tt-help-cli-ycl 1.3.0 → 1.3.2

package/src/lib/parser.js

@@ -1,47 +1,47 @@
-import { USER_SECTION_SIZE } from './constants.js';
-
-export function extractUserSection(html) {
-  const idx = html.indexOf('"uniqueId"');
-  if (idx < 0) return null;
-  return html.substring(idx, idx + USER_SECTION_SIZE);
-}
-
-export function parseUserSection(section) {
-  const data = {};
-
-  for (const key of ['uniqueId', 'uid', 'secUid']) {
-    const m = section.match(new RegExp(`"${key}":"([^"]*)`));
-    if (m) data[key] = m[1];
-  }
-
-  for (const key of ['nickname', 'signature']) {
-    const m = section.match(new RegExp(`"${key}":"((?:[^"\\\\]|\\\\.)*)"`, 'g'));
-    if (m) {
-      const raw = m[0].replace(`"${key}":"`, '').replace(/"$/, '');
-      data[key] = raw.replace(/\\n/g, '\n').replace(/\\\\/g, '\\');
-    }
-  }
-
-  for (const key of ['ttSeller', 'verified']) {
-    const m = section.match(new RegExp(`"${key}":\\s*(true|false)`));
-    data[key] = m ? m[1] === 'true' : undefined;
-  }
-
-  for (const key of ['followerCount', 'followingCount', 'heartCount', 'videoCount', 'diggCount']) {
-    const m = section.match(new RegExp(`"${key}":(\\d+)`));
-    if (m) data[key] = parseInt(m[1], 10);
-  }
-
-  const mt = section.match(/"createTime":(\d+)/);
-  if (mt) data.createTime = parseInt(mt[1], 10);
-
-  const ma = section.match(/"avatarLarger":"([^"]*)/);
-  if (ma) data.avatarLarger = ma[1].replace(/\\u002F/g, '/');
-
-  return data;
-}
-
-export function extractLocationCreated(html) {
-  const m = html.match(/"locationCreated":"([^"]*)/);
-  return m ? m[1] : null;
-}
+export const USER_SECTION_SIZE = 12000;
+
+export function extractUserSection(html) {
+  const idx = html.indexOf('"uniqueId"');
+  if (idx < 0) return null;
+  return html.substring(idx, idx + USER_SECTION_SIZE);
+}
+
+export function parseUserSection(section) {
+  const data = {};
+
+  for (const key of ['uniqueId', 'uid', 'secUid']) {
+    const m = section.match(new RegExp(`"${key}":"([^"]*)`));
+    if (m) data[key] = m[1];
+  }
+
+  for (const key of ['nickname', 'signature']) {
+    const m = section.match(new RegExp(`"${key}":"((?:[^"\\\\]|\\\\.)*)"`, 'g'));
+    if (m) {
+      const raw = m[0].replace(`"${key}":"`, '').replace(/"$/, '');
+      data[key] = raw.replace(/\\n/g, '\n').replace(/\\\\/g, '\\');
+    }
+  }
+
+  for (const key of ['ttSeller', 'verified']) {
+    const m = section.match(new RegExp(`"${key}":\\s*(true|false)`));
+    data[key] = m ? m[1] === 'true' : undefined;
+  }
+
+  for (const key of ['followerCount', 'followingCount', 'heartCount', 'videoCount', 'diggCount']) {
+    const m = section.match(new RegExp(`"${key}":(\\d+)`));
+    if (m) data[key] = parseInt(m[1], 10);
+  }
+
+  const mt = section.match(/"createTime":(\d+)/);
+  if (mt) data.createTime = parseInt(mt[1], 10);
+
+  const ma = section.match(/"avatarLarger":"([^"]*)/);
+  if (ma) data.avatarLarger = ma[1].replace(/\\u002F/g, '/');
+
+  return data;
+}
+
+export function extractLocationCreated(html) {
+  const m = html.match(/"locationCreated":"([^"]*)/);
+  return m ? m[1] : null;
+}

package/src/lib/retry.js

@@ -0,0 +1,44 @@
+import { delay } from './delay.js';
+
+const RETRYABLE_PATTERNS = [
+  'interrupted',
+  'Navigation.*interrupted',
+  'net::',
+  'ECONN',
+  'ETIMEDOUT',
+  'ENOTFOUND',
+  'EAI_AGAIN',
+  'ESOCKETRESET',
+  'connection.*refused',
+  'connection.*reset',
+  'failed.*navigate',
+  'target.*closed',
+  'crash',
+];
+
+export function isRetryableError(error) {
+  if (!error) return false;
+  const msg = (error.message || error.toString() || '').toLowerCase();
+  return RETRYABLE_PATTERNS.some(p => new RegExp(p, 'i').test(msg));
+}
+
+export async function retryWithBackoff(fn, { maxRetries = 3, baseDelay = 3000, log } = {}) {
+  let lastError;
+  for (let attempt = 0; attempt <= maxRetries; attempt++) {
+    try {
+      return await fn();
+    } catch (error) {
+      lastError = error;
+      if (attempt >= maxRetries || !isRetryableError(error)) {
+        throw error;
+      }
+      const jitter = Math.random() * 2000;
+      const waitTime = baseDelay * Math.pow(2, attempt) + jitter;
+      if (log) {
+        log(`  [重试] ${attempt + 1}/${maxRetries}，${Math.round(waitTime / 1000)}s 后重试...`);
+      }
+      await delay(Math.round(waitTime), Math.round(waitTime));
+    }
+  }
+  throw lastError;
+}

package/src/lib/scrape.js

@@ -1,39 +1,40 @@
-import { extractUserSection, parseUserSection, extractLocationCreated } from './parser.js';
-import { fetchHtml, makeProfileUrl, isProfileUrl, isVideoUrl, extractProfileHandle } from './fetcher.js';
-
-export async function extractUserData(profileUrl, proxyUrl) {
-  const profileHtml = await fetchHtml(profileUrl, proxyUrl);
-  const section = extractUserSection(profileHtml);
-  if (!section) throw new Error('无法解析用户信息');
-  const data = parseUserSection(section);
-  data.locationCreated = extractLocationCreated(profileHtml);
-  return data;
-}
-
-export async function extractVideoLocation(videoUrl, proxyUrl) {
-  const videoHtml = await fetchHtml(videoUrl, proxyUrl);
-  return extractLocationCreated(videoHtml);
-}
-
-export async function processUrl(url, proxyUrl) {
-  if (isProfileUrl(url)) {
-    const profileUrl = makeProfileUrl(url);
-    const profileData = await extractUserData(profileUrl, proxyUrl);
-    return [profileData];
-  }
-
-  if (isVideoUrl(url)) {
-    const profileHandle = extractProfileHandle(url);
-    if (!profileHandle) throw new Error(`无法从视频URL提取用户主页: ${url}`);
-
-    const profileUrl = makeProfileUrl(profileHandle);
-    const [profileData, locationCreated] = await Promise.all([
-      extractUserData(profileUrl, proxyUrl),
-      extractVideoLocation(url, proxyUrl),
-    ]);
-
-    return [{ ...profileData, locationCreated }];
-  }
-
-  return [];
-}
+import { extractUserSection, parseUserSection, extractLocationCreated } from './parser.js';
+import { fetchHtml, isProfileUrl } from './fetcher.js';
+import { toProfileUrl, isVideoUrl, extractUniqueId } from './url.js';
+
+export async function extractUserData(profileUrl, proxyUrl) {
+  const profileHtml = await fetchHtml(profileUrl, proxyUrl);
+  const section = extractUserSection(profileHtml);
+  if (!section) throw new Error('无法解析用户信息');
+  const data = parseUserSection(section);
+  data.locationCreated = extractLocationCreated(profileHtml);
+  return data;
+}
+
+export async function extractVideoLocation(videoUrl, proxyUrl) {
+  const videoHtml = await fetchHtml(videoUrl, proxyUrl);
+  return extractLocationCreated(videoHtml);
+}
+
+export async function processUrl(url, proxyUrl) {
+  if (isProfileUrl(url)) {
+    const profileUrl = toProfileUrl(url);
+    const profileData = await extractUserData(profileUrl, proxyUrl);
+    return [profileData];
+  }
+
+  if (isVideoUrl(url)) {
+    const profileHandle = extractUniqueId(url);
+    if (!profileHandle) throw new Error(`无法从视频URL提取用户主页: ${url}`);
+
+    const profileUrl = toProfileUrl(profileHandle);
+    const [profileData, locationCreated] = await Promise.all([
+      extractUserData(profileUrl, proxyUrl),
+      extractVideoLocation(url, proxyUrl),
+    ]);
+
+    return [{ ...profileData, locationCreated }];
+  }
+
+  return [];
+}

package/src/lib/url.js

@@ -0,0 +1,52 @@
+const BASE_URL = 'https://www.tiktok.com';
+
+export function extractUniqueId(url) {
+  const m = url.match(/\/@([^/]+)/);
+  return m ? m[1] : null;
+}
+
+export function extractVideoId(url) {
+  const m = url.match(/\/video\/(\d+)/);
+  return m ? m[1] : null;
+}
+
+export function normalizeUsername(input) {
+  return (input || '').replace(/^@/, '');
+}
+
+export function toProfileUrl(handle) {
+  const clean = normalizeUsername(handle);
+  return `${BASE_URL}/@${clean}`;
+}
+
+export function toVideoUrl(handle, videoId) {
+  const clean = normalizeUsername(handle);
+  return `${BASE_URL}/@${clean}/video/${videoId}`;
+}
+
+export function ensureAbsoluteUrl(href) {
+  if (href.startsWith('http')) return href;
+  return `${BASE_URL}${href}`;
+}
+
+export function isProfileUrl(url) {
+  return /\/@[\w-]+(?:$|[?#])/.test(url);
+}
+
+export function isVideoUrl(url) {
+  return /\/video\/\d+/.test(url);
+}
+
+export function extractDisplayPath(url) {
+  try {
+    const parts = new URL(url).pathname.split('/').filter(Boolean);
+    return parts.slice(-2).join('/');
+  } catch {
+    return url;
+  }
+}
+
+export function extractAuthorFromVideoUrl(url) {
+  const m = url.match(/@([^/]+)\/video/);
+  return m ? '@' + m[1] : null;
+}