opex-yt-id/index.js

Extracts YouTube video, channel, and playlist IDs from various URLs

/*
opex-yt-id/
├── index.js        # Основной код функций
├── index.d.ts      # Файл описания типов
├── LICENSE         # Файл лицензии (Assuming MIT, not included here)
├── package.json    # Файл манифеста
├── README.md       # Документация
└── test.js         # Файл тестов
*/

import { URL, URLSearchParams } from 'url';

// --- Validation Functions ---

function isValidYouTubeVideoId(id) {
  return typeof id === 'string' && /^[a-zA-Z0-9_-]{11}$/.test(id);
}

function isValidYouTubeChannelId(id) {
  // Standard channel IDs start with 'UC' and are 24 chars total.
  return typeof id === 'string' && /^UC[a-zA-Z0-9_-]{22}$/.test(id);
}

function isValidYouTubePlaylistId(id) {
  // Common playlist IDs start with 'PL' and vary in length.
  // Let's use a reasonable length range (e.g., 10-40 chars after 'PL').
  return typeof id === 'string' && /^PL[a-zA-Z0-9_-]{10,40}$/.test(id);
}

// --- Helper for URL Parsing and Basic Checks ---

function parseUrlAndBasicInfo(urlString) {
    if (!urlString || typeof urlString !== 'string' || urlString.trim() === '') {
        return null;
    }

    let url;
    let originalInput = urlString;
    let potentialUrlString = urlString;

    try {
        // Attempt to fix common non-standard URL inputs before parsing
        if (potentialUrlString.startsWith('//')) potentialUrlString = 'https:' + potentialUrlString;
        else if (!potentialUrlString.match(/^[a-zA-Z]+:/) && potentialUrlString.includes('.') && potentialUrlString.includes('/')) potentialUrlString = 'https:' + potentialUrlString;

        url = new URL(potentialUrlString);
        return {
            url,
            hostname: url.hostname.toLowerCase(),
            pathname: url.pathname,
            searchParams: url.searchParams,
            originalInput: originalInput
        };
    } catch (e) {
        // If URL parsing fails, return null but keep original input for regex checks
        return { originalInput: originalInput, error: true };
    }
}

// --- Core Extraction Functions ---

/**
 * Extracts YouTube Video ID.
 */
export function getYouTubeVideoId(urlString) {
    const parsed = parseUrlAndBasicInfo(urlString);
    if (!parsed) return null;

    const { url, hostname, pathname, searchParams, originalInput, error } = parsed;

    // Phase 1: Direct String/Regex Matching (before full URL parse or if parse failed)
    if (originalInput.trim().startsWith('<iframe')) {
        const iframeSrcMatch = originalInput.match(/src=["'](?:https?:)?\/\/(?:www\.)?(?:youtube\.com|youtube-nocookie\.com)\/embed\/([a-zA-Z0-9_-]{11})/);
        if (iframeSrcMatch && iframeSrcMatch[1] && isValidYouTubeVideoId(iframeSrcMatch[1])) return iframeSrcMatch[1];
    }
    if (originalInput.startsWith('youtube://')) {
        const schemeMatch = originalInput.match(/^youtube:\/\/(?:(?:watch\?v=|youtu\.be\/)?)([a-zA-Z0-9_-]{11})/);
        if (schemeMatch && schemeMatch[1] && isValidYouTubeVideoId(schemeMatch[1])) return schemeMatch[1];
    }
    const simplePattern = /(?:youtu\.be|y2u\.be)\/([a-zA-Z0-9_-]{11})|(?:v=|vi=|video_id=|docid=|embed\/|shorts\/|live\/|v\/|vi\/|e\/|user\/.+\/|attribution_link.*v(?:%3D|=))([a-zA-Z0-9_-]{11})|i\.ytimg\.com\/(?:vi|an_webp)\/([a-zA-Z0-9_-]{11})\//;
    const simpleMatch = originalInput.match(simplePattern);
    const potentialIdFromSimpleMatch = simpleMatch ? (simpleMatch[1] || simpleMatch[2] || simpleMatch[3]) : null;

    // If URL parsing failed, rely solely on regex match
    if (error) {
        return potentialIdFromSimpleMatch && isValidYouTubeVideoId(potentialIdFromSimpleMatch) ? potentialIdFromSimpleMatch : null;
    }

    // Phase 2: Full URL Parsing Logic
    let potentialId = null;

    // *** Removed Google Redirect Check ***

    // Check known domains and patterns
    if (hostname === 'youtu.be' || hostname === 'y2u.be') {
        potentialId = pathname.split('/')[1];
    } else if (hostname === 'i.ytimg.com') {
        const imgMatch = pathname.match(/^\/(?:vi|an_webp)\/([a-zA-Z0-9_-]{11})\//);
        if (imgMatch && imgMatch[1]) potentialId = imgMatch[1];
    } else {
        const officialDomains = ['youtube.com', 'youtube-nocookie.com', 'music.youtube.com', 'kids.youtube.com', 'youtubekids.com', 'gaming.youtube.com'];
        const thirdPartyDomains = ['ymusicapp.com', 'yewtu.be', 'vid.puffyan.us', 'invidious.snopyta.org', 'inv.riverside.rocks', 'invidious.io', 'piped.video', 'piped.kavin.rocks', 'piped.syncpundit.io']; // Example list
        const isOfficial = officialDomains.some(domain => hostname === domain || hostname.endsWith('.' + domain));
        const isKnownThirdParty = thirdPartyDomains.includes(hostname);

        if (isOfficial || isKnownThirdParty) {
            const pathMatch = pathname.match(/^\/(?:embed|v|vi|e|shorts|live)\/([a-zA-Z0-9_-]{11})/);
            if (pathMatch && pathMatch[1]) {
                potentialId = pathMatch[1];
            } else {
                potentialId = searchParams.get('v') || searchParams.get('vi') || searchParams.get('video_id') || searchParams.get('docid');
                if (!potentialId || !isValidYouTubeVideoId(potentialId)) { // Check validity early
                    potentialId = null; // Reset if invalid param found
                    if (pathname === '/attribution_link' && searchParams.has('u')) {
                        const innerUrlPath = searchParams.get('u'); // .get() decodes
                        if (innerUrlPath) {
                            try {
                                // No need to decode innerUrlPath again here either
                                const innerUrlToParse = innerUrlPath.startsWith('http') ? innerUrlPath : `https://www.youtube.com${innerUrlPath.startsWith('/') ? innerUrlPath : '/' + innerUrlPath}`;
                                potentialId = getYouTubeVideoId(innerUrlToParse); // Recursive call
                            } catch (recursiveError) {
                                // Fallback attempt if parsing the constructed URL fails
                                const rawParamMatch = innerUrlPath.match(/(?:v|vi)(?:%3D|=)([a-zA-Z0-9_-]{11})/); // Check raw string
                                if (rawParamMatch && rawParamMatch[1]) potentialId = rawParamMatch[1];
                            }
                        }
                    } else if (pathname.startsWith('/user/')) {
                        // Legacy user channel page fragment identifier for videos
                        const fragment = url.hash || (originalInput.includes('#') ? originalInput.substring(originalInput.indexOf('#')) : '');
                        const fragmentMatch = fragment.match(/#p\/(?:[a-z]\/)?u\/\d+\/([a-zA-Z0-9_-]{11})/);
                        if (fragmentMatch && fragmentMatch[1]) potentialId = fragmentMatch[1];
                    }
                }
            }
        }
    }

    // Phase 3: Generic Fallback for /watch?v={ID} on ANY domain
    if (!potentialId || !isValidYouTubeVideoId(potentialId)) {
        if (pathname === '/watch' && searchParams.has('v')) {
            const genericV = searchParams.get('v');
            if (isValidYouTubeVideoId(genericV)) potentialId = genericV;
        }
    }

    // Phase 4: Final Validation & Return
    if (potentialId && isValidYouTubeVideoId(potentialId)) return potentialId;
    // Fallback to simple regex match if URL parsing didn't yield a valid ID
    if (potentialIdFromSimpleMatch && isValidYouTubeVideoId(potentialIdFromSimpleMatch)) return potentialIdFromSimpleMatch;

    return null;
}

/**
 * Extracts YouTube Channel ID (starting with UC).
 * Does NOT resolve custom URLs (/c/), legacy usernames (/user/), or handles (/@/).
 */
export function getYouTubeChannelId(urlString) {
    const parsed = parseUrlAndBasicInfo(urlString);
    if (!parsed) return null;

    const { url, hostname, pathname, searchParams, originalInput, error } = parsed;
    let potentialId = null;

     // *** Removed Google Redirect Check ***

    // Regex for direct match in the original string (useful if URL parsing fails or for non-standard formats)
    const simpleMatch = originalInput.match(/(?:channel|c|user)\/(UC[a-zA-Z0-9_-]{22})/);
    const potentialIdFromSimpleMatch = simpleMatch ? simpleMatch[1] : null;

    if (error) {
        return potentialIdFromSimpleMatch && isValidYouTubeChannelId(potentialIdFromSimpleMatch) ? potentialIdFromSimpleMatch : null;
    }

    // Check path segment for /channel/UC...
    const pathMatch = pathname.match(/^\/channel\/(UC[a-zA-Z0-9_-]{22})/);
    if (pathMatch && pathMatch[1]) {
        potentialId = pathMatch[1];
    } else {
        // Check known third-party frontend patterns (example)
        const thirdPartyDomains = ['yewtu.be', 'vid.puffyan.us', 'invidious.snopyta.org', 'inv.riverside.rocks', 'invidious.io', 'piped.video', 'piped.kavin.rocks', 'piped.syncpundit.io'];
        if (thirdPartyDomains.includes(hostname)) {
             const thirdPartyPathMatch = pathname.match(/^\/channel\/(UC[a-zA-Z0-9_-]{22})/);
             if (thirdPartyPathMatch && thirdPartyPathMatch[1]) {
                 potentialId = thirdPartyPathMatch[1];
             }
        }
        // Note: We intentionally do NOT try to resolve /c/, /user/, or /@ handles here
    }

    // Final Validation & Return
    if (potentialId && isValidYouTubeChannelId(potentialId)) return potentialId;
    if (potentialIdFromSimpleMatch && isValidYouTubeChannelId(potentialIdFromSimpleMatch)) return potentialIdFromSimpleMatch;


    return null;
}

/**
 * Extracts YouTube Playlist ID (starting with PL).
 */
export function getYouTubePlaylistId(urlString) {
    const parsed = parseUrlAndBasicInfo(urlString);
    if (!parsed) return null;

    const { url, hostname, pathname, searchParams, originalInput, error } = parsed;
    let potentialId = null;

     // *** Removed Google Redirect Check ***

    // Regex for direct match in the original string
    const simpleMatch = originalInput.match(/[?&]list=(PL[a-zA-Z0-9_-]{10,40})/);
    const potentialIdFromSimpleMatch = simpleMatch ? simpleMatch[1] : null;

    if (error) {
        return potentialIdFromSimpleMatch && isValidYouTubePlaylistId(potentialIdFromSimpleMatch) ? potentialIdFromSimpleMatch : null;
    }

    // Check search parameters for 'list' parameter
    if (searchParams.has('list')) {
        potentialId = searchParams.get('list');
    } else if (pathname === '/playlist' && searchParams.has('list')) {
         // Some structures might have /playlist?list=...
         potentialId = searchParams.get('list');
    }

    // Final Validation & Return
    if (potentialId && isValidYouTubePlaylistId(potentialId)) return potentialId;
    if (potentialIdFromSimpleMatch && isValidYouTubePlaylistId(potentialIdFromSimpleMatch)) return potentialIdFromSimpleMatch;

    return null;
}