mirror of
https://github.com/jxxghp/MoviePilot.git
synced 2026-10-31 04:14:38 +08:00
bcbf79f030
* fix(media): preserve recognition and rule mutation identities * test: cover typed classification facts with alias policies --------- Co-authored-by: jxxghp <jxxghp@gmail.com>
223 lines
8.4 KiB
Python
223 lines
8.4 KiB
Python
"""
|
||
读取 Kodi 风格媒体 NFO，并按标准媒体信息构造元数据文档和图片清单。
|
||
"""
|
||
|
||
import xml.etree.ElementTree as ET
|
||
from pathlib import Path
|
||
from typing import List, Optional
|
||
from urllib.parse import urlparse
|
||
from xml.dom import minidom
|
||
|
||
from app.domain.context import MediaInfo
|
||
from app.foundation.dom import DomUtils
|
||
from app.schemas.types import MediaSource, MediaType
|
||
|
||
|
||
class NfoReader:
|
||
"""
|
||
读取 Kodi 风格 NFO XML 中的元素值。
|
||
"""
|
||
|
||
def __init__(self, xml_file_path: Path) -> None:
|
||
"""
|
||
解析指定 NFO 文件并保存根元素。
|
||
"""
|
||
self.xml_file_path = xml_file_path
|
||
self.tree = ET.parse(xml_file_path)
|
||
self.root = self.tree.getroot()
|
||
|
||
def get_element_value(self, element_path: str) -> Optional[str]:
|
||
"""
|
||
返回首个匹配元素的文本。
|
||
"""
|
||
element = self.root.find(element_path)
|
||
return element.text if element is not None else None
|
||
|
||
def get_elements(self, element_path: str) -> List[ET.Element]:
|
||
"""
|
||
返回路径匹配的全部 XML 元素。
|
||
"""
|
||
return self.root.findall(element_path)
|
||
|
||
|
||
class MediaScraperHelper:
|
||
"""
|
||
基于统一媒体信息生成通用 NFO 与图片清单，供缺少专用刮削格式的数据源复用。
|
||
"""
|
||
|
||
@staticmethod
|
||
def _media_identity(
|
||
mediainfo: MediaInfo,
|
||
season: Optional[int] = None,
|
||
episode: Optional[int] = None,
|
||
) -> tuple[Optional[MediaSource], Optional[str]]:
|
||
"""
|
||
获取媒体信息中的来源与来源原生 ID。
|
||
|
||
:param mediainfo: 统一媒体信息
|
||
:param season: 单集所属季号；主媒体或季元数据时为空
|
||
:param episode: 单集集号；为空时使用主媒体身份
|
||
:return: 数据源名称与原生 ID；IMDb 单集缺少独立身份时 ID 为 None，不使用剧集主 ID
|
||
"""
|
||
if mediainfo.media_source == MediaSource.IMDb and season is not None and episode is not None:
|
||
for detail in mediainfo.season_info:
|
||
if detail.get("season_number") != season:
|
||
continue
|
||
for item in detail.get("episodes") or []:
|
||
if item.get("episode_number") == episode:
|
||
return MediaSource.IMDb, item.get("id")
|
||
return MediaSource.IMDb, None
|
||
return mediainfo.media_source, mediainfo.media_id
|
||
|
||
@staticmethod
|
||
def _image_extension(url: str) -> str:
|
||
"""
|
||
从图片 URL 中提取可用于本地文件名的扩展名。
|
||
|
||
:param url: 图片地址
|
||
:return: 图片扩展名，无法确定时返回 .jpg
|
||
"""
|
||
extension = Path(urlparse(url).path).suffix.lower()
|
||
return extension if extension in {".jpg", ".jpeg", ".png", ".webp"} else ".jpg"
|
||
|
||
@classmethod
|
||
def _append_common_nodes(
|
||
cls,
|
||
mediainfo: MediaInfo,
|
||
doc: minidom.Document,
|
||
root: minidom.Element,
|
||
season: Optional[int] = None,
|
||
episode: Optional[int] = None,
|
||
) -> None:
|
||
"""
|
||
向 NFO 根节点写入各媒体类型共享的标准字段。
|
||
|
||
:param mediainfo: 统一媒体信息
|
||
:param doc: XML 文档
|
||
:param root: NFO 根节点
|
||
:param season: 单集季号，用于选择来源原生身份
|
||
:param episode: 单集集号，不使用剧集主身份代替单集
|
||
"""
|
||
DomUtils.add_node(doc, root, "title", mediainfo.title or "")
|
||
DomUtils.add_node(doc, root, "originaltitle", mediainfo.original_title or "")
|
||
DomUtils.add_node(doc, root, "year", mediainfo.year or "")
|
||
DomUtils.add_node(doc, root, "premiered", mediainfo.release_date or "")
|
||
DomUtils.add_node(doc, root, "rating", mediainfo.vote_average or "0")
|
||
|
||
plot = DomUtils.add_node(doc, root, "plot")
|
||
plot.appendChild(doc.createCDATASection(mediainfo.overview or ""))
|
||
outline = DomUtils.add_node(doc, root, "outline")
|
||
outline.appendChild(doc.createCDATASection(mediainfo.overview or ""))
|
||
|
||
source, media_id = cls._media_identity(mediainfo, season=season, episode=episode)
|
||
if source and media_id:
|
||
unique_id = DomUtils.add_node(doc, root, "uniqueid", media_id)
|
||
unique_id.setAttribute("type", str(source))
|
||
unique_id.setAttribute("default", "true")
|
||
|
||
for genre in mediainfo.genres or []:
|
||
genre_name = genre.get("name") if isinstance(genre, dict) else str(genre)
|
||
if genre_name:
|
||
DomUtils.add_node(doc, root, "genre", genre_name)
|
||
|
||
for company in mediainfo.production_companies or []:
|
||
company_name = company.get("name") if isinstance(company, dict) else str(company)
|
||
if company_name:
|
||
DomUtils.add_node(doc, root, "studio", company_name)
|
||
|
||
for director in mediainfo.directors or []:
|
||
director_name = director.get("name") if isinstance(director, dict) else str(director)
|
||
if director_name:
|
||
DomUtils.add_node(doc, root, "director", director_name)
|
||
|
||
for actor in mediainfo.actors or []:
|
||
if not isinstance(actor, dict):
|
||
continue
|
||
actor_node = DomUtils.add_node(doc, root, "actor")
|
||
DomUtils.add_node(doc, actor_node, "name", actor.get("name") or "")
|
||
DomUtils.add_node(
|
||
doc,
|
||
actor_node,
|
||
"role",
|
||
actor.get("character") or actor.get("role") or "",
|
||
)
|
||
avatar = actor.get("avatar") or actor.get("images") or {}
|
||
if isinstance(avatar, dict):
|
||
DomUtils.add_node(
|
||
doc,
|
||
actor_node,
|
||
"thumb",
|
||
avatar.get("large") or avatar.get("medium") or avatar.get("normal") or "",
|
||
)
|
||
|
||
@classmethod
|
||
def get_metadata_nfo(
|
||
cls,
|
||
mediainfo: MediaInfo,
|
||
season: Optional[int] = None,
|
||
episode: Optional[int] = None,
|
||
) -> Optional[str]:
|
||
"""
|
||
根据统一媒体信息生成电影、剧集、季或单集 NFO。
|
||
|
||
:param mediainfo: 统一媒体信息
|
||
:param season: 季号
|
||
:param episode: 集号
|
||
:return: UTF-8 编码声明的 NFO XML 文本；媒体信息为空时返回 None
|
||
"""
|
||
if not mediainfo:
|
||
return None
|
||
|
||
doc = minidom.Document()
|
||
if mediainfo.type == MediaType.MOVIE:
|
||
root = DomUtils.add_node(doc, doc, "movie")
|
||
cls._append_common_nodes(mediainfo, doc, root)
|
||
elif season is not None and episode is not None:
|
||
root = DomUtils.add_node(doc, doc, "episodedetails")
|
||
cls._append_common_nodes(mediainfo, doc, root, season=season, episode=episode)
|
||
DomUtils.add_node(doc, root, "season", str(season))
|
||
DomUtils.add_node(doc, root, "episode", str(episode))
|
||
DomUtils.add_node(
|
||
doc,
|
||
root,
|
||
"showtitle",
|
||
mediainfo.title or "",
|
||
)
|
||
elif season is not None:
|
||
root = DomUtils.add_node(doc, doc, "season")
|
||
cls._append_common_nodes(mediainfo, doc, root)
|
||
DomUtils.add_node(doc, root, "seasonnumber", str(season))
|
||
else:
|
||
root = DomUtils.add_node(doc, doc, "tvshow")
|
||
cls._append_common_nodes(mediainfo, doc, root)
|
||
DomUtils.add_node(doc, root, "season", "-1")
|
||
DomUtils.add_node(doc, root, "episode", "-1")
|
||
|
||
return doc.toprettyxml(indent=" ", encoding="utf-8").decode("utf-8")
|
||
|
||
@classmethod
|
||
def get_metadata_img(
|
||
cls,
|
||
mediainfo: MediaInfo,
|
||
season: Optional[int] = None,
|
||
episode: Optional[int] = None,
|
||
) -> dict[str, str]:
|
||
"""
|
||
根据统一媒体信息生成主海报和背景图下载清单。
|
||
|
||
:param mediainfo: 统一媒体信息
|
||
:param season: 季号
|
||
:param episode: 集号
|
||
:return: 图片文件名与下载地址映射
|
||
"""
|
||
if not mediainfo or season is not None or episode is not None:
|
||
return {}
|
||
images = {}
|
||
if mediainfo.poster_path:
|
||
extension = cls._image_extension(mediainfo.poster_path)
|
||
images[f"poster{extension}"] = mediainfo.poster_path
|
||
if mediainfo.backdrop_path:
|
||
extension = cls._image_extension(mediainfo.backdrop_path)
|
||
images[f"backdrop{extension}"] = mediainfo.backdrop_path
|
||
return images
|