Agent/examples_how/overall_derivation/overall_derivation_agent_run.py

"""
选题点整体推导 Agent（增强版）

参考 examples/how/run.py，提供：
1. 命令行交互：输入 'p' 暂停、'q' 退出
2. 暂停后可插入干预消息、触发经验总结、查看 GoalTree、手动压缩上下文
3. 支持 --trace <ID> 恢复已有 Trace 继续执行
4. 使用 SimplePrompt 加载 derivation_main.md，支持评估子 agent（agent_type=evaluate_derivation）
"""

import argparse
import os
import sys
import select
import asyncio
from datetime import datetime
from pathlib import Path

# 与 examples/how/run.py 一致：禁止 httpx/urllib 自动检测系统 HTTP 代理
# os.environ.setdefault("no_proxy", "*")

# 添加项目根目录到 Python 路径
sys.path.insert(0, str(Path(__file__).parent.parent.parent))

from dotenv import load_dotenv

from agent.llm.prompts import SimplePrompt
from agent.core.runner import AgentRunner, RunConfig
from agent.core.presets import AgentPreset, register_preset
from agent.trace import (
    FileSystemTraceStore,
    Trace,
    Message,
)
from agent.llm import create_openrouter_llm_call
from agent.trace.compaction import build_reflect_prompt

load_dotenv()


# ===== 非阻塞 stdin 检测 =====
if sys.platform == 'win32':
    import msvcrt


def check_stdin() -> str | None:
    """
    跨平台非阻塞检查 stdin 输入。
    Windows: msvcrt.kbhit()；macOS/Linux: select.select()
    """
    if sys.platform == 'win32':
        if msvcrt.kbhit():
            ch = msvcrt.getwch().lower()
            if ch == 'p':
                return 'pause'
            if ch == 'q':
                return 'quit'
        return None
    else:
        ready, _, _ = select.select([sys.stdin], [], [], 0)
        if ready:
            line = sys.stdin.readline().strip().lower()
            if line in ('p', 'pause'):
                return 'pause'
            if line in ('q', 'quit'):
                return 'quit'
        return None


# ===== 交互菜单 =====

def _read_multiline() -> str:
    """读取多行输入，以连续两次回车（空行）结束。"""
    print("\n请输入干预消息（连续输入两次回车结束）：")
    lines: list[str] = []
    blank_count = 0
    while True:
        line = input()
        if line == "":
            blank_count += 1
            if blank_count >= 2:
                break
            lines.append("")
        else:
            blank_count = 0
            lines.append(line)
    while lines and lines[-1] == "":
        lines.pop()
    return "\n".join(lines)


async def show_interactive_menu(
    runner: AgentRunner,
    trace_id: str,
    current_sequence: int,
    store: FileSystemTraceStore,
):
    """显示交互式菜单，让用户选择操作。"""
    print("\n" + "=" * 60)
    print("  执行已暂停")
    print("=" * 60)
    print("请选择操作：")
    print("  1. 插入干预消息并继续")
    print("  2. 触发经验总结（reflect）")
    print("  3. 查看当前 GoalTree")
    print("  4. 手动压缩上下文（compact）")
    print("  5. 继续执行")
    print("  6. 停止执行")
    print("=" * 60)

    while True:
        choice = input("请输入选项 (1-6): ").strip()

        if choice == "1":
            text = _read_multiline()
            if not text:
                print("未输入任何内容，取消操作")
                continue
            print("\n将插入干预消息并继续执行...")
            live_trace = await store.get_trace(trace_id)
            actual_sequence = live_trace.last_sequence if live_trace and live_trace.last_sequence else current_sequence
            return {
                "action": "continue",
                "messages": [{"role": "user", "content": text}],
                "after_sequence": actual_sequence,
            }

        elif choice == "2":
            print("\n触发经验总结...")
            focus = input("请输入反思重点（可选，直接回车跳过）: ").strip()
            trace = await store.get_trace(trace_id)
            saved_head = trace.head_sequence
            prompt = build_reflect_prompt()
            if focus:
                prompt += f"\n\n请特别关注：{focus}"
            print("正在生成反思...")
            reflect_cfg = RunConfig(trace_id=trace_id, max_iterations=1, tools=[])
            reflection_text = ""
            try:
                result = await runner.run_result(
                    messages=[{"role": "user", "content": prompt}],
                    config=reflect_cfg,
                )
                reflection_text = result.get("summary", "")
            finally:
                await store.update_trace(trace_id, head_sequence=saved_head)
            if reflection_text:
                from datetime import datetime
                experiences_path = runner.experiences_path or "./.cache/experiences_overall_derivation.md"
                os.makedirs(os.path.dirname(experiences_path), exist_ok=True)
                header = f"\n\n---\n\n## {trace_id} ({datetime.now().strftime('%Y-%m-%d %H:%M')})\n\n"
                with open(experiences_path, "a", encoding="utf-8") as f:
                    f.write(header + reflection_text + "\n")
                print(f"\n反思已保存到: {experiences_path}")
                print("\n--- 反思内容 ---")
                print(reflection_text)
                print("--- 结束 ---\n")
            else:
                print("未生成反思内容")
            continue

        elif choice == "3":
            goal_tree = await store.get_goal_tree(trace_id)
            if goal_tree and goal_tree.goals:
                print("\n当前 GoalTree:")
                print(goal_tree.to_prompt())
            else:
                print("\n当前没有 Goal")
            continue

        elif choice == "4":
            print("\n正在执行上下文压缩（compact）...")
            try:
                goal_tree = await store.get_goal_tree(trace_id)
                trace = await store.get_trace(trace_id)
                if not trace:
                    print("未找到 Trace，无法压缩")
                    continue
                main_path = await store.get_main_path_messages(trace_id, trace.head_sequence)
                history = [msg.to_llm_dict() for msg in main_path]
                head_seq = main_path[-1].sequence if main_path else 0
                next_seq = head_seq + 1
                compact_config = RunConfig(trace_id=trace_id)
                new_history, new_head, new_seq = await runner._compress_history(
                    trace_id=trace_id,
                    history=history,
                    goal_tree=goal_tree,
                    config=compact_config,
                    sequence=next_seq,
                    head_seq=head_seq,
                )
                print(f"\n✅ 压缩完成: {len(history)} 条消息 → {len(new_history)} 条")
            except Exception as e:
                print(f"\n❌ 压缩失败: {e}")
            continue

        elif choice == "5":
            print("\n继续执行...")
            return {"action": "continue"}

        elif choice == "6":
            print("\n停止执行...")
            return {"action": "stop"}

        else:
            print("无效选项，请重新输入")


def _replace_prompt_placeholders(
    messages: list,
    account_name: str,
    post_id: str,
    log_id: str,
    post_point_count: int,
) -> None:
    """在 messages 的 content 中用 replace 替换 {account_name}, {帖子ID}, {log_id}, {post_point_count}。"""
    post_point_count_str = str(post_point_count)
    for m in messages:
        content = m.get("content")
        if isinstance(content, str):
            m["content"] = (
                content.replace("{account_name}", account_name)
                .replace("{帖子ID}", post_id)
                .replace("{log_id}", log_id)
                .replace("{post_point_count}", post_point_count_str)
            )
        elif isinstance(content, list):
            for part in content:
                if isinstance(part, dict) and part.get("type") == "text":
                    part["text"] = (
                        (part.get("text") or "")
                        .replace("{account_name}", account_name)
                        .replace("{帖子ID}", post_id)
                        .replace("{log_id}", log_id)
                        .replace("{post_point_count}", post_point_count_str)
                    )


async def main(account_name, post_id):
    parser = argparse.ArgumentParser(description="选题点整体推导 Agent（支持交互与恢复）")
    parser.add_argument(
        "--trace", type=str, default=None,
        help="已有的 Trace ID，用于恢复继续执行（不指定则新建）",
    )
    args = parser.parse_args()

    base_dir = Path(__file__).parent
    prompt_path = base_dir / "derivation_main.md"
    output_dir = base_dir / "output"
    output_dir.mkdir(exist_ok=True)

    # 加载项目级 presets（evaluate_derivation、derivation_search 等）
    presets_path = base_dir / "presets.json"
    if presets_path.exists():
        import json
        with open(presets_path, "r", encoding="utf-8") as f:
            project_presets = json.load(f)
        for name, cfg in project_presets.items():
            register_preset(name, AgentPreset(**cfg))
        print(f"   - 已加载项目 presets: {list(project_presets.keys())}")

    # 注册选题点推导专用工具（主 agent 与评估子 agent 会调用）
    import importlib.util
    tools_dir = base_dir / "tools"
    for mod_name, file_name in [
        ("find_tree_node", "find_tree_node.py"),
        ("find_pattern", "find_pattern.py"),
        ("point_match", "point_match.py"),
    ]:
        path = tools_dir / file_name
        if path.is_file():
            spec = importlib.util.spec_from_file_location(f"overall_derivation.{mod_name}", path)
            mod = importlib.util.module_from_spec(spec)
            spec.loader.exec_module(mod)
            print(f"   - 已注册推导工具: {mod_name}")

    skills_dir = str(base_dir / "skills")

    print("=" * 60)
    print("选题点整体推导 Agent（交互增强）")
    print("=" * 60)
    print()
    print("💡 交互提示：")
    print("   - 执行过程中输入 'p' 或 'pause' 暂停并进入交互模式")
    print("   - 执行过程中输入 'q' 或 'quit' 停止执行")
    print("=" * 60)
    print()

    # 在读取 prompt 前生成 log_id（格式 yyyyMMddHHmmss），保证每次运行使用同一 log_id（用于推导日志输出路径）
    log_id = datetime.now().strftime("%Y%m%d%H%M%S")
    print(f"   - 本次运行 log_id: {log_id}")
    print(f"   - account_name: {account_name}")
    print(f"   - post_id: {post_id}")


    # 读取选题点列表，得到 post_point_count（用于 prompt 占位符）
    input_dir = base_dir / "input" / account_name / "post_topic"
    post_topic_path = input_dir / f"{post_id}.json"
    post_point_count = 0
    if post_topic_path.exists():
        import json
        with open(post_topic_path, "r", encoding="utf-8") as f:
            post_topics = json.load(f)
        post_point_count = len(post_topics) if isinstance(post_topics, list) else 0
        print(f"   - 选题点数量 post_point_count: {post_point_count} (来自 {post_topic_path.relative_to(base_dir)})")
    else:
        print(f"   - 未找到选题点文件: {post_topic_path}，post_point_count 使用 0")

    print("1. 加载 prompt 配置...")
    prompt = SimplePrompt(prompt_path)

    print("2. 构建任务消息...")
    messages = prompt.build_messages()
    _replace_prompt_placeholders(messages, account_name, post_id, log_id, post_point_count)

    print("3. 创建 Agent Runner...")
    print(f"   - Skills 目录: {skills_dir}")
    model_key = prompt.config.get("model", "google/gemini-3-flash-preview")
    # model_id = f"google/{model_key}" if not model_key.startswith("google/") else model_key
    model_id = model_key
    print(f"   - 模型: {model_id}")

    store = FileSystemTraceStore(base_path=".trace")
    runner = AgentRunner(
        trace_store=store,
        llm_call=create_openrouter_llm_call(model=model_id),
        skills_dir=skills_dir,
        # experiences_path="./.cache/experiences_overall_derivation.md",
        debug=True,
    )

    resume_trace_id = args.trace
    if resume_trace_id:
        existing_trace = await store.get_trace(resume_trace_id)
        if not existing_trace:
            print(f"\n错误: Trace 不存在: {resume_trace_id}")
            sys.exit(1)
        print(f"4. 恢复已有 Trace: {resume_trace_id[:8]}...")
        print(f"   - 状态: {existing_trace.status}")
        print(f"   - 消息数: {existing_trace.total_messages}")
        print(f"   - 任务: {existing_trace.task}")
    else:
        print("4. 启动新 Agent 模式...")

    print()

    final_response = ""
    current_trace_id = resume_trace_id
    current_sequence = 0
    should_exit = False

    try:
        if resume_trace_id:
            initial_messages = None
            config = RunConfig(
                model=model_id,
                temperature=float(prompt.config.get("temperature", 0.3)),
                max_iterations=200,
                trace_id=resume_trace_id,
            )
        else:
            initial_messages = messages
            config = RunConfig(
                model=model_id,
                temperature=float(prompt.config.get("temperature", 0.3)),
                max_iterations=200,
                name="选题点整体推导任务",
            )

        while not should_exit:
            if current_trace_id:
                config.trace_id = current_trace_id

            final_response = ""

            if current_trace_id and initial_messages is None:
                check_trace = await store.get_trace(current_trace_id)
                if check_trace and check_trace.status in ("completed", "failed"):
                    if check_trace.status == "completed":
                        print("\n[Trace] ✅ 已完成")
                        print(f"  - Total messages: {check_trace.total_messages}")
                        print(f"  - Total cost: ${check_trace.total_cost:.4f}")
                    else:
                        print(f"\n[Trace] ❌ 已失败: {check_trace.error_message}")
                    current_sequence = check_trace.head_sequence
                    menu_result = await show_interactive_menu(
                        runner, current_trace_id, current_sequence, store
                    )
                    if menu_result["action"] == "stop":
                        break
                    elif menu_result["action"] == "continue":
                        new_messages = menu_result.get("messages", [])
                        if new_messages:
                            initial_messages = new_messages
                            config.after_sequence = menu_result.get("after_sequence")
                        else:
                            initial_messages = []
                            config.after_sequence = None
                        continue
                    break
                initial_messages = []

            print(f"{'▶️ 开始执行...' if not current_trace_id else '▶️ 继续执行...'}")

            paused = False
            try:
                async for item in runner.run(messages=initial_messages, config=config):
                    cmd = check_stdin()
                    if cmd == 'pause':
                        print("\n⏸️ 正在暂停执行...")
                        if current_trace_id:
                            await runner.stop(current_trace_id)
                        await asyncio.sleep(0.5)
                        menu_result = await show_interactive_menu(
                            runner, current_trace_id, current_sequence, store
                        )
                        if menu_result["action"] == "stop":
                            should_exit = True
                            paused = True
                            break
                        elif menu_result["action"] == "continue":
                            new_messages = menu_result.get("messages", [])
                            if new_messages:
                                initial_messages = new_messages
                                after_seq = menu_result.get("after_sequence")
                                if after_seq is not None:
                                    config.after_sequence = after_seq
                                paused = True
                                break
                            else:
                                initial_messages = []
                                config.after_sequence = None
                                paused = True
                                break

                    elif cmd == 'quit':
                        print("\n🛑 用户请求停止...")
                        if current_trace_id:
                            await runner.stop(current_trace_id)
                        should_exit = True
                        break

                    if isinstance(item, Trace):
                        current_trace_id = item.trace_id
                        if item.status == "running":
                            print(f"[Trace] 开始: {item.trace_id[:100]}...")
                        elif item.status == "completed":
                            print("\n[Trace] ✅ 完成")
                            print(f"  - Total messages: {item.total_messages}")
                            print(f"  - Total tokens: {item.total_tokens}")
                            print(f"  - Total cost: ${item.total_cost:.4f}")
                        elif item.status == "failed":
                            print(f"\n[Trace] ❌ 失败: {item.error_message}")
                        elif item.status == "stopped":
                            print("\n[Trace] ⏸️ 已停止")

                    elif isinstance(item, Message):
                        current_sequence = item.sequence
                        if item.role == "assistant":
                            content = item.content
                            if isinstance(content, dict):
                                text = content.get("text", "")
                                tool_calls = content.get("tool_calls")
                                if text and not tool_calls:
                                    final_response = text
                                    print("\n[Response] Agent 回复：")
                                    print(text)
                                elif text:
                                    preview = text[:500] + "..." if len(text) > 500 else text
                                    print(f"[Assistant] {preview}")
                                if tool_calls:
                                    for tc in tool_calls:
                                        tool_name = tc.get("function", {}).get("name", "unknown")
                                        tool_args = tc.get("function", {}).get("arguments", "")
                                        print(f"[Tool Call] 🛠️  {tool_name}")
                                        print(f"  params: {tool_args}")
                        elif item.role == "tool":
                            content = item.content
                            if isinstance(content, dict):
                                tool_name = content.get("tool_name", "unknown")
                                tool_result = content.get("result", content)
                                print(f"[Tool Result] ✅ {tool_name}")
                                print(f"  result: {tool_result}")
                            if item.description:
                                desc = item.description[:500] if len(item.description) > 500 else item.description
                                print(f"  {desc}...")

            except Exception as e:
                print(f"\n执行出错: {e}")
                import traceback
                traceback.print_exc()

            if paused:
                if should_exit:
                    break
                continue

            if should_exit:
                break

            if current_trace_id:
                menu_result = await show_interactive_menu(
                    runner, current_trace_id, current_sequence, store
                )
                if menu_result["action"] == "stop":
                    break
                elif menu_result["action"] == "continue":
                    new_messages = menu_result.get("messages", [])
                    if new_messages:
                        initial_messages = new_messages
                        config.after_sequence = menu_result.get("after_sequence")
                    else:
                        initial_messages = []
                        config.after_sequence = None
                    continue
            break

    except KeyboardInterrupt:
        print("\n\n用户中断 (Ctrl+C)")
        if current_trace_id:
            await runner.stop(current_trace_id)

    if final_response:
        print()
        print("=" * 60)
        print("Agent 响应:")
        print("=" * 60)
        print(final_response)
        print("=" * 60)
        print()
        output_file = output_dir / account_name / "推导日志" / current_trace_id / log_id / "result.txt"
        with open(output_file, 'w', encoding='utf-8') as f:
            f.write(final_response)
        print(f"✓ 结果已保存到: {output_file}")
        print()

    if current_trace_id:
        print("=" * 60)
        print("可视化 Step Tree:")
        print("=" * 60)
        print("1. 启动 API Server:")
        print("   python3 api_server.py")
        print()
        print("2. 浏览器访问:")
        print("   http://localhost:8000/api/traces")
        print()
        print(f"3. Trace ID: {current_trace_id}")
        print(f"4. Log ID（推导日志目录）: {log_id}")
        print("=" * 60)


if __name__ == "__main__":
    # anthropic/claude-sonnet-4.6
    # google/gemini-3-flash-preview
    asyncio.run(main(account_name="家有大志", post_id="68fb6a5c000000000302e5de"))