feat: align with upstream 1.5.9.9
This commit is contained in:
@@ -0,0 +1,300 @@
|
||||
/**
|
||||
* 歌词数据适配器
|
||||
* 将现有的歌词数据格式转换为 WordByWordLyrics 组件所需的 AMLL 格式
|
||||
*/
|
||||
|
||||
import type { ParsedLyricLine as ExistingLyricLine } from '~/composables/useLyrics'
|
||||
|
||||
// AMLL 格式的接口定义
|
||||
export interface LyricWord {
|
||||
startTime: number
|
||||
endTime: number
|
||||
word: string
|
||||
romanWord: string
|
||||
obscene: boolean
|
||||
}
|
||||
|
||||
export interface AdapterLyricLine {
|
||||
words: LyricWord[]
|
||||
translatedLyric: string
|
||||
romanLyric: string
|
||||
startTime: number
|
||||
endTime: number
|
||||
isBG: boolean
|
||||
isDuet: boolean
|
||||
}
|
||||
|
||||
/**
|
||||
* 将现有的歌词数据转换为 AMLL 格式
|
||||
* @param existingLyrics 现有的歌词数据
|
||||
* @param translationLyrics 翻译歌词数据
|
||||
* @param options 转换选项
|
||||
*/
|
||||
export function convertToAmllFormat(
|
||||
existingLyrics: ExistingLyricLine[],
|
||||
translationLyrics: ExistingLyricLine[] = [],
|
||||
options: {
|
||||
enableWordByWord?: boolean
|
||||
defaultDuration?: number
|
||||
romanization?: boolean
|
||||
} = {}
|
||||
): AdapterLyricLine[] {
|
||||
const { enableWordByWord = true, defaultDuration = 1000, romanization = false } = options
|
||||
|
||||
return existingLyrics.map((line, index) => {
|
||||
// 计算行的结束时间
|
||||
const nextLine = existingLyrics[index + 1]
|
||||
const endTime = nextLine ? nextLine.time : line.time + defaultDuration
|
||||
|
||||
// 查找对应的翻译歌词
|
||||
const translation = findMatchingTranslation(line, translationLyrics)
|
||||
|
||||
// 处理逐字歌词
|
||||
const words = processWords(line, endTime, enableWordByWord)
|
||||
|
||||
return {
|
||||
words,
|
||||
translatedLyric:
|
||||
translation &&
|
||||
translation.content &&
|
||||
translation.content.trim() &&
|
||||
translation.content !== '//'
|
||||
? translation.content
|
||||
: '',
|
||||
romanLyric: romanization ? generateRomanization(line.content) : '',
|
||||
startTime: line.time,
|
||||
endTime,
|
||||
isBG: false, // 可以根据需要扩展
|
||||
isDuet: false // 可以根据需要扩展
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
/**
|
||||
* 处理单行歌词的词汇数据
|
||||
*/
|
||||
function processWords(
|
||||
line: ExistingLyricLine,
|
||||
lineEndTime: number,
|
||||
enableWordByWord: boolean
|
||||
): LyricWord[] {
|
||||
// 如果有逐字歌词数据，直接使用
|
||||
if (enableWordByWord && line.words && line.words.length > 0) {
|
||||
return line.words.map((word) => ({
|
||||
startTime: word.time,
|
||||
endTime: word.time + word.duration,
|
||||
word: word.content,
|
||||
romanWord: '',
|
||||
obscene: false
|
||||
}))
|
||||
}
|
||||
|
||||
// 如果没有逐字歌词，将整行作为一个词处理
|
||||
if (line.content.trim()) {
|
||||
return [
|
||||
{
|
||||
startTime: line.time,
|
||||
endTime: lineEndTime,
|
||||
word: line.content,
|
||||
romanWord: '',
|
||||
obscene: false
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
return []
|
||||
}
|
||||
|
||||
/**
|
||||
* 查找匹配的翻译歌词
|
||||
*/
|
||||
function findMatchingTranslation(
|
||||
line: ExistingLyricLine,
|
||||
translations: ExistingLyricLine[]
|
||||
): ExistingLyricLine | null {
|
||||
if (!translations.length) return null
|
||||
|
||||
// 查找时间最接近的翻译
|
||||
let bestMatch: ExistingLyricLine | null = null
|
||||
let minTimeDiff = Infinity
|
||||
|
||||
for (const translation of translations) {
|
||||
const timeDiff = Math.abs(translation.time - line.time)
|
||||
if (timeDiff < minTimeDiff) {
|
||||
minTimeDiff = timeDiff
|
||||
bestMatch = translation
|
||||
}
|
||||
}
|
||||
|
||||
// 如果时间差超过5秒，认为不匹配
|
||||
return minTimeDiff <= 5000 ? bestMatch : null
|
||||
}
|
||||
|
||||
/**
|
||||
* 生成音译（简单实现，可以根据需要扩展）
|
||||
*/
|
||||
function generateRomanization(text: string): string {
|
||||
// 这里可以集成更复杂的音译逻辑
|
||||
// 目前返回空字符串，可以根据需要实现
|
||||
return ''
|
||||
}
|
||||
|
||||
/**
|
||||
* 智能分词处理（用于没有逐字歌词的情况）
|
||||
*/
|
||||
export function smartWordSegmentation(
|
||||
text: string,
|
||||
startTime: number,
|
||||
endTime: number,
|
||||
options: {
|
||||
minWordDuration?: number
|
||||
maxWordDuration?: number
|
||||
} = {}
|
||||
): LyricWord[] {
|
||||
const { minWordDuration = 200, maxWordDuration = 2000 } = options
|
||||
|
||||
if (!text.trim()) return []
|
||||
|
||||
// 分词逻辑：
|
||||
// 1. 对于英文，按空格分割单词
|
||||
// 2. 对于中文，只在句末标点符号处分割，避免拆分歌名中的数字和符号
|
||||
|
||||
// 检测是否包含英文字母
|
||||
const hasEnglish = /[a-zA-Z]/.test(text)
|
||||
|
||||
let segments: string[]
|
||||
|
||||
if (hasEnglish) {
|
||||
// 英文内容：按空格和句末标点符号分割
|
||||
segments = text.split(/(\s+|[.!?。！？])/g).filter((s) => s.trim())
|
||||
} else {
|
||||
// 中文内容：只按句末标点符号分割，避免拆分歌名中的数字和符号
|
||||
segments = text.split(/([。！？])/g).filter((s) => s.trim() && !/^[。！？\s]+$/.test(s))
|
||||
}
|
||||
|
||||
// 如果没有明显的分割点，或者分割后只有一个片段，将整行作为一个词处理
|
||||
if (segments.length <= 1) {
|
||||
return [
|
||||
{
|
||||
startTime,
|
||||
endTime,
|
||||
word: text.trim(),
|
||||
romanWord: '',
|
||||
obscene: false
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
const totalDuration = endTime - startTime
|
||||
const avgDuration = Math.max(
|
||||
minWordDuration,
|
||||
Math.min(maxWordDuration, totalDuration / segments.length)
|
||||
)
|
||||
|
||||
return segments.map((segment, index) => {
|
||||
const wordStartTime = startTime + index * avgDuration
|
||||
const wordEndTime = Math.min(wordStartTime + avgDuration, endTime)
|
||||
|
||||
return {
|
||||
startTime: wordStartTime,
|
||||
endTime: wordEndTime,
|
||||
word: segment.trim(),
|
||||
romanWord: '',
|
||||
obscene: false
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
/**
|
||||
* 检测和标记背景歌词
|
||||
*/
|
||||
export function detectBackgroundLyrics(lyrics: AdapterLyricLine[]): AdapterLyricLine[] {
|
||||
return lyrics.map((line) => {
|
||||
// 简单的背景歌词检测逻辑
|
||||
const isBG =
|
||||
line.words.length === 0 ||
|
||||
line.words.every((word) => word.word.trim().length === 0) ||
|
||||
(line.translatedLyric.includes('(') && line.translatedLyric.includes(')'))
|
||||
|
||||
return {
|
||||
...line,
|
||||
isBG
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
/**
|
||||
* 检测和标记对唱歌词
|
||||
*/
|
||||
export function detectDuetLyrics(lyrics: AdapterLyricLine[]): AdapterLyricLine[] {
|
||||
return lyrics.map((line) => {
|
||||
// 简单的对唱检测逻辑
|
||||
const isDuet =
|
||||
line.translatedLyric.includes('(对唱)') ||
|
||||
line.translatedLyric.includes('(合唱)') ||
|
||||
line.words.some((word) => word.word.includes('(') && word.word.includes(')'))
|
||||
|
||||
return {
|
||||
...line,
|
||||
isDuet
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
/**
|
||||
* 完整的歌词数据转换流程
|
||||
*/
|
||||
export function fullLyricConversion(
|
||||
existingLyrics: ExistingLyricLine[],
|
||||
translationLyrics: ExistingLyricLine[] = [],
|
||||
options: {
|
||||
enableWordByWord?: boolean
|
||||
enableSmartSegmentation?: boolean
|
||||
enableBackgroundDetection?: boolean
|
||||
enableDuetDetection?: boolean
|
||||
romanization?: boolean
|
||||
} = {}
|
||||
): AdapterLyricLine[] {
|
||||
const {
|
||||
enableWordByWord = true,
|
||||
enableSmartSegmentation = true,
|
||||
enableBackgroundDetection = true,
|
||||
enableDuetDetection = true,
|
||||
romanization = false
|
||||
} = options
|
||||
|
||||
// 基础转换
|
||||
let convertedLyrics = convertToAmllFormat(existingLyrics, translationLyrics, {
|
||||
enableWordByWord,
|
||||
romanization
|
||||
})
|
||||
|
||||
// 智能分词（对于没有逐字歌词的行）
|
||||
if (enableSmartSegmentation) {
|
||||
convertedLyrics = convertedLyrics.map((line) => {
|
||||
if (line.words.length === 1 && line.words[0].word === line.words[0].word.trim()) {
|
||||
const smartWords = smartWordSegmentation(line.words[0].word, line.startTime, line.endTime)
|
||||
|
||||
if (smartWords.length > 1) {
|
||||
return {
|
||||
...line,
|
||||
words: smartWords
|
||||
}
|
||||
}
|
||||
}
|
||||
return line
|
||||
})
|
||||
}
|
||||
|
||||
// 背景歌词检测
|
||||
if (enableBackgroundDetection) {
|
||||
convertedLyrics = detectBackgroundLyrics(convertedLyrics)
|
||||
}
|
||||
|
||||
// 对唱歌词检测
|
||||
if (enableDuetDetection) {
|
||||
convertedLyrics = detectDuetLyrics(convertedLyrics)
|
||||
}
|
||||
|
||||
return convertedLyrics
|
||||
}
|
||||
Reference in New Issue
Block a user