feat(collect): 纯离线生成打通 1ms 签到文字点选 —— errorCode=0 收口

做了什么:
- 新增 tcaptcha/tdc_values.py: 37 个语义名的取值生成器(静态画像 / 会话量 /
  每次采集量三层)+ build_plain 四段组装 + generate_collect 纯离线出包
- crypto: 新增 generate_telemetry(mode="python_values");轨迹生成统一到
  trajectory_points()(首点绝对坐标 + 增量 + 固定收尾 [1,1,12])
- crypto.split_collect_segments 重写: pad 只有 1 个空格时旧 ` {2,}` 判据会漏段,
  且 UA 串里的普通空格会假阳性 —— 改为「模 24 算术 + 不落在 JSON 字符串内部 + 结构校验」
- tools/tdc_modules: trajectory 指纹原先指向只做 JSON.stringify 的模块,改为
  mouseMoveEvent/touchStartEvent(真轨迹模块),37/37 不变
- client 默认 telemetry_mode 改为 python_values
- 新增 scripts/verify/{align_session,audit_names,verify_values,verify_offline,send_test}.py
  与 tools/{cdp_collect,cdp_dump_scripts,tdc_profile}.py

为什么:
- 账本 #40「谁在发 HTTP 是 12 的剩余自变量」被推翻。同 aid/同客户端/同求解器、
  答案都对,只换 collect 构造方式: 本次会话构建生成 -> 0 (4/4),
  旧模板回填 -> 12 (2/2)。真因是 collect 必须与本次会话构建同源。

怎么验证:
- .venv/bin/python -m tcaptcha.selftest  全绿(含新增四段切分与纯离线闭环回归)
- scripts/verify/verify_values.py        4 份真机明文 152/152 逐槽全等
- scripts/verify/verify_offline.py       逐槽全等 + 分段几何一致
- scripts/verify/send_test.py --mode python_values  1ms 签到页文字点选 errorCode=0 x4
This commit is contained in:
yml2213
2026-10-12 03:38:18 +08:00
parent 80a122245a
commit 51dff870cf
18 changed files with 2102 additions and 91 deletions
+178
View File
@@ -0,0 +1,178 @@
#!/usr/bin/env node
/**
* cdp_module_probe.js —— 在真 Chrome 里**抓到每个采集器的真实取值**(不需要手工配平)。
*
* 背景(2026-10-12):
* 腾讯验证码 iframe (captcha.gtimg.com) 与页面 (1ms.run) **不同站**,Chrome 站点隔离下
* 它是独立 target —— 直接在页面里包 Function.prototype.apply **进不去**(实测:
* 钩子装上时模块系统已经初始化完,__mods 恒为空)。
*
* 解法(CDP 标准做法):
* Target.setAutoAttach(autoAttach, waitForDebuggerOnStart, flatten)
* -> 新的 iframe target 一创建就**暂停在任何脚本执行之前**;
* -> 此时 Page.addScriptToEvaluateOnNewDocument 注册钩子,再 runIfWaitingForDebugger 放行;
* -> tdc.js 执行时钩子已就位,模块工厂的 require/get 全被记账。
*
* 产出: {url, mods: [[module_id, get()返回值], ...], reqs: [module_id,...]}
* 配合同一会话的 tdc.js 静态分析(tdc_modules.analyze)即可得到「语义名 -> 真实取值」,
* 无需任何人工对齐。
*
* 用法:
* node tools/cdp_module_probe.js --aid 195672344 --out /tmp/1ms_bridge/modvals.json
*/
const fs = require('fs');
function arg(name, def) {
const i = process.argv.indexOf('--' + name);
if (i < 0) return def;
const v = process.argv[i + 1];
return (v === undefined || v.startsWith('--')) ? true : v;
}
const sleep = ms => new Promise(r => setTimeout(r, ms));
class Cdp {
constructor(ws) { this.ws = ws; this.id = 0; this.waits = new Map(); this.handlers = []; }
static async connect(url) {
const ws = new WebSocket(url);
await new Promise((res, rej) => { ws.onopen = res; ws.onerror = rej; });
const c = new Cdp(ws);
ws.onmessage = ev => {
const m = JSON.parse(ev.data);
if (m.id && c.waits.has(m.id)) {
const { res, rej } = c.waits.get(m.id);
c.waits.delete(m.id);
m.error ? rej(new Error(JSON.stringify(m.error))) : res(m.result);
} else if (m.method) { for (const h of c.handlers) h(m); }
};
return c;
}
send(method, params = {}, sessionId) {
const id = ++this.id;
const msg = { id, method, params };
if (sessionId) msg.sessionId = sessionId;
this.ws.send(JSON.stringify(msg));
return new Promise((res, rej) => this.waits.set(id, { res, rej }));
}
on(fn) { this.handlers.push(fn); }
close() { try { this.ws.close(); } catch (e) {} }
}
// 在任何页面脚本之前注入的钩子:记账「模块工厂 require 了谁」+「每个模块 .get() 返回什么」
const HOOK = `(() => {
if (window.__mp) return;
window.__mp = true;
window.__mods = [];
window.__reqs = [];
const oa = Function.prototype.apply;
Function.prototype.apply = function (t, a) {
try {
if (a && a.length >= 3 && typeof a[a.length - 1] === 'function') {
const req = a[a.length - 1];
if (typeof req === 'function' && !req.__w) {
const w = function (id) {
try { window.__reqs.push(String(id)); } catch (e) {}
const m = req.apply(this, arguments);
try {
if (m && typeof m.get === 'function' && !m.__g) {
m.__g = true;
const og = m.get;
m.get = function () {
let v;
try { v = og.apply(this, arguments); } catch (e) { v = '__ERR__:' + e; }
try { window.__mods.push([String(id), v]); } catch (e) {}
return v;
};
}
} catch (e) {}
return m;
};
w.__w = true;
const na = Array.prototype.slice.call(a);
na[na.length - 1] = w;
return oa.call(this, t, na);
}
}
} catch (e) {}
return oa.call(this, t, a);
};
})();`;
const SHOW = aid => `(async () => {
const load = (src) => new Promise((res, rej) => {
const s = document.createElement('script');
s.src = src; s.onload = () => res(true); s.onerror = () => rej(new Error('fail'));
document.head.appendChild(s);
});
if (typeof window.TencentCaptcha !== 'function') await load('https://ssl.captcha.qq.com/TCaptcha.js');
window.__capres = null;
window.__cap = new window.TencentCaptcha('${aid}', function (r) { window.__capres = r; }, {});
window.__cap.show();
return {ok: true};
})()`;
(async () => {
const port = parseInt(arg('port', '9222'), 10);
const aid = arg('aid', '195672344');
const out = arg('out', '/tmp/1ms_bridge/modvals.json');
const waitMs = parseInt(arg('wait', '9000'), 10);
const list = await (await fetch(`http://127.0.0.1:${port}/json/list`)).json();
const page = list.find(t => t.type === 'page' && !(t.url || '').startsWith('chrome'));
if (!page) throw new Error('没有可用 page target');
const pageCdp = await Cdp.connect(page.webSocketDebuggerUrl);
await pageCdp.send('Page.enable');
await pageCdp.send('Runtime.enable');
const ver = await (await fetch(`http://127.0.0.1:${port}/json/version`)).json();
const brCdp = await Cdp.connect(ver.webSocketDebuggerUrl);
const sessions = [];
brCdp.on(async m => {
if (m.method === 'Target.attachedToTarget') {
const { sessionId, targetInfo } = m.params;
sessions.push({ sessionId, targetInfo });
console.error(`[attach] ${targetInfo.type} ${(targetInfo.url || '').slice(0, 80)}`);
try {
await brCdp.send('Page.enable', {}, sessionId);
await brCdp.send('Page.addScriptToEvaluateOnNewDocument', { source: HOOK }, sessionId);
console.error(' -> hook registered');
} catch (e) { console.error(' -> hook fail', e.message); }
try { await brCdp.send('Runtime.runIfWaitingForDebugger', {}, sessionId); }
catch (e) {}
}
});
await brCdp.send('Target.setAutoAttach', {
autoAttach: true, waitForDebuggerOnStart: true, flatten: true,
});
console.error('[1] autoAttach 已开(新 target 会先暂停,等我们注册钩子)');
console.error('[2] 导航到 1ms.run 并拉起验证码');
await pageCdp.send('Page.navigate', { url: 'https://1ms.run/' });
await sleep(4000);
await pageCdp.send('Runtime.evaluate', {
expression: SHOW(aid), awaitPromise: true, returnByValue: true,
});
await sleep(waitMs);
console.error('[3] 在各 session 里读取记账结果');
const results = [];
for (const s of sessions) {
try {
const r = await brCdp.send('Runtime.evaluate', {
expression: `(() => { try { return {u: location.href, n: (window.__mods||[]).length,
mods: window.__mods||[], reqs: window.__reqs||[]}; } catch(e) { return {err:String(e)}; } })()`,
returnByValue: true,
}, s.sessionId);
const v = r.result && r.result.value;
if (v && v.mods && v.mods.length) { results.push(v); }
else if (v) console.error(` ${(v.u || '?').slice(0, 60)}: mods=${v.n || 0}`);
} catch (e) { console.error(' read fail', e.message); }
}
fs.writeFileSync(out, JSON.stringify(results, null, 1));
console.error(`[4] -> ${out} (命中 ${results.length} 个 session 有记账)`);
for (const r of results) {
console.error(` ${(r.u || '').slice(0, 70)} mods=${r.mods.length} reqs=${r.reqs.length}`);
}
brCdp.close(); pageCdp.close();
})().catch(e => { console.error('ERR', e.message); process.exit(1); });