Implement full media crawler workflow with Flask backend and Vue frontend.
Add TMDB search and media detail pages, HDHive resource ingestion flow, unified error handling, Docker single-container runtime, and project docs/config updates for local deployment. Co-authored-by: Cursor <cursoragent@cursor.com>
This commit is contained in:
renjue
0
backend/services/__init__.py
Normal file
0
backend/services/__init__.py
Normal file
73
backend/services/media_service.py
Normal file
73
backend/services/media_service.py
Normal file
@@ -0,0 +1,73 @@
|
||||
from adapters.hdhive_adapter import normalize_resource, search_resource, unlock_link
|
||||
from adapters.tmdb_adapter import get_media_detail, search_media
|
||||
from error_handling import AppServiceError
|
||||
|
||||
|
||||
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)
|
||||
hdhive = search_resource(media_type, tmdb_id)
|
||||
search_data = hdhive.get("data") or []
|
||||
if isinstance(search_data, dict):
|
||||
search_data = search_data.get("items") or []
|
||||
|
||||
resources = []
|
||||
for item in search_data:
|
||||
slug = (item or {}).get("slug")
|
||||
unlock_data = {}
|
||||
unlock_error = None
|
||||
if slug:
|
||||
try:
|
||||
unlock = unlock_link(slug)
|
||||
unlock_data = unlock.get("data") or {}
|
||||
except Exception as error:
|
||||
unlock_error = str(error)
|
||||
normalized = normalize_resource(item, unlock_data)
|
||||
normalized["unlockError"] = unlock_error
|
||||
resources.append(normalized)
|
||||
|
||||
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",
|
||||
)
|
||||
176
backend/services/orchestrator.py
Normal file
176
backend/services/orchestrator.py
Normal file
@@ -0,0 +1,176 @@
|
||||
import time
|
||||
from datetime import datetime
|
||||
|
||||
from adapters.cms_adapter import create_resource
|
||||
from adapters.emby_adapter import exists_by_tmdb_id
|
||||
from adapters.hdhive_adapter import normalize_resource, search_resource, unlock_link
|
||||
from adapters.tmdb_adapter import get_media_detail
|
||||
from error_handling import AppServiceError, normalize_exception
|
||||
from storage import find_media_item, insert_log, upsert_media_item, upsert_task
|
||||
|
||||
|
||||
def now_iso():
|
||||
return datetime.utcnow().isoformat() + "Z"
|
||||
|
||||
|
||||
def new_task_id():
|
||||
return f"task_{int(time.time() * 1000)}"
|
||||
|
||||
|
||||
def log(task_id, step, level, message, detail=None):
|
||||
insert_log(task_id, step, level, message, detail or {}, now_iso())
|
||||
|
||||
|
||||
def run_ingest_task(payload):
|
||||
task_id = new_task_id()
|
||||
trace_id = f"{task_id}_{hex(int(time.time() * 1000))[-6:]}"
|
||||
task = {
|
||||
"taskId": task_id,
|
||||
"traceId": trace_id,
|
||||
"status": "RUNNING",
|
||||
"inputPayload": payload,
|
||||
"startedAt": now_iso(),
|
||||
"finishedAt": None,
|
||||
"summary": None,
|
||||
}
|
||||
upsert_task(task)
|
||||
log(task_id, "START", "INFO", "任务开始", {"payload": payload, "traceId": trace_id})
|
||||
|
||||
try:
|
||||
deduped = find_media_item(payload["tmdbId"])
|
||||
if deduped and deduped.get("ingest_status") == "SUCCESS":
|
||||
task["status"] = "SUCCESS"
|
||||
task["finishedAt"] = now_iso()
|
||||
task["summary"] = {
|
||||
"result": "SKIPPED_ALREADY_EXISTS",
|
||||
"tmdbId": payload["tmdbId"],
|
||||
"cmsId": deduped.get("cms_id"),
|
||||
}
|
||||
upsert_task(task)
|
||||
log(task_id, "DEDUPE", "INFO", "命中本地幂等，跳过", task["summary"])
|
||||
return task
|
||||
|
||||
tmdb_result = get_media_detail(payload["tmdbId"], payload["type"])
|
||||
log(task_id, "TMDB_DETAIL", "INFO", "TMDB 元数据获取成功", {"status": tmdb_result["status"]})
|
||||
|
||||
hdhive_search = search_resource(payload["type"], payload["tmdbId"])
|
||||
hdhive_first = None
|
||||
search_data = hdhive_search.get("data") or {}
|
||||
preferred_slug = str(payload.get("slug") or "").strip()
|
||||
if isinstance(search_data, list) and search_data:
|
||||
if preferred_slug:
|
||||
hdhive_first = next(
|
||||
(item for item in search_data if (item or {}).get("slug") == preferred_slug),
|
||||
None,
|
||||
)
|
||||
if not hdhive_first:
|
||||
hdhive_first = search_data[0]
|
||||
elif isinstance(search_data, dict):
|
||||
items = search_data.get("items") or []
|
||||
if preferred_slug:
|
||||
hdhive_first = next(
|
||||
(item for item in items if (item or {}).get("slug") == preferred_slug),
|
||||
None,
|
||||
)
|
||||
if items:
|
||||
hdhive_first = hdhive_first or items[0]
|
||||
if not hdhive_first:
|
||||
raise AppServiceError(
|
||||
"HDHIVE 未检索到可用资源",
|
||||
category="not_found",
|
||||
code="HDHIVE_RESOURCE_NOT_FOUND",
|
||||
provider="hdhive",
|
||||
)
|
||||
log(task_id, "HDHIVE_SEARCH", "INFO", "HDHIVE 检索成功")
|
||||
|
||||
slug = hdhive_first.get("slug")
|
||||
if not slug:
|
||||
raise AppServiceError(
|
||||
"HDHIVE 返回资源缺少 slug，无法解锁",
|
||||
category="validation",
|
||||
code="HDHIVE_INVALID_RESOURCE",
|
||||
provider="hdhive",
|
||||
)
|
||||
hdhive_unlock = unlock_link(slug)
|
||||
normalized_resource = normalize_resource(hdhive_first, hdhive_unlock.get("data"))
|
||||
log(task_id, "HDHIVE_UNLOCK", "INFO", "HDHIVE 解锁成功", {"unlockUrl": normalized_resource["unlockUrl"]})
|
||||
|
||||
emby_exists = exists_by_tmdb_id(payload["tmdbId"])
|
||||
log(task_id, "EMBY_EXISTS", "INFO", "Emby 查询完成", {"exists": emby_exists["exists"]})
|
||||
if emby_exists["exists"]:
|
||||
upsert_media_item(
|
||||
{
|
||||
"tmdbId": payload["tmdbId"],
|
||||
"type": payload["type"],
|
||||
"title": tmdb_result["normalized"]["title"],
|
||||
"year": tmdb_result["normalized"]["year"],
|
||||
"tmdbRaw": tmdb_result["data"],
|
||||
"hdhiveRaw": hdhive_search["data"],
|
||||
"cmsId": None,
|
||||
"ingestStatus": "SKIPPED_ALREADY_EXISTS",
|
||||
}
|
||||
)
|
||||
task["status"] = "SUCCESS"
|
||||
task["finishedAt"] = now_iso()
|
||||
task["summary"] = {"result": "SKIPPED_ALREADY_EXISTS", "source": "EMBY"}
|
||||
upsert_task(task)
|
||||
return task
|
||||
|
||||
cms_payload = {
|
||||
"tmdbId": payload["tmdbId"],
|
||||
"mediaType": payload["type"],
|
||||
"title": tmdb_result["normalized"]["title"],
|
||||
"originalTitle": tmdb_result["normalized"]["originalTitle"],
|
||||
"year": tmdb_result["normalized"]["year"],
|
||||
"overview": tmdb_result["normalized"]["overview"],
|
||||
"posterPath": tmdb_result["normalized"]["posterPath"],
|
||||
"rating": tmdb_result["normalized"]["rating"],
|
||||
"genres": tmdb_result["normalized"]["genres"],
|
||||
"resource": {
|
||||
"title": normalized_resource["resourceTitle"],
|
||||
"quality": normalized_resource["quality"],
|
||||
"size": normalized_resource["size"],
|
||||
"diskType": normalized_resource["diskType"],
|
||||
"slug": normalized_resource["slug"],
|
||||
"source": normalized_resource["source"],
|
||||
"subtitleLanguage": normalized_resource["subtitleLanguage"],
|
||||
"unlockUrl": normalized_resource["unlockUrl"],
|
||||
},
|
||||
"traceId": trace_id,
|
||||
}
|
||||
cms_result = create_resource(cms_payload)
|
||||
cms_data = cms_result.get("data") or {}
|
||||
cms_id = (
|
||||
cms_data.get("id")
|
||||
or cms_data.get("resourceId")
|
||||
or (cms_data.get("data") or {}).get("id")
|
||||
or (cms_data.get("data") or {}).get("resourceId")
|
||||
or normalized_resource["slug"]
|
||||
)
|
||||
log(task_id, "CMS_CREATE", "INFO", "CMS 入库成功", {"cmsId": cms_id})
|
||||
|
||||
upsert_media_item(
|
||||
{
|
||||
"tmdbId": payload["tmdbId"],
|
||||
"type": payload["type"],
|
||||
"title": tmdb_result["normalized"]["title"],
|
||||
"year": tmdb_result["normalized"]["year"],
|
||||
"tmdbRaw": tmdb_result["data"],
|
||||
"hdhiveRaw": hdhive_search["data"],
|
||||
"cmsId": cms_id,
|
||||
"ingestStatus": "SUCCESS",
|
||||
}
|
||||
)
|
||||
task["status"] = "SUCCESS"
|
||||
task["finishedAt"] = now_iso()
|
||||
task["summary"] = {"result": "CREATED", "cmsId": cms_id}
|
||||
upsert_task(task)
|
||||
return task
|
||||
except Exception as error:
|
||||
normalized_error = normalize_exception(error)
|
||||
log(task_id, "FAILED", "ERROR", str(normalized_error), normalized_error.to_dict())
|
||||
task["status"] = "FAILED"
|
||||
task["finishedAt"] = now_iso()
|
||||
task["summary"] = {"error": normalized_error.to_dict()}
|
||||
upsert_task(task)
|
||||
return task
|
||||
Reference in New Issue
Block a user