"""Researcher Agent — gathers information and analyzes topics.

Handles steps that involve research, analysis, or information gathering.
Uses the LLM to generate insights and summaries.
"""

from __future__ import annotations

import logging
from typing import Any

from .agent_base import BaseAgent
from ..memory.goal_memory import Goal

logger = logging.getLogger(__name__)


class ResearcherAgent(BaseAgent):
    """Researches topics and gathers information for goal steps."""

    def __init__(self, goal_memory, persistent_memory=None, generate_fn=None):
        super().__init__(
            name="researcher",
            role="Researcher",
            description="Gathers information, analyzes topics, and generates insights",
            goal_memory=goal_memory,
            persistent_memory=persistent_memory,
            generate_fn=generate_fn,
            poll_interval_s=3.0,
        )

    def _can_handle(self, goal: Goal) -> bool:
        """Researcher handles goals related to research/analysis."""
        keywords = ["research", "analyze", "investigate", "study", "learn", "understand",
                    "explore", "gather", "find", "search", "compare", "evaluate"]
        text = (goal.title + " " + goal.description).lower()
        return any(kw in text for kw in keywords)

    def process_goal(self, goal: Goal) -> dict[str, Any]:
        """Execute a research step."""
        if goal.current_step >= len(goal.steps):
            return {"success": True, "output": "No more steps"}

        step = goal.steps[goal.current_step]
        prompt = (
            f"You are a research agent. Execute this step:\n"
            f"Goal: {goal.title}\n"
            f"Step: {step['title']}\n"
            f"Description: {step['description']}\n"
            f"Research and provide a concise summary of findings.\n"
        )

        response = self._generate(prompt)

        # Store findings in persistent memory
        if self.persistent_memory and response:
            self.persistent_memory.add_episodic(
                "research", f"Research for '{goal.title}': {response[:200]}",
                importance=0.7, tags=["research", goal.title[:20]]
            )

        return {"success": True, "output": response[:200]}