": self.order_id,
            "quantity": self.quantity,
            "price": self.price,
            "reason": self.reason,
            "trade_items": [
                {
                    "order_id": t.order_id,
                    "resource_type": t.resource_type.value,
                    "quantity": t.quantity,
                    "price_per_unit": t.price_per_unit,
                }
                for t in self.trade_items
            ],
            "adjust_order_id": self.adjust_order_id,
            "new_price": self.new_price,
        }


# Resource to action for gathering
RESOURCE_ACTIONS: dict[ResourceType, ActionType] = {
    ResourceType.MEAT: ActionType.HUNT,
    ResourceType.BERRIES: ActionType.GATHER,
    ResourceType.WATER: ActionType.GET_WATER,
    ResourceType.WOOD: ActionType.CHOP_WOOD,
    ResourceType.HIDE: ActionType.HUNT,
    ResourceType.CLOTHES: ActionType.WEAVE,
}

# Energy cost to gather each resource (used for efficiency calculations)
def get_energy_cost(resource_type: ResourceType) -> int:
    """Get the energy cost to produce one unit of a resource."""
    action = RESOURCE_ACTIONS.get(resource_type)
    if not action:
        return 10
    config = ACTION_CONFIG.get(action)
    if not config:
        return 10
    energy_cost = abs(config.energy_cost)
    avg_output = max(1, (config.min_output + config.max_output) / 2) if config.output_resource else 1
    return int(energy_cost / avg_output)


# Cached config values to avoid repeated lookups
_cached_ai_config = None
_cached_economy_config = None


def _get_ai_config():
    """Get AI-relevant configuration values (cached)."""
    global _cached_ai_config
    if _cached_ai_config is None:
        from backend.config import get_config
        _cached_ai_config = get_config().agent_stats
    return _cached_ai_config


def _get_economy_config():
    """Get economy/market configuration values (cached)."""
    global _cached_economy_config
    if _cached_economy_config is None:
        from backend.config import get_config
        _cached_economy_config = getattr(get_config(), 'economy', None)
    return _cached_economy_config


def reset_ai_config_cache():
    """Reset the cached config values (call after config reload)."""
    global _cached_ai_config, _cached_economy_config
    _cached_ai_config = None
    _cached_economy_config = None


def get_ai_decision(
    agent: Agent,
    market: "OrderBook",
    step_in_day: int = 1,
    day_steps: int = 10,
    current_turn: int = 0,
    is_night: bool = False,
) -> AIDecision:
    """Get an AI decision for an agent using GOAP (Goal-Oriented Action Planning).

    Args:
        agent: The agent to make a decision for
        market: The market order book
        step_in_day: Current step within the day
        day_steps: Total steps per day
        current_turn: Current simulation turn
        is_night: Whether it's currently night time

    Returns:
        AIDecision with the chosen action and parameters
    """
    from backend.core.goap.goap_ai import get_goap_decision
    return get_goap_decision(
        agent=agent,
        market=market,
        step_in_day=step_in_day,
        day_steps=day_steps,
        current_turn=current_turn,
        is_night=is_night,
    )