openclaw-voc-skill/tools/probe_douyin_media_from_search.js

#!/usr/bin/env node
const { DouyinApi } = require('../claude-code/claude-code-voc-intelligence/mcp/src/providers/douyin-api');

function parseArgs(argv) {
  const args = {};
  for (let i = 0; i < argv.length; i += 1) {
    const token = argv[i];
    if (!token.startsWith('--')) continue;
    const eq = token.indexOf('=');
    if (eq > 0) {
      args[token.slice(2, eq)] = token.slice(eq + 1);
      continue;
    }
    const key = token.slice(2);
    const next = argv[i + 1];
    if (next && !next.startsWith('--')) {
      args[key] = next;
      i += 1;
    } else {
      args[key] = true;
    }
  }
  return args;
}

function decodeMaybeBase64Url(value) {
  const text = String(value || '').trim();
  if (!text) return '';
  if (/^https?:\/\//i.test(text)) return text;
  if (!/^[A-Za-z0-9+/=_-]+$/.test(text) || text.length < 24) return '';
  try {
    const decoded = Buffer.from(text.replace(/-/g, '+').replace(/_/g, '/'), 'base64').toString('utf8');
    return /^https?:\/\//i.test(decoded) ? decoded : '';
  } catch {
    return '';
  }
}

function collectUrls(node, path = [], out = []) {
  if (!node) return out;
  if (Array.isArray(node)) {
    node.forEach((item, index) => collectUrls(item, [...path, String(index)], out));
    return out;
  }
  if (typeof node !== 'object') {
    if (typeof node === 'string') {
      const url = decodeMaybeBase64Url(node);
      if (url) out.push({ url, path: path.join('.') });
    }
    return out;
  }
  Object.entries(node).forEach(([key, value]) => collectUrls(value, [...path, key], out));
  return out;
}

function isMediaUrl(url) {
  const lower = String(url || '').toLowerCase();
  return lower.includes('douyinvod.com')
    || lower.includes('/aweme/v1/play')
    || lower.includes('mime_type=video')
    || /\.(mp4|m4a|mp3|aac|wav|webm)(?:\?|$)/i.test(lower);
}

function findAweme(node, awemeId, depth = 0) {
  if (!node || depth > 8) return null;
  if (Array.isArray(node)) {
    for (const item of node) {
      const found = findAweme(item, awemeId, depth + 1);
      if (found) return found;
    }
    return null;
  }
  if (typeof node !== 'object') return null;
  const id = String(node.aweme_id || node.id || '');
  if (id === String(awemeId)) return node;
  for (const value of Object.values(node)) {
    const found = findAweme(value, awemeId, depth + 1);
    if (found) return found;
  }
  return null;
}

async function main() {
  const args = parseArgs(process.argv.slice(2));
  const token = process.env.VOC_TOKEN || process.env.VOC_DOUYIN_TOKEN || process.env.DOUYIN_TOKEN;
  if (!token) throw new Error('VOC_TOKEN is required in environment');
  if (!args.keyword) throw new Error('--keyword is required');

  const api = new DouyinApi({ token });
  const response = await api.searchVideos({
    keyword: args.keyword,
    cursor: Number(args.cursor || 0),
    sortType: args.sortType || '1',
    publishTime: args.publishTime || '180',
    filterDuration: args.filterDuration || '0',
    contentType: args.contentType || '1',
  });

  const target = args.awemeId ? findAweme(response.data, args.awemeId) : response.data;
  const urls = collectUrls(target || response.data)
    .filter(item => isMediaUrl(item.url))
    .filter((item, index, list) => list.findIndex(other => other.url === item.url) === index)
    .slice(0, 20);

  console.log(JSON.stringify({
    status: 'ok',
    foundTarget: Boolean(target),
    mediaUrlCount: urls.length,
    mediaCandidates: urls.map(item => ({
      path: item.path,
      urlPrefix: item.url.slice(0, 180),
    })),
  }, null, 2));
}

main().catch(error => {
  console.error(JSON.stringify({ status: 'error', message: error.message }, null, 2));
  process.exit(1);
});