media_crawler/backend/services/media_service.py

import json
import time

from adapters.hdhive_adapter import normalize_resource, search_resource
from adapters.tmdb_adapter import get_media_detail, search_media
from error_handling import AppServiceError, normalize_exception
from storage import find_media_item

RESOURCE_CACHE_TTL_SECONDS = 15 * 60
_RESOURCE_CACHE = {}


def _extract_hdhive_items(hdhive_result):
    payload = (hdhive_result or {}).get("data")
    # HDHive openapi uses envelope: { success, data, ... }
    if isinstance(payload, dict) and isinstance(payload.get("data"), list):
        return payload.get("data") or []
    if isinstance(payload, list):
        return payload
    if isinstance(payload, dict):
        return payload.get("items") or []
    return []


def _resource_cache_key(media_type, tmdb_id):
    return f"{media_type}:{tmdb_id}"


def _get_cached_resources(media_type, tmdb_id):
    key = _resource_cache_key(media_type, tmdb_id)
    cached = _RESOURCE_CACHE.get(key)
    if not cached:
        return None
    if cached.get("expireAt", 0) < time.time():
        _RESOURCE_CACHE.pop(key, None)
        return None
    return cached.get("resources") or []


def _set_cached_resources(media_type, tmdb_id, resources):
    key = _resource_cache_key(media_type, tmdb_id)
    _RESOURCE_CACHE[key] = {
        "expireAt": time.time() + RESOURCE_CACHE_TTL_SECONDS,
        "resources": resources or [],
    }


def _extract_hdhive_raw_items(hdhive_raw):
    payload = hdhive_raw
    if isinstance(payload, str):
        try:
            payload = json.loads(payload)
        except Exception:
            return []
    if isinstance(payload, dict) and isinstance(payload.get("data"), list):
        return payload.get("data") or []
    if isinstance(payload, dict) and isinstance(payload.get("items"), list):
        return payload.get("items") or []
    if isinstance(payload, list):
        return payload
    return []


def _fallback_resources_from_db(tmdb_id, media_type):
    item = find_media_item(tmdb_id)
    if not item:
        return []
    if item.get("type") != media_type:
        return []

    raw_items = _extract_hdhive_raw_items(item.get("hdhive_raw"))
    resources = []
    for raw_item in raw_items:
        normalized = normalize_resource(raw_item, {})
        normalized["unlockError"] = "当前 HDHive 接口限流，已回退展示缓存资源（未实时解锁）"
        resources.append(normalized)
    return resources


def search_media_by_keyword(query, media_type):
    result = search_media(query, media_type)
    raw_items = result.get("items") or []
    items = []
    for item in raw_items:
        items.append(
            {
                "id": item.get("id"),
                "type": media_type,
                "title": item.get("title") or item.get("name"),
                "overview": item.get("overview") or "",
                "posterPath": item.get("poster_path") or "",
                "releaseDate": item.get("release_date") or item.get("first_air_date") or "",
                "voteAverage": item.get("vote_average"),
            }
        )
    return {"items": items}


def get_media_resources(media_type, tmdb_id):
    detail = get_media_detail(tmdb_id, media_type)
    cached_resources = _get_cached_resources(media_type, tmdb_id)
    if cached_resources:
        return {
            "media": detail.get("normalized"),
            "resources": cached_resources,
            "resourceNotice": "命中本地缓存，已减少上游请求压力",
        }

    try:
        hdhive = search_resource(media_type, tmdb_id)
    except Exception as error:
        normalized_error = normalize_exception(error)
        if normalized_error.provider == "hdhive" and normalized_error.category == "rate_limit":
            fallback_resources = _fallback_resources_from_db(tmdb_id, media_type)
            if fallback_resources:
                _set_cached_resources(media_type, tmdb_id, fallback_resources)
                return {
                    "media": detail.get("normalized"),
                    "resources": fallback_resources,
                    "resourceNotice": "HDHive 当前限流，已回退到历史缓存资源",
                }
        raise

    search_data = _extract_hdhive_items(hdhive)

    resources = []
    for item in search_data:
        normalized = normalize_resource(item, {})
        normalized["unlockError"] = "详情页不自动解锁，点击“使用该资源入库”时才会请求解锁并扣积分"
        resources.append(normalized)
    _set_cached_resources(media_type, tmdb_id, resources)

    return {
        "media": detail.get("normalized"),
        "resources": resources,
    }


def validate_media_query(query, media_type):
    if not query:
        raise AppServiceError(
            "query is required",
            category="validation",
            code="INVALID_INPUT",
            status=400,
            provider="api",
        )
    validate_media_type(media_type)


def validate_media_type(media_type):
    if media_type not in ("movie", "tv"):
        raise AppServiceError(
            "type must be movie or tv",
            category="validation",
            code="INVALID_INPUT",
            status=400,
            provider="api",
        )