tiktok-scripts/tiktok-auto/tools/pc-selftest-id.js

/**
 * pc-selftest-id.js — 在 PC 上验证 ID 提取的纯函数（不碰设备）
 *
 * 为什么单独一份：设备的自检要过 MCP 下发，一轮几十秒；而 parse/filters
 * 是**零依赖纯函数**，拼起来就能在 PC 上跑。先用它把逻辑验对，
 * 再上设备验"真机环境里也没问题"，两步分开，排查时能立刻区分是逻辑错还是环境错。
 *
 * 用法：node tools/pc-selftest-id.js
 */
'use strict';

const fs = require('fs');
const path = require('path');
const vm = require('vm');

const SRC = path.join(__dirname, '..', 'src');

/* 只拼纯函数模块 —— 它们不引用 files/device/sleep 这些 AutoX 全局 */
const MODULES = ['parse', 'filters', 'uidset'];

const sandbox = {
    console: console,
    Date: Date, Math: Math, JSON: JSON,
    String: String, Number: Number, Object: Object,
    Array: Array, RegExp: RegExp, isNaN: isNaN, parseInt: parseInt,
    TikTokAuto: {}
};
/* uidset 与 filters 在函数体内取 logger / files，这里给最小桩 */
sandbox.TikTokAuto.logger = {
    info: () => { }, warn: () => { }, error: () => { }, step: () => { }
};
sandbox.files = {
    exists: () => false,
    read: () => '',
    append: () => { },
    ensureDir: () => { }
};

vm.createContext(sandbox);

let source = 'var TikTokAuto = this.TikTokAuto;\n';
for (const name of MODULES) {
    source += fs.readFileSync(path.join(SRC, `${name}.js`), 'utf8') + '\n';
}
vm.runInContext(source, sandbox);

const parse = sandbox.TikTokAuto.parse;
const filters = sandbox.TikTokAuto.filters;
const uidset = sandbox.TikTokAuto.uidset;

let pass = 0;
let fail = 0;

function check(name, condition, detail) {
    if (condition === true) {
        pass += 1;
        return;
    }
    fail += 1;
    console.log('FAIL ' + name + (detail === undefined ? '' : '  → ' + detail));
}

/* ---------------------------------------------------------- 计数解析 */
const CASES = [
    ['594', 594], ['2,491', 2491], ['1.6M', 1600000], ['253.9M', 253900000],
    ['245K', 245000], ['371.7K', 371700], ['1.2万', 12000], ['1.5亿', 150000000],
    ['1.6M followers', 1600000], ['594 followers · 2.5K likes', 594],
    ['0', 0], ['12.3K', 12300], ['1,234', 1234], ['100万', 1000000]
];
for (const [input, want] of CASES) {
    const got = parse.count(input);
    check(`parseCount("${input}") = ${want}`, got === want, `实际 ${got}`);
}
check('parseCount 空串 → null', parse.count('') === null);
check('parseCount 非数字 → null', parse.count('abc') === null);
check('parseCount 全角数字 "１.２万"', parse.count('１.２万') === 12000, String(parse.count('１.２万')));

/* ---------------------------------------------------------- 语言 */
check('中文 → zh', parse.language('小美爱穿搭') === 'zh');
check('英文 → en', parse.language('Fashion Girl') === 'en');
check('中文混英文 → zh', parse.language('小美的Fashion') === 'zh', parse.language('小美的Fashion'));
check('英文混一个字 → en', parse.language('Fashion Girl 小') === 'en', parse.language('Fashion Girl 小'));
check('纯 emoji → none', parse.language('🥷🏿💗✨') === 'none');
check('阿拉伯文 → none（没有中英线索）', parse.language('عايشه حسين') === 'none', parse.language('عايشه حسين'));
check('zh 不放行 en', parse.languageMatches('zh', 'en') === false);
check('en 不放行 zh', parse.languageMatches('en', 'zh') === false);
check('any 全放行', parse.languageMatches('any', 'en') === true && parse.languageMatches('any', 'none') === true);

/* ---------------------------------------------------------- 区间 */
check('未设区间 → 无脑通过', parse.inRange(null, 0, 0) === true);
check('值在内', parse.inRange(500, 100, 1000) === true);
check('值在下限外', parse.inRange(50, 100, 1000) === false);
check('上限 0 = 不设限', parse.inRange(999999, 100, 0) === true);
check('只设上限', parse.inRange(500, 0, 1000) === true && parse.inRange(5000, 0, 1000) === false);
check('设了区间但值为 null → 拦', parse.inRange(null, 100, 1000) === false);
check('isRangeOpen 双向为 0 才算空', parse.isRangeOpen(0, 0) === true && parse.isRangeOpen(1, 0) === false);

/* ---------------------------------------------------------- 用户名 */
check('剥 @', parse.username('@abc') === 'abc');
check('剥 FID 隔离符', parse.username('\u2068ajman_fashion_week\u2069') === 'ajman_fashion_week',
    JSON.stringify(parse.username('\u2068ajman_fashion_week\u2069')));
check('合法用户名', parse.looksLikeUsername('user3670597276145') === true);
check('带点的用户名', parse.looksLikeUsername('mukhawar.shop') === true);
check('文案不是用户名', parse.looksLikeUsername('This is a caption') === false);
check('过长不是用户名', parse.looksLikeUsername('a'.repeat(31)) === false);

/* ---------------------------------------------------------- 去重集合 */
const set = uidset.create();
check('首次 add → true', set.add('a') === true);
check('重复 add → false', set.add('a') === false);
check('size 只算唯一值', set.size() === 1);
check('has 正确', set.has('a') === true && set.has('b') === false);

/* ---------------------------------------------------------- 过滤 */
check('无过滤全放行', filters.judge({ uid: 'x' }, {}).ok === true);
check('私密被拦', filters.judge({ uid: 'x', isPrivate: true }, { private: true }).ok === false);
check('公开放行', filters.judge({ uid: 'x', isPrivate: false }, { private: true }).ok === true);

check('仅中文 + 中文昵称', filters.judge({ uid: 'x', nickname: '小美' }, { langRegion: 'zh' }).ok === true);
check('仅中文 + 英文昵称被拦', filters.judge({ uid: 'x', nickname: 'Fashion' }, { langRegion: 'zh' }).ok === false);
check('仅中文 + 纯 emoji 被拦', filters.judge({ uid: 'x', nickname: '🥷🏿' }, { langRegion: 'zh' }).ok === false);
check('取样=个签：昵称英文但个签中文 → 放行',
    filters.judge({ uid: 'x', nickname: 'Fashion', signature: '小美' },
        { langRegion: 'zh', nicknameCheck: 'sign' }).ok === true);
check('取样=昵称：同上应当被拦',
    filters.judge({ uid: 'x', nickname: 'Fashion', signature: '小美' },
        { langRegion: 'zh', nicknameCheck: 'nick' }).ok === false);

check('粉丝达标', filters.judge({ uid: 'x', fans: 5940 }, { fans: { min: 1000, max: 0 } }).ok === true);
check('粉丝不足', filters.judge({ uid: 'x', fans: 594 }, { fans: { min: 1000, max: 0 } }).ok === false);
check('粉丝读不到 → 拦', filters.judge({ uid: 'x', fans: null }, { fans: { min: 1000, max: 0 } }).ok === false);
check('粉丝上界', filters.judge({ uid: 'x', fans: 99999 }, { fans: { min: 0, max: 10000 } }).ok === false);

const videoVerdict = filters.judge({ uid: 'x' }, { videos: { min: 10, max: 0 } });
check('视频量不拦人', videoVerdict.ok === true);
check('视频量要如实告知未生效', videoVerdict.soft.indexOf('视频量') >= 0, videoVerdict.soft);

check('无头像：默认剪影被拦', filters.judge({ uid: 'x', hasAvatar: false }, { noAvatar: true }).ok === false);
check('无头像：有头像放行', filters.judge({ uid: 'x', hasAvatar: true }, { noAvatar: true }).ok === true);
const noAvatarUnknown = filters.judge({ uid: 'x', hasAvatar: null }, { noAvatar: true });
check('无头像：判不了时放行', noAvatarUnknown.ok === true);
check('无头像：判不了要告知', noAvatarUnknown.soft.indexOf('无头像') >= 0, noAvatarUnknown.soft);

check('无过滤 → 不必开用户页', filters.needsProfile({}) === false);
check('只有视频量 → 不必开用户页', filters.needsProfile({ videos: { min: 10, max: 0 } }) === false);
check('有粉丝量 → 必须开', filters.needsProfile({ fans: { min: 10, max: 0 } }) === true);
check('有语言 → 必须开', filters.needsProfile({ langRegion: 'zh' }) === true);

/* 「读不到资料」必须拦下 —— 放行等于"打不开的人全都通过"，
 * 过滤看着在跑、实际放进一批没检查过的账号，比不过滤更糟。
 * 但快路径（本来不需要资料）仍然要放行。 */
check('需要资料但拿不到 → 拦下', filters.judge(null, { fans: { min: 1000, max: 0 } }).ok === false);
check('不需要资料且没有资料 → 放行（快路径）', filters.judge(null, {}).ok === true);

/* ---------------------------------------------------------- 文件名 */
const name = parse.timestampName('uid', new Date(2026, 9, 20, 18, 31, 0));
check('文件名带时分', name === 'uid_20261020_1831.txt', name);

console.log('');
console.log(`通过 ${pass} 项，失败 ${fail} 项`);
process.exit(fail === 0 ? 0 : 1);