ys/GenericAgent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

TMWebDriver: add debug logs + CDP bridge refactor (action→cmd, extract buildExecScript, handleWsExec routing)

Browse Source
This commit is contained in:
Liang Jiaqing
2026-03-31 20:37:05 +08:00
parent 86210296c1
commit d922303645
4 changed files with 136 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
TMWebDriver.py
View File

@@ -30,6 +30,7 @@ class Session:
self.connect_at = time.time() self.connect_at = time.time()
self.disconnect_at = None self.disconnect_at = None
def mark_disconnected(self): def mark_disconnected(self):
print(f"Tab disconnected: {self.url} (Session: {self.id})")
self.disconnect_at = time.time() self.disconnect_at = time.time()
@@ -131,6 +132,7 @@ class TMWebDriver:
elif data.get('type') in ['ext_ready', 'tabs_update']: elif data.get('type') in ['ext_ready', 'tabs_update']:
tabs = data.get('tabs', []) tabs = data.get('tabs', [])
current_tab_ids = {str(tab['id']) for tab in tabs} current_tab_ids = {str(tab['id']) for tab in tabs}
print(f"Received tabs update: {current_tab_ids}")
for sid in list(driver.sessions.keys()): for sid in list(driver.sessions.keys()):
sess = driver.sessions[sid] sess = driver.sessions[sid]
if sess.type == 'ext_ws' and sid not in current_tab_ids: if sess.type == 'ext_ws' and sid not in current_tab_ids:
@@ -150,7 +152,9 @@ class TMWebDriver:
print(f"Error handling message: {e}") print(f"Error handling message: {e}")
if hasattr(self, 'data'): print(self.data) if hasattr(self, 'data'): print(self.data)
def connected(self): (f"New connection from {self.address}") def connected(self): (f"New connection from {self.address}")
def handle_close(self): driver._unregister_client(self) def handle_close(self):
print(f"WS Connection closed: {self.address}")
driver._unregister_client(self)
self.server = WebSocketServer(self.host, self.port, JSExecutor) self.server = WebSocketServer(self.host, self.port, JSExecutor)
server_thread = threading.Thread(target=self.server.serve_forever) server_thread = threading.Thread(target=self.server.serve_forever)

245
assets/tmwd_cdp_bridge/background.js
View File

@@ -15,40 +15,38 @@ chrome.runtime.onInstalled.addListener(() => {
}); });
}); });
async function handleExtMessage(msg, sender) {
if (msg.cmd === 'cookies') return await handleCookies(msg, sender);
if (msg.cmd === 'cdp') return await handleCDP(msg, sender);
if (msg.cmd === 'batch') return await handleBatch(msg, sender);
if (msg.cmd === 'tabs') {
try {
if (msg.method === 'switch') {
const tab = await chrome.tabs.update(msg.tabId, { active: true });
await chrome.windows.update(tab.windowId, { focused: true });
return { ok: true };
} else {
const tabs = (await chrome.tabs.query({})).filter(t => isScriptable(t.url));
const data = tabs.map(t => ({ id: t.id, url: t.url, title: t.title, active: t.active, windowId: t.windowId }));
return { ok: true, data };
}
} catch (e) { return { ok: false, error: e.message }; }
}
return { ok: false, error: 'Unknown cmd: ' + msg.cmd };
}
chrome.runtime.onMessage.addListener((msg, sender, sendResponse) => { chrome.runtime.onMessage.addListener((msg, sender, sendResponse) => {
if (msg.action === 'cookies') { handleExtMessage(msg, sender).then(sendResponse);
handleCookies(msg, sender).then(sendResponse); return true;
return true;
}
if (msg.action === 'cdp') {
handleCDP(msg, sender).then(sendResponse);
return true;
}
if (msg.action === 'batch') {
handleBatch(msg, sender).then(sendResponse);
return true;
}
if (msg.action === 'tabs') {
(async () => {
try {
if (msg.method === 'switch') {
const tab = await chrome.tabs.update(msg.tabId, { active: true });
await chrome.windows.update(tab.windowId, { focused: true });
sendResponse({ ok: true });
} else {
const tabs = (await chrome.tabs.query({})).filter(t => isScriptable(t.url));
const data = tabs.map(t => ({ id: t.id, url: t.url, title: t.title, active: t.active, windowId: t.windowId }));
sendResponse({ ok: true, data });
}
} catch (e) { sendResponse({ ok: false, error: e.message }); }
})();
return true;
}
}); });
async function handleCookies(msg, sender) { async function handleCookies(msg, sender) {
try { try {
const url = msg.url || sender.tab?.url; let url = msg.url || sender.tab?.url;
if (!url && msg.tabId) {
const tab = await chrome.tabs.get(msg.tabId);
url = tab.url;
}
const origin = url.match(/^https?:\/\/[^\/]+/)[0]; const origin = url.match(/^https?:\/\/[^\/]+/)[0];
const all = await chrome.cookies.getAll({ url }); const all = await chrome.cookies.getAll({ url });
const part = await chrome.cookies.getAll({ url, partitionKey: { topLevelSite: origin } }).catch(() => []); const part = await chrome.cookies.getAll({ url, partitionKey: { topLevelSite: origin } }).catch(() => []);
@@ -69,6 +67,7 @@ async function handleBatch(msg, sender) {
(_, i, path) => { let v = R[+i]; for (const k of path.split('.')) v = v[k]; return JSON.stringify(v); })); (_, i, path) => { let v = R[+i]; for (const k of path.split('.')) v = v[k]; return JSON.stringify(v); }));
try { try {
for (const c of msg.commands) { for (const c of msg.commands) {
if (c.tabId === undefined && msg.tabId !== undefined) c.tabId = msg.tabId;
if (c.cmd === 'cookies') { if (c.cmd === 'cookies') {
R.push(await handleCookies(c, sender)); R.push(await handleCookies(c, sender));
} else if (c.cmd === 'tabs') { } else if (c.cmd === 'tabs') {
@@ -110,8 +109,8 @@ async function handleCDP(msg, sender) {
// Filter out chrome:// and other internal tabs that can't be scripted // Filter out chrome:// and other internal tabs that can't be scripted
const isScriptable = url => url && /^https?:/.test(url); const isScriptable = url => url && /^https?:/.test(url);
// --- Shared page-script builder (used by both executeScript and CDP fallback) --- // --- Shared page/CDP script builder core ---
function buildPageScript(code) { function buildExecScript(code, errorHandler) {
return `(async () => { return `(async () => {
function smartProcessResult(result) { function smartProcessResult(result) {
if (result === null || result === undefined || typeof result !== 'object') return result; if (result === null || result === undefined || typeof result !== 'object') return result;
@@ -146,52 +145,23 @@ function buildPageScript(code) {
} }
return { ok: true, data: smartProcessResult(r) }; return { ok: true, data: smartProcessResult(r) };
} catch (e) { } catch (e) {
const errMsg = e.message || String(e); ${errorHandler}
return { ok: false, error: { name: e.name || 'Error', message: errMsg, stack: e.stack || '' },
csp: errMsg.includes('Refused to evaluate') || errMsg.includes('unsafe-eval') || errMsg.includes('Content Security Policy') };
} }
})()`; })()`;
} }
// --- CDP script: includes smartProcessResult to avoid "Object reference chain is too long" --- function buildPageScript(code) {
return buildExecScript(code, `
const errMsg = e.message || String(e);
return { ok: false, error: { name: e.name || 'Error', message: errMsg, stack: e.stack || '' },
csp: errMsg.includes('Refused to evaluate') || errMsg.includes('unsafe-eval') || errMsg.includes('Content Security Policy') };
`);
}
function buildCdpScript(code) { function buildCdpScript(code) {
return `(async () => { return buildExecScript(code, `
function smartProcessResult(result) {
if (result === null || result === undefined || typeof result !== 'object') return result;
if (typeof jQuery !== 'undefined' && result instanceof jQuery) {
const elements = []; for (let i = 0; i < result.length; i++) { if (result[i] && result[i].nodeType === 1) elements.push(result[i].outerHTML); } return elements;
}
if (result instanceof NodeList || result instanceof HTMLCollection) {
const elements = []; for (let i = 0; i < result.length; i++) { if (result[i] && result[i].nodeType === 1) elements.push(result[i].outerHTML); } return elements;
}
if (result.nodeType === 1) return result.outerHTML;
if (!Array.isArray(result) && typeof result === 'object' && 'length' in result && typeof result.length === 'number') {
const firstElement = result[0];
if (firstElement && firstElement.nodeType === 1) {
const elements = []; const length = Math.min(result.length, 100);
for (let i = 0; i < length; i++) { const elem = result[i]; if (elem && elem.nodeType === 1) elements.push(elem.outerHTML); } return elements;
}
}
try { return JSON.parse(JSON.stringify(result, function(key, value) { if (typeof value === 'object' && value !== null) { if (value.nodeType === 1) return value.outerHTML; if (value === window || value === document) return '[Object]'; } return value; })); } catch (e) { return '[无法序列化: ' + e.message + ']'; }
}
try {
const jsCode = ${JSON.stringify(code)}.trim();
const lines = jsCode.split(/\\r?\\n/).filter(l => l.trim());
const lastLine = lines.length > 0 ? lines[lines.length - 1].trim() : '';
const AsyncFunction = Object.getPrototypeOf(async function(){}).constructor;
let r;
if (lastLine.startsWith('return')) {
r = await (new AsyncFunction(jsCode))();
} else {
try { r = eval(jsCode); if (r instanceof Promise) r = await r; } catch (e) {
if (e instanceof SyntaxError && (/return/i.test(e.message) || /await/i.test(e.message))) { r = await (new AsyncFunction(jsCode))(); } else throw e;
}
}
return { ok: true, data: smartProcessResult(r) };
} catch (e) {
return { ok: false, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' } }; return { ok: false, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' } };
} `);
})()`;
} }
// --- WebSocket Client for TMWebDriver --- // --- WebSocket Client for TMWebDriver ---
@@ -242,6 +212,66 @@ chrome.alarms.onAlarm.addListener(async (alarm) => {
} }
}); });
async function handleWsExec(data) {
const tabId = data.tabId;
console.log('[TMWD-WS] Exec request', data.id, 'on tab', tabId);
ws.send(JSON.stringify({ type: 'ack', id: data.id }));
if (!tabId) {
ws.send(JSON.stringify({ type: 'error', id: data.id, error: 'No tabId provided' }));
return;
}
try {
const tabsBefore = new Set((await chrome.tabs.query({})).map(t => t.id));
let res;
try {
const result = await chrome.scripting.executeScript({
target: { tabId },
world: 'MAIN',
func: async (s) => await eval(s),
args: [buildPageScript(data.code)]
});
res = result[0]?.result;
if (res === null || res === undefined) {
console.log('[TMWD-WS] executeScript returned null/undefined, treating as CSP issue');
res = { ok: false, error: { name: 'Error', message: 'executeScript returned null (possible CSP or context issue)', stack: '' }, csp: true };
}
} catch (e) {
console.log('[TMWD-WS] scripting.executeScript failed:', e.message);
res = { ok: false, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' }, csp: true };
}
// CDP fallback for CSP-restricted pages
if (res && !res.ok && res.csp) {
console.log('[TMWD-WS] CDP fallback for tab', tabId);
const wrappedCode = buildCdpScript(data.code);
try {
await chrome.debugger.attach({ tabId }, '1.3');
const cdpRes = await chrome.debugger.sendCommand({ tabId }, 'Runtime.evaluate', {
expression: wrappedCode, awaitPromise: true, returnByValue: true
});
await chrome.debugger.detach({ tabId });
if (cdpRes.exceptionDetails) {
const desc = cdpRes.exceptionDetails.exception?.description || 'CDP Error';
res = { ok: false, error: { name: 'Error', message: desc, stack: desc } };
} else {
res = cdpRes.result.value;
}
} catch (cdpErr) {
try { await chrome.debugger.detach({ tabId }); } catch (_) {}
res = { ok: false, error: { name: 'Error', message: 'CDP fallback failed: ' + cdpErr.message, stack: '' } };
}
}
const newTabs = (await chrome.tabs.query({})).filter(t => !tabsBefore.has(t.id)).map(t => ({id: t.id, url: t.url, title: t.title}));
if (res?.ok) {
ws.send(JSON.stringify({ type: 'result', id: data.id, result: res.data, newTabs }));
} else {
console.log(res);
ws.send(JSON.stringify({ type: 'error', id: data.id, error: res?.error || 'Unknown error', newTabs }));
}
} catch (e) {
ws.send(JSON.stringify({ type: 'error', id: data.id, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' } }));
}
}
function connectWS() { function connectWS() {
if (ws && ws.readyState <= 1) return; // CONNECTING or OPEN if (ws && ws.readyState <= 1) return; // CONNECTING or OPEN
ws = null; ws = null;
@@ -268,63 +298,24 @@ function connectWS() {
try { try {
const data = JSON.parse(event.data); const data = JSON.parse(event.data);
if (data.id && data.code) { if (data.id && data.code) {
const tabId = data.tabId; let code = data.code;
console.log('[TMWD-WS] Exec request', data.id, 'on tab', tabId); // If code is a JSON string representing an object, parse it
// Send ACK immediately so Python side resets timeout timer if (typeof code === 'string') {
ws.send(JSON.stringify({ type: 'ack', id: data.id })); try { const p = JSON.parse(code); if (p && typeof p === 'object') code = p; } catch (_) {}
if (!tabId) {
ws.send(JSON.stringify({ type: 'error', id: data.id, error: 'No tabId provided' }));
return;
} }
try { if (typeof code === 'object' && code !== null && code.cmd) {
const tabsBefore = new Set((await chrome.tabs.query({})).map(t => t.id)); // Custom protocol message → route to handleExtMessage
let res; if (code.tabId === undefined && data.tabId !== undefined) code.tabId = data.tabId;
try { const res = await handleExtMessage(code, {});
const result = await chrome.scripting.executeScript({ ws.send(JSON.stringify({ type: res.ok ? 'result' : 'error', id: data.id, result: res.data ?? res.results ?? res, error: res.error }));
target: { tabId }, } else if (typeof code === 'string') {
world: 'MAIN', // Plain JS code
func: async (s) => await eval(s), await handleWsExec(data);
args: [buildPageScript(data.code)] } else if (typeof code === 'object' && code !== null) {
}); // Object without cmd → legacy extension message
res = result[0]?.result; const msg = code.tabId === undefined && data.tabId !== undefined ? { ...code, tabId: data.tabId } : code;
if (res === null || res === undefined) { const res = await handleExtMessage(msg, {});
console.log('[TMWD-WS] executeScript returned null/undefined, treating as CSP issue'); ws.send(JSON.stringify({ type: res.ok ? 'result' : 'error', id: data.id, result: res.data ?? res.results ?? res, error: res.error }));
res = { ok: false, error: { name: 'Error', message: 'executeScript returned null (possible CSP or context issue)', stack: '' }, csp: true };
}
} catch (e) {
console.log('[TMWD-WS] scripting.executeScript failed:', e.message);
res = { ok: false, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' }, csp: true };
}
// CDP fallback for CSP-restricted pages
if (res && !res.ok && res.csp) {
console.log('[TMWD-WS] CDP fallback for tab', tabId);
const wrappedCode = buildCdpScript(data.code);
try {
await chrome.debugger.attach({ tabId }, '1.3');
const cdpRes = await chrome.debugger.sendCommand({ tabId }, 'Runtime.evaluate', {
expression: wrappedCode, awaitPromise: true, returnByValue: true
});
await chrome.debugger.detach({ tabId });
if (cdpRes.exceptionDetails) {
const desc = cdpRes.exceptionDetails.exception?.description || 'CDP Error';
res = { ok: false, error: { name: 'Error', message: desc, stack: desc } };
} else {
res = cdpRes.result.value; // Already {ok, data/error} from the wrapper
}
} catch (cdpErr) {
try { await chrome.debugger.detach({ tabId }); } catch (_) {}
res = { ok: false, error: { name: 'Error', message: 'CDP fallback failed: ' + cdpErr.message, stack: '' } };
}
}
const newTabs = (await chrome.tabs.query({})).filter(t => !tabsBefore.has(t.id)).map(t => ({id: t.id, url: t.url, title: t.title}));
if (res?.ok) {
ws.send(JSON.stringify({ type: 'result', id: data.id, result: res.data, newTabs }));
} else {
console.log(res);
ws.send(JSON.stringify({ type: 'error', id: data.id, error: res?.error || 'Unknown error', newTabs }));
}
} catch (e) {
ws.send(JSON.stringify({ type: 'error', id: data.id, error: { name: e.name || 'Error', message: e.message || String(e), stack: e.stack || '' } }));
} }
} }
} catch (e) { } catch (e) {

8
assets/tmwd_cdp_bridge/content.js
View File

@@ -28,13 +28,13 @@ async function handle(el) {
const cmd = req.cmd || 'cookies'; const cmd = req.cmd || 'cookies';
let resp; let resp;
if (cmd === 'cookies') { if (cmd === 'cookies') {
resp = await chrome.runtime.sendMessage({ action: 'cookies', url: req.url || location.href }); resp = await chrome.runtime.sendMessage({ cmd: 'cookies', url: req.url || location.href });
} else if (cmd === 'cdp') { } else if (cmd === 'cdp') {
resp = await chrome.runtime.sendMessage({ action: 'cdp', method: req.method, params: req.params || {}, tabId: req.tabId }); resp = await chrome.runtime.sendMessage({ cmd: 'cdp', method: req.method, params: req.params || {}, tabId: req.tabId });
} else if (cmd === 'batch') { } else if (cmd === 'batch') {
resp = await chrome.runtime.sendMessage({ action: 'batch', commands: req.commands, tabId: req.tabId }); resp = await chrome.runtime.sendMessage({ cmd: 'batch', commands: req.commands, tabId: req.tabId });
} else if (cmd === 'tabs') { } else if (cmd === 'tabs') {
resp = await chrome.runtime.sendMessage({ action: 'tabs', method: req.method, tabId: req.tabId }); resp = await chrome.runtime.sendMessage({ cmd: 'tabs', method: req.method, tabId: req.tabId });
} else { } else {
resp = { ok: false, error: 'unknown cmd: ' + cmd }; resp = { ok: false, error: 'unknown cmd: ' + cmd };
} }

17
memory/tmwebdriver_sop.md
View File

@@ -53,19 +53,20 @@ fetch('PDF_URL').then(r=>r.blob()).then(b=>{
## CDP桥(tmwd_cdp_bridge扩展) ⭐首选 ## CDP桥(tmwd_cdp_bridge扩展) ⭐首选
扩展路径：`assets/tmwd_cdp_bridge/`(需安装，含debugger权限) 扩展路径：`assets/tmwd_cdp_bridge/`(需安装，含debugger权限)
⚠TID密钥：首次运行自动生成到`assets/tmwd_cdp_bridge/config.js`(已gitignore)，扩展通过manifest引用 ⚠TID密钥：首次运行自动生成到`assets/tmwd_cdp_bridge/config.js`(已gitignore)，扩展通过manifest引用
调用：MutationObserver监听addedNodes(id=TID)，⚠每次必须remove旧→createElement新→设textContent JSON→appendChild 调用：`web_execute_js` script直传JSON字符串（工具层自动识别对象格式，走WS→background.js cmd路由）
```js ```js
// TID从assets/tmwd_cdp_bridge/config.js读取 // 直接传JSON字符串作为script参数，无需DOM操作
const old = document.getElementById(TID); web_execute_js script='{"cmd": "cookies"}'
if (old) old.remove(); web_execute_js script='{"cmd": "tabs"}'
const el = document.createElement('div'); web_execute_js script='{"cmd": "cdp", "tabId": N, "method": "...", "params": {...}}'
el.id = TID; el.style.display = 'none'; web_execute_js script='{"cmd": "batch", "commands": [...]}'
el.textContent = JSON.stringify({cmd:'...', ...}); // 返回值直接是JSON结果
document.body.appendChild(el); // 响应写回el.textContent
``` ```
⚠旧DOM方式(TID元素+MutationObserver)仍可用但已不推荐
单命令：`{cmd:'tabs'}` | `{cmd:'cookies'}` | `{cmd:'cdp', tabId:N, method:'...', params:{...}}` 单命令：`{cmd:'tabs'}` | `{cmd:'cookies'}` | `{cmd:'cdp', tabId:N, method:'...', params:{...}}`
- ⭐batch混合：`{cmd:'batch', commands:[{cmd:'cookies'},{cmd:'tabs'},{cmd:'cdp',...},...]}` - ⭐batch混合：`{cmd:'batch', commands:[{cmd:'cookies'},{cmd:'tabs'},{cmd:'cdp',...},...]}`
- 返回`{ok:true, results:[...]}`，一次请求多命令，CDP懒attach复用session - 返回`{ok:true, results:[...]}`，一次请求多命令，CDP懒attach复用session
- 子命令会自动继承外层batch的tabId（如cookies命令可正确获取当前页面URL）
- `$N.path`引用第N个结果字段(0-indexed)，如`"nodeId":"$2.root.nodeId"` - `$N.path`引用第N个结果字段(0-indexed)，如`"nodeId":"$2.root.nodeId"`
- ⚠batch前序命令失败时后续`$N`引用拿到undefined，整条链路**静默失败不报错**，需检查返回results数组中每项的ok状态（未验证，BBS#46） - ⚠batch前序命令失败时后续`$N`引用拿到undefined，整条链路**静默失败不报错**，需检查返回results数组中每项的ok状态（未验证，BBS#46）
- 典型：文件上传三连 getDocument(**depth:1**性能优化，200ms+→个位数ms)→querySelector(input[type=file])→setFileInputFiles（未验证，BBS#38） - 典型：文件上传三连 getDocument(**depth:1**性能优化，200ms+→个位数ms)→querySelector(input[type=file])→setFileInputFiles（未验证，BBS#38）