@rolme/ytscript/dist/services/transcript/index.js

A CLI tool to download YouTube transcripts and generate summaries

import { TranscriptError } from '../../errors.js';
function parseTimestamp(text) {
    const match = text.match(/<text start="([0-9.]+)" dur="([0-9.]+)"/);
    if (!match)
        return null;
    return {
        offset: parseFloat(match[1]),
        duration: parseFloat(match[2])
    };
}
function decodeHtmlEntities(text) {
    return text
        .replace(/&#39;/g, "'")
        .replace(/&quot;/g, '"')
        .replace(/&amp;/g, '&')
        .replace(/&lt;/g, '<')
        .replace(/&gt;/g, '>')
        .trim();
}
function parseXmlToSegments(xml) {
    const segments = [];
    const lines = xml.split('\n');
    for (const line of lines) {
        const timing = parseTimestamp(line);
        if (!timing)
            continue;
        const text = line.replace(/<\/?[^>]+(>|$)/g, '').trim(); // Remove XML tags
        if (text) {
            segments.push({
                text: decodeHtmlEntities(text),
                ...timing
            });
        }
    }
    return segments;
}
export async function getTranscript(videoInfo, options = {}) {
    try {
        const captions = videoInfo.player_response.captions;
        if (!captions || !captions.playerCaptionsTracklistRenderer) {
            throw new TranscriptError('No captions available for this video');
        }
        const captionTracks = captions.playerCaptionsTracklistRenderer.captionTracks;
        if (!captionTracks || captionTracks.length === 0) {
            throw new TranscriptError('No caption tracks found');
        }
        // Find the requested language or default to English
        const targetLang = options.lang || 'en';
        const caption = captionTracks.find(track => track.languageCode === targetLang);
        if (!caption) {
            throw new TranscriptError(`No captions found for language: ${targetLang}`);
        }
        // Download the caption track
        const response = await fetch(caption.baseUrl);
        if (!response.ok) {
            throw new TranscriptError(`Failed to download caption track: ${response.statusText}`);
        }
        const rawText = await response.text();
        let text = '';
        let segments = [];
        // Check if the response is XML by looking for transcript tags
        const isXml = rawText.includes('<transcript>') && rawText.includes('</transcript>');
        if (isXml) {
            // Try to parse as XML
            try {
                segments = parseXmlToSegments(rawText);
                text = segments.map(s => s.text).join('\n');
            }
            catch (e) {
                segments = [];
                text = '';
            }
        }
        else {
            // Invalid XML or plain text
            segments = [];
            text = '';
        }
        return {
            transcript: text,
            segments,
            videoId: videoInfo.videoDetails.videoId
        };
    }
    catch (error) {
        if (error instanceof TranscriptError) {
            throw error;
        }
        const errorMessage = error instanceof Error ? error.message : 'Unknown error occurred';
        throw new TranscriptError(`Failed to download transcript: ${errorMessage}`);
    }
}