TTMusic/entry/src/main/ets/lyric/parse/LyricParser.ts

import { IParser } from './IParser';
import { Lyric } from '../bean/Lyric';
import { LyricLine } from '../bean/LyricLine';
import { LyricWord } from '../bean/LyricWord';

/**
 * The parser to parse the string array of a standard lyric file.
 */
export class LyricParser implements IParser {

    /**
     * Parse the string array to a Lyric.
     * @param src The lyric source, a string array.
     * @returns A lyric instance.
     */
    parse(src: Array<string>): Lyric {
        let lyricLines = new Array<LyricLine>()
        let title = ""
        let artist = ""
        let album = ""
        let by = ""
        let offset = 0
        const ignoredTags = [ 'id', 'hash', 'sign', 'qq', 'total','Outro','Intro']; // 定义需要忽略的标签

        // 首先检测是否是纯文本歌词（没有时间标签）
        let hasValidTimeTag = false;
        for (let i = 0; i < src.length; i++) {
            let line = src[i].trim();
            if (line.length > 0 && /^\[\d{2}:\d{2}\.\d{2,3}\]/.test(line)) {
                hasValidTimeTag = true;
                break;
            }
        }

        // 如果没有时间标签,作为纯文本歌词处理
        if (!hasValidTimeTag) {
            // printD("检测到纯文本歌词(无时间标签),不添加时间标签");
            return this.parsePlainTextLyric(src);
        }

        for (let i = 0; i < src.length; i++) {
            let line = src[i]
            if (line == "" || line == "\n" || line == "\r" || line == "\r\n" || line == "[Verse]" || line == "[Chorus]"
                || line == "[PreChorus]" || line == "[PreChorus]"||line == "[Bridge]") {
                // printW("the lyric line is empty, carriage return or line feed, line index= " + i)
                continue
            }

            // 检查是否是需要忽略的标签
            // 修改后的标签检测逻辑：只有在[]中的标签才忽略
            const shouldIgnore = ignoredTags.some(tag  => {
                const tagPattern = new RegExp(`\\[${tag}[^\\]]*\\]`, 'i'); // 匹配[tag...]格式，忽略大小写
                return tagPattern.test(line);
            });
            if (shouldIgnore) {
                // printW(`the lyric line contains ignored tag, line index= ${i}`);
                continue;
            }
            // 修改后的标签检测逻辑，修复英文歌词的时候，部分歌词没有显示出来。
            if (line.startsWith("[ti:"))  {
                title = this.parseIdTag(line);
            }
            else if (line.startsWith("[ar:"))  {
                artist = this.parseIdTag(line);
            }
            else if (line.startsWith("[al:"))  {
                album = this.parseIdTag(line);
            }
            else if (line.startsWith("[by:"))  {
                by = this.parseIdTag(line);
            }
            else if (line.startsWith("[offset:"))  {
                offset = Number.parseInt(this.parseIdTag(line));
            }
            else {

                // 处理双语歌词的特殊情况（英文行+中文行交替）
                if (this.isBilingualLyric(src,  i)) {
                    const englishLine = src[i];
                    const chineseLine = src[i+1];

                    // 解析英文逐字歌词
                    const { timeline, words } = this.parseWordByWordLine(englishLine,  offset);

                    // 获取中文文本（去掉时间戳）
                    const chineseText = chineseLine.replace(/^\[\d{2}:\d{2}\.\d{2}\]/,  '').trim();

                    // 创建双语歌词行（将中文作为附加文本）
                    const bilingualLine = new LyricLine('', timeline, -1, words);
                    bilingualLine.translation  = chineseText; // 新增translation字段存储翻译

                    lyricLines.push(bilingualLine);
                    i++; // 跳过已处理的中文行
                    continue;
                }

                // 新增逐字歌词解析逻辑[mm:ss.xx] <mm:ss.xx>
                if (this.isWordByWordLyric(line)) {
                    const { timeline, words } = this.parseWordByWordLine(line, offset);
                    lyricLines.push(new LyricLine('', timeline, -1, words))
                    continue;
                }

                    // 新增：逐字歌词[]检测方括号逐字歌词格式 [mm:ss.xxx] 文字
                if (this.isSquareBracketWordByWordLyric(line))  {
                    // console.info(`onecold 普通歌词处理中`)
                    const { timeline, words,translation  } = this.parseSquareBracketWordLine(line,  offset,src[i+1]);
                    // if (words.length  > 0) {
                    //     lyricLines.push(new  LyricLine('', timeline, -1, words))
                    // }
                    if (words.length  > 0) {
                        const lyricLine = new LyricLine('', timeline, -1, words, translation);
                        lyricLines.push(lyricLine);
                        if (translation) i++; // 跳过已处理的中文行
                    }
                    continue;
                } else {
                    // 原逻辑处理，但支持逐字歌词
                    // [00:00.10]画心 - 张靓颖
                    // [01:05.49][02:08.40]看不穿 是你失落的魂魄
                    let spr = line.split(']');
                    if (spr.length <= 1) {
                        continue
                    }
                    // parse text
                    let text = spr[spr.length-1]
                    // ... 原来的文本解析逻辑保持不变 ...
                    for (let i = 0; i < spr.length - 1; i++) {
                        let timeline = spr[i].replace("[", "");
                        let timeStamp = this.parseTimeline(timeline);
                        lyricLines.push(new LyricLine(text, timeStamp - offset, -1));
                    }
                }


            }
        }
        lyricLines.sort((l1, l2) => {
            return l1.beginTime - l2.beginTime
        })
        for (let i = 0;i < lyricLines.length; i++) {
            let lyricLine = lyricLines[i]
            if (i == lyricLines.length - 1) {
                lyricLine.nextTime = lyricLine.beginTime + 1000 - offset
            } else {
                let next = lyricLines[i+1]
                lyricLine.nextTime = next.beginTime
            }
        }
        // 为逐字歌词填充text（拼接所有歌词词）
        this.populateTextForWordLyrics(lyricLines);
        let result = new Lyric(artist, title, album, by, offset, lyricLines)
        return result
    }

    // 新增判断是否为双语歌词的方法
    private isBilingualLyric(lines: string[], currentIndex: number): boolean {
        if (currentIndex + 1 >= lines.length)  return false;

        const currentLine = lines[currentIndex];
        const nextLine = lines[currentIndex + 1];

        // 检查当前行是否是英文逐字歌词
        const isEnglishLine = this.isWordByWordLyric(currentLine)  &&
        /[a-zA-Z]/.test(currentLine);

        // 检查下一行是否是纯中文歌词（带时间戳但不含逐字标记）
        const isChineseLine = /^\[\d{2}:\d{2}\.\d{2}\]/.test(nextLine) &&
            !this.isWordByWordLyric(nextLine)  &&
        /[\u4e00-\u9fa5]/.test(nextLine);

        return isEnglishLine && isChineseLine;
    }

    // 检测方括号格式的逐字歌词 [00:00.000]文[00:01.000]字
    private isSquareBracketWordByWordLyric(line: string): boolean {
        return /\[\d{2}:\d{2}\.\d{2,3}\]\S/.test(line);
    }



    private parseSquareBracketWordLine(
        line: string,
        offset: number,
        nextLine?: string
    ): { timeline: number, words: LyricWord[], translation?: string } {
        const words: LyricWord[] = [];
        let firstTimeline = -1;

        // 解析逐字歌词（原有逻辑）
        const regex = /\[(\d{2}:\d{2}[.:]\d{2,5})\]([^\[]*)/g;
        let match;
        while ((match = regex.exec(line))  !== null) {
            const timeStr = match[1];
            const word = match[2].trim();
            if (!word) continue;

            const timeline = this.parseTimeline(timeStr)  - offset;
            if (firstTimeline < 0) firstTimeline = timeline;
            words.push(new  LyricWord(word, timeline, 0));
        }

        // 计算词持续时间
        for (let i = 0; i < words.length  - 1; i++) {
            words[i].duration = words[i + 1].startTime - words[i].startTime;
        }
        if (words.length  > 0 && words[words.length - 1].duration === 0) {
            words[words.length - 1].duration = 200;
        }

        // 双语支持
        if (nextLine && /^\[\d{2}:\d{2}\.\d{2,3}\]/.test(nextLine)) {
            // 提取两行时间戳（需完全一致）
            const currentLineTime = line.match(/^\[(\d{2}:\d{2}\.\d{2,3})\]/)?.[1];
            const nextLineTime = nextLine.match(/^\[(\d{2}:\d{2}\.\d{2,3})\]/)?.[1];

            if (currentLineTime && nextLineTime && currentLineTime === nextLineTime) {
                const chineseText = nextLine.replace(/^\[\d{2}:\d{2}\.\d{2,3}\]/,  '').trim();
                if (chineseText) { // 非空文本才作为翻译
                    return {
                        timeline: firstTimeline,
                        words,
                        translation: chineseText
                    };
                }
            }
        }

        return { timeline: firstTimeline, words };
    }

    /******************** 时间解析增强 ********************/
    private parseTimeline2(timeString: string): number {
        // 增强支持毫秒/厘秒解析
        const parts = timeString.split(':');
        const minutes = parseInt(parts[0], 10);

        const secondParts = parts[1].split('.');
        const seconds = parseInt(secondParts[0], 10);
        const fraction = parseInt(secondParts[1], 10);

        // 根据小数位长度判断时间精度
        const milliseconds = secondParts[1].length === 2 ?
            fraction * 10 : // 厘秒转毫秒 (01 -> 10ms)
            fraction;       // 毫秒直接使用

        return minutes * 60000 + seconds * 1000 + milliseconds;
    }

    // 判断是否是逐字歌词行
    private isWordByWordLyric(line: string): boolean {
        const hasBracketTimestamp = /$$\d{2}:\d{2}\.\d{2,3}$$\S/.test(line);
        const hasAngleTimestamp = /<\d{2}:\d{2}\.\d{2,3}>/.test(line);
        return hasBracketTimestamp || hasAngleTimestamp;
    }

    // 解析逐字歌词行
    private parseWordByWordLine(line: string, offset: number): { timeline: number, words: LyricWord[] } {
        const words: LyricWord[] = [];
        let firstTimeline = -1;
        const regex = /((?:<|$$)(\d{2}:\d{2}\.\d{2,3})(?:>|$$))([^<\[]*)/g;
        let match;

        while ((match = regex.exec(line)) !== null) {
            const [_, tag, timeStr, word] = match;
            if (word.trim() === '') continue;

            const timeline = this.parseTimeline(timeStr) - offset;
            if (firstTimeline < 0) firstTimeline = timeline;

            words.push(new LyricWord(word.trim(), timeline, 0));
        }

        // 设置每个词持续时间（下一个词开始时间-当前词开始时间）
        for (let i = 0; i < words.length - 1; i++) {
            words[i].duration = words[i + 1].startTime - words[i].startTime;
        }
        if (words.length > 0 && words[words.length - 1].duration === 0) {
            // 最后一个词持续200ms
            words[words.length - 1].duration = 200;
        }

        return { timeline: firstTimeline, words };
    }

    // 为逐字歌词拼接整行文本
    private populateTextForWordLyrics(lyricLines: LyricLine[]) {
        lyricLines.forEach(line => {
            if (line.words.length > 0) {
                line.text = line.words.map(word => word.word).join('');
            }
        });
    }

    private parseIdTag(line: string): string {
        let spr = line.split(":")
        let spr1 = spr[1]
        let result = spr1.replace("]", "")
        return result
    }
    //
    // private parseTimeline(timeString: string): number {
    //     // 00:00.50
    //     let timeStringList = timeString.split(':')
    //     let minuteString = timeStringList[0] //00
    //     let minute = Number.parseInt(minuteString)
    //     let secondStrings = timeStringList[1] //00.50
    //     let secondStringList = secondStrings.split(".")
    //     let secondString = secondStringList[0] //00
    //     let millionSecondString = secondStringList[1] //50
    //     let seconds = Number.parseInt(secondString)
    //     let millionSecond = Number.parseInt(millionSecondString)
    //     return minute * 60000 + seconds * 1000 + millionSecond // covert to million seconds
    // }

    private parseTimeline(timeString: string): number {
        // 支持多种分隔符：`.` 或 `:`
        const parts = timeString.split(/[:.]/);
        if (parts.length  < 2) {
            // printW(`Invalid timeline format: ${timeString}`);
            return 0;
        }

        // 解析分钟、秒、毫秒
        const minutes = parseInt(parts[0], 10) || 0;
        const seconds = parseInt(parts[1], 10) || 0;
        let milliseconds = 0;

        // 处理毫秒部分（可能为厘秒或毫秒）
        if (parts.length  > 2) {
            const fraction = parts[2];
            // 根据小数位长度判断精度：2位为厘秒（需*10），3位以上为毫秒
            milliseconds = fraction.length  <= 2 ?
                parseInt(fraction, 10) * 10 : // 厘秒转毫秒（如 "54" -> 540ms）
            parseInt(fraction.substring(0,  3), 10); // 截取前3位（如 "54001" -> 540ms）
        }

        return minutes * 60000 + seconds * 1000 + milliseconds;
    }

    /**
     * 解析纯文本歌词（没有时间标签的歌词）
     * @param src 歌词行数组
     * @returns Lyric 对象
     */
    private parsePlainTextLyric(src: Array<string>): Lyric {
        let lyricLines = new Array<LyricLine>()
        let title = ""
        let artist = ""
        let album = ""
        let by = ""
        let offset = 0

        for (let i = 0; i < src.length; i++) {
            let line = src[i].trim()

            // 跳过空行
            if (line.length === 0 || line === "\n" || line === "\r" || line === "\r\n") {
                continue
            }

            // 跳过元数据行（以 [ti:、[ar:、[al:、[by:、[offset: 等开头）
            if (line.startsWith('[ti:') || line.startsWith('[ar:') || line.startsWith('[al:') ||
                line.startsWith('[by:') || line.startsWith('[offset:') || line.startsWith('[length:') ||
                line.startsWith('[id:') || line.startsWith('[hash:') || line.startsWith('[sign:')) {
                // 解析元数据
                if (line.startsWith("[ti:")) {
                    title = this.parseIdTag(line);
                } else if (line.startsWith("[ar:")) {
                    artist = this.parseIdTag(line);
                } else if (line.startsWith("[al:")) {
                    album = this.parseIdTag(line);
                } else if (line.startsWith("[by:")) {
                    by = this.parseIdTag(line);
                } else if (line.startsWith("[offset:")) {
                    offset = Number.parseInt(this.parseIdTag(line));
                }
                continue
            }

            // 跳过标签行（如 [Verse]、[Chorus] 等）
            if (line === "[Verse]" || line === "[Chorus]" || line === "[PreChorus]" ||
                line === "[Bridge]" || line === "[Intro]" || line === "[Outro]") {
                continue
            }

            // 跳过其他以 [ 开头但不是时间标签的行
            if (line.startsWith('[') && !/^\[\d{2}:\d{2}\.\d{2,3}\]/.test(line)) {
                continue
            }

            // 不添加时间标签，使用 -1 表示纯文本歌词
            lyricLines.push(new LyricLine(line, -1, -1))
        }

        // printD(`纯文本歌词解析完成，共 ${lyricLines.length} 行歌词`)
        // 标记为纯文本歌词
        let result = new Lyric(artist, title, album, by, offset, lyricLines, true)
        return result
    }
}