mirror of
https://github.com/jxxghp/MoviePilot.git
synced 2026-10-30 20:05:42 +08:00
107 lines
3.9 KiB
Python
107 lines
3.9 KiB
Python
"""
|
||
AMLL 歌词来源的录音身份核对规则。
|
||
"""
|
||
|
||
import re
|
||
import unicodedata
|
||
from typing import Any
|
||
|
||
from app.domain.context import MusicInfo
|
||
from app.domain.meta.metamusic import MetaMusic
|
||
|
||
|
||
def normalize_text(value: str) -> str:
|
||
"""
|
||
统一字形与大小写并移除标点，保留版本文字用于精确比较。
|
||
|
||
:param value: 待比较的标题、艺术家或专辑文本
|
||
:return: 规范化的匹配文本，保留下划线及文字、数字
|
||
"""
|
||
return re.sub(r"[^\w]", "", unicodedata.normalize("NFKC", value).casefold())
|
||
|
||
|
||
def normalize_isrc(value: str | None) -> str:
|
||
"""
|
||
去除空格和连字符并校验 ISRC 结构。
|
||
|
||
:param value: 原始录音标识，可为空
|
||
:return: 规范化的大写 ISRC；缺失或格式不完整时返回空字符串
|
||
"""
|
||
normalized = re.sub(r"[\s-]", "", value or "").upper()
|
||
return normalized if re.fullmatch(r"[A-Z]{2}[A-Z0-9]{3}\d{7}", normalized) else ""
|
||
|
||
|
||
def text_values(payload: dict[str, Any], key: str) -> list[str]:
|
||
"""
|
||
提取来源字段中有效的非空字符串列表。
|
||
|
||
:param payload: AMLL 来源响应
|
||
:param key: 需要读取的字段名称
|
||
:return: 去除首尾空白后的字符串列表；字段不是列表时为空
|
||
"""
|
||
values = payload.get(key)
|
||
if not isinstance(values, list):
|
||
return []
|
||
return [value.strip() for value in values if isinstance(value, str) and value.strip()]
|
||
|
||
|
||
def search_identity(music: MetaMusic | MusicInfo) -> tuple[str, list[str], str]:
|
||
"""
|
||
提取单曲搜索身份，保留独立的版本标记。
|
||
|
||
版本未出现在标题尾部括号内时补入标题，避免现场版等版本与原版混淆。
|
||
|
||
:param music: 识别线索或已确认的单曲信息
|
||
:return: 歌曲标题、艺术家列表和专辑名；艺术家缺失时尝试专辑艺术家
|
||
"""
|
||
title = str(music.title or "").strip()
|
||
version = str(music.version or "").strip()
|
||
suffix = re.search(r"[\[(（]([^\])）]+)[\])）]\s*$", title)
|
||
has_version = suffix is not None and normalize_text(suffix.group(1)) == normalize_text(version)
|
||
if version and not has_version:
|
||
title = f"{title} ({version})" if title else ""
|
||
artist_values = list(music.artists or [])
|
||
if not artist_values and music.album_artist:
|
||
artist_values.append(music.album_artist)
|
||
artists = [artist.strip() for artist in artist_values if artist.strip()]
|
||
return title, artists, str(music.album or "").strip()
|
||
|
||
|
||
def match_score(music: MetaMusic | MusicInfo, payload: dict[str, Any]) -> int:
|
||
"""
|
||
严格核对标题别名、艺术家集合与已知专辑。
|
||
|
||
AMLL 时间戳描述歌词位置，不作为音轨总时长参与比较。
|
||
|
||
:param music: 待查单曲信息
|
||
:param payload: 来源候选身份字段
|
||
:return: 不匹配为 0，标题和艺术家匹配为 90，同时匹配已知专辑为 95
|
||
"""
|
||
title, artists, album = search_identity(music)
|
||
if not title or not artists:
|
||
return 0
|
||
if normalize_text(title) not in {normalize_text(value) for value in text_values(payload, "musicNames")}:
|
||
return 0
|
||
expected_artists = {normalize_text(value) for value in artists}
|
||
source_artists = {normalize_text(value) for value in text_values(payload, "artistNames")}
|
||
if not expected_artists <= source_artists:
|
||
return 0
|
||
if album and normalize_text(album) not in {
|
||
normalize_text(value) for value in text_values(payload, "albumNames")
|
||
}:
|
||
return 0
|
||
return 95 if album else 90
|
||
|
||
|
||
def lyric_id(payload: dict[str, Any]) -> str:
|
||
"""
|
||
校验可安全传入原生 API 的歌词标识。
|
||
|
||
:param payload: 包含 id 的歌词来源记录
|
||
:return: 正整数标识的字符串；布尔值、非整数或不在 1 至 2**53-1 范围时为空字符串
|
||
"""
|
||
value = payload.get("id")
|
||
if isinstance(value, bool) or not isinstance(value, int) or not 0 < value < 2 ** 53:
|
||
return ""
|
||
return str(value)
|