mirror of
https://github.com/jxxghp/MoviePilot.git
synced 2026-10-31 12:24:39 +08:00
6d4ce83837
严格检查全部宿主源码，清零类型错误预算并移除按模块类型豁免。 按原因、所属方法、错误码及语法指纹锁定必要兼容例外和站点类型桩， CI 阻断新增忽略、检查范围退化、本地桩替代真实接口及新的类型错误。 修复空值、异步和供应方类型边界，保持插件与 API 兼容合同， 并补充存储、订阅、通知、Agent 和工作流边界的回归覆盖。
107 lines
3.6 KiB
Python
107 lines
3.6 KiB
Python
"""
|
|
为工作流中的文件刮削媒体资料与图片。
|
|
"""
|
|
|
|
from app.chain.media import MediaChain
|
|
from app.chain.scraping import ScrapingChain
|
|
from app.chain.storage import StorageChain
|
|
from app.runtime.log import logger
|
|
from app.runtime.stop import runtime_stop_state
|
|
from app.schemas.file import FileItem
|
|
from app.schemas.workflow import ActionContext, ActionParams
|
|
from app.workflow.actions import BaseAction
|
|
|
|
|
|
class ScrapeFileParams(ActionParams):
|
|
"""
|
|
刮削文件参数
|
|
"""
|
|
pass
|
|
|
|
|
|
class ScrapeFileAction(BaseAction):
|
|
"""
|
|
刮削文件
|
|
"""
|
|
|
|
contract = {
|
|
"inputs": [{"name": "fileitems", "label": "文件", "kind": "list"}],
|
|
"outputs": [{"name": "fileitems", "label": "文件", "kind": "list"}],
|
|
}
|
|
|
|
def __init__(self, action_id: str) -> None:
|
|
"""
|
|
建立当前动作的已刮削文件集合与错误状态。
|
|
|
|
:param action_id: 当前工作流动作的标识
|
|
"""
|
|
super().__init__(action_id)
|
|
self._scraped_files: list[FileItem] = []
|
|
self._has_error = False
|
|
|
|
name = "刮削文件"
|
|
description = "刮削媒体信息和图片"
|
|
data = ScrapeFileParams().model_dump()
|
|
|
|
@property
|
|
def success(self) -> bool:
|
|
"""
|
|
返回文件刮削过程是否未记录错误。
|
|
|
|
:return: 未遇到刮削错误时为 True
|
|
"""
|
|
return not self._has_error
|
|
|
|
def execute(self, workflow_id: int, params: dict[str, object], context: ActionContext) -> ActionContext:
|
|
"""
|
|
刮削上下文中的文件并记录成功项。
|
|
|
|
:param workflow_id: 当前工作流标识，用于检查停止状态与缓存
|
|
:param params: 动作参数；此动作沿用上下文中的文件列表
|
|
:param context: 包含待刮削文件的工作流上下文
|
|
:return: 原工作流上下文，动作状态由成功数与失败数更新
|
|
"""
|
|
# 失败次数
|
|
_failed_count = 0
|
|
for fileitem in context.fileitems or []:
|
|
if runtime_stop_state.is_workflow_stopped(workflow_id):
|
|
break
|
|
if fileitem in self._scraped_files:
|
|
continue
|
|
if not StorageChain().exists(fileitem):
|
|
continue
|
|
# 检查缓存
|
|
cache_key = f"{fileitem.path}"
|
|
if self.check_cache(workflow_id, cache_key):
|
|
logger.info(f"{fileitem.path} 已刮削过，跳过")
|
|
continue
|
|
mediachain = MediaChain()
|
|
media_context = mediachain.recognize_by_path(
|
|
fileitem.path,
|
|
obtain_images=True,
|
|
)
|
|
if not media_context or not media_context.media_info:
|
|
_failed_count += 1
|
|
logger.info(f"{fileitem.path} 未识别到媒体信息，无法刮削")
|
|
continue
|
|
scrape_result = ScrapingChain().scrape_metadata(
|
|
fileitem=fileitem,
|
|
meta=media_context.meta_info,
|
|
mediainfo=media_context.media_info
|
|
)
|
|
if isinstance(scrape_result, tuple) and not scrape_result[0]:
|
|
_failed_count += 1
|
|
logger.info(
|
|
f"{fileitem.path} 刮削失败：{scrape_result[1] or '未知错误'}"
|
|
)
|
|
continue
|
|
self._scraped_files.append(fileitem)
|
|
# 保存缓存
|
|
self.save_cache(workflow_id, cache_key)
|
|
|
|
if not self._scraped_files and _failed_count:
|
|
self._has_error = True
|
|
|
|
self.job_done(f"成功刮削 {len(self._scraped_files)} 个文件，失败 {_failed_count} 个")
|
|
return context
|