FishServer/fish_api/app/services/action_watch.py

from __future__ import annotations

import asyncio
import json
import traceback
from pathlib import Path
from typing import Any, Dict, Set

from app.services import action as action_svc
from app.settings import Settings
from app.state import HealthSnapshot, app_state


def _state_path(settings: Settings) -> Path:
    if settings.action_watch_state_file is not None:
        return settings.action_watch_state_file
    assert settings.action_watch_dir is not None
    return settings.action_watch_dir / ".fishaction_watch_processed.json"


def load_processed(path: Path) -> Set[str]:
    if not path.is_file():
        return set()
    try:
        with open(path, encoding="utf-8") as f:
            data: Any = json.load(f)
        if isinstance(data, list):
            return set(str(x) for x in data)
        if isinstance(data, dict) and "processed" in data:
            return set(str(x) for x in data["processed"])
    except (json.JSONDecodeError, OSError):
        pass
    return set()


def save_processed(path: Path, processed: Set[str]) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    tmp = path.with_suffix(path.suffix + ".tmp")
    with open(tmp, "w", encoding="utf-8") as f:
        json.dump(sorted(processed), f, indent=0, ensure_ascii=False)
    tmp.replace(path)


def iter_mp4(watch_dir: Path, recursive: bool) -> list[Path]:
    if recursive:
        return sorted(
            p
            for p in watch_dir.rglob("*")
            if p.is_file() and p.suffix.lower() == ".mp4"
        )
    return sorted(
        p
        for p in watch_dir.iterdir()
        if p.is_file() and p.suffix.lower() == ".mp4"
    )


async def _run_inference_and_state(
    mp4: Path,
    settings: Settings,
    processed: Set[str],
    state_file: Path,
) -> None:
    key = str(mp4.resolve())
    if key in processed:
        return
    print(f"[action-watch] inference: {mp4}", flush=True)
    async with app_state.action_lock:
        app_state.action_status = "running"
        try:
            snap = await asyncio.to_thread(action_svc.run_full_action, mp4, settings)
            app_state.last_health = snap
            app_state.action_status = "idle"
            processed.add(key)
            if settings.action_watch_use_state_file:
                save_processed(state_file, processed)
            pred = (snap.raw_class_en or "").strip()
            print(f"[action-watch] done: {mp4.name} -> {pred}", flush=True)
        except Exception as e:
            app_state.last_health = HealthSnapshot(
                behavior_result="",
                health_result="",
                error=str(e),
            )
            app_state.action_status = "error"
            print(f"[action-watch] error on {mp4}: {e}", flush=True)
            traceback.print_exc()
            raise


async def watch_tick(
    settings: Settings,
    processed: Set[str],
    stability: Dict[str, tuple[int, int]],
    state_file: Path,
) -> bool:
    """处理一轮目录扫描；若处理了至少一个文件返回 True。"""
    assert settings.action_watch_dir is not None
    watch_dir = settings.action_watch_dir
    did = False
    seen_keys: Set[str] = set()
    for mp4 in iter_mp4(watch_dir, settings.action_watch_recursive):
        key = str(mp4.resolve())
        seen_keys.add(key)
        if key in processed:
            continue
        try:
            st = mp4.stat()
        except OSError:
            continue
        size = int(st.st_size)
        if size <= 0:
            stability.pop(key, None)
            continue
        last = stability.get(key)
        if last is None or last[0] != size:
            stability[key] = (size, 1)
        else:
            _, cnt = last
            stability[key] = (size, cnt + 1)
        _, cnt = stability[key]
        if cnt >= settings.action_watch_stable_polls:
            try:
                await _run_inference_and_state(mp4, settings, processed, state_file)
                stability.pop(key, None)
                did = True
            except Exception:
                stability[key] = (size, 1)
    for k in list(stability.keys()):
        if k not in seen_keys:
            del stability[k]
    return did


async def run_action_watch_loop(settings: Settings) -> None:
    assert settings.action_watch_dir is not None
    wd = settings.action_watch_dir
    if not wd.is_dir():
        print(f"[action-watch] skip: not a directory: {wd}", flush=True)
        return

    state_file = _state_path(settings)
    processed: Set[str] = (
        load_processed(state_file) if settings.action_watch_use_state_file else set()
    )
    stability: Dict[str, tuple[int, int]] = {}

    print(
        f"[action-watch] watching {wd} "
        f"(poll={settings.action_watch_poll_interval}s, "
        f"stable_polls={settings.action_watch_stable_polls}, "
        f"state={'on' if settings.action_watch_use_state_file else 'off'} "
        f"{state_file if settings.action_watch_use_state_file else ''})",
        flush=True,
    )

    while True:
        await watch_tick(settings, processed, stability, state_file)
        await asyncio.sleep(max(settings.action_watch_poll_interval, 0.1))