feat(monitor): 抖音 Web 接口客户端 —— 绕开爬虫子进程，直接发 HTTP
移植自 mac-agent-os 的 mediacrawler_adapter：不起子进程、不开页面，用浏览器里那份
登录态直接调抖音 Web 接口。产物键名照抄 store/douyin，所以 ingest 那条链路一个字不用改。
**目前能用的（真环境实测，非推断）**：
profile/other : 200, 7075 字节 —— 博主主页指标（粉丝/获赞/作品数/昵称）
aweme/detail : 200, 45425 字节 —— 单条作品详情（含点赞/评论/收藏/分享）
**目前不能用的：作品列表 `aweme/post`。** 两个互相独立的原因：
1. 这个接口被抖音单独升级成了真校验：不带 x-tt-argus 回 403「Uifid Not Found」，
带上 dummy 值回 200 + **空 body**。也就是说「头在不在」骗得过，「真校验」过不了。
同一套头打 profile/other 和 aweme/detail 都是通的 —— 抖音是挑着接口加保护的，
挑中的恰好是「批量拉作品列表」这个最敏感的动作。
2. 改走页面截获也不行：CDP 浏览器打开博主主页会落到「验证码中间页」（当天大量探测的
代价，过几小时要重测）。
所以现在的边界是：**已知作品的指标刷新能做，自动发现新作品做不了**。
**排查中控住变量后得到的两条事实**（都写进注释了）：
· `Accept` / `Accept-Language` / `Referer` 才是主页接口能返回真数据的原因 —— 只有
UA+client hints+Cookie 时是 200 但仅 121 字节的空壳，补上这三个头变 7074 字节。
（我先前猜的 sec-ch-ua 不是关键。）
· 因此 UA 与 client hints 必须**成套地取自同一个浏览器**，所以 BrowserIdentity 一次
从 CDP 取齐 cookie + UA + hints，而不是各自写死。
「200 + 空 body 必须当场报错」也是刻意写死的：放过去它会在下游变成「这个博主没作品」，
把一次失败伪装成一条正常结果 —— 爬虫那条路正是这么栽的，还被翻译成「账号被封」。
测试 +11：cookie 解析、请求头成套性（含 uifid 缺失/回退）、产物键名与 store 对齐、
以及 _get 的三条失败路径（空 body / 403 带网关原话 / 正常返回）。
This commit is contained in:
@@ -0,0 +1,202 @@
|
||||
"""
|
||||
联邦数据聚合器
|
||||
|
||||
从所有在线机器拉取数据并合并，供 Dashboard 展示。
|
||||
所有机器通过 Tailscale 网络互联，端口统一 9988。
|
||||
"""
|
||||
|
||||
import json, logging, sys
|
||||
from pathlib import Path
|
||||
|
||||
logger = logging.getLogger("dashboard.federation")
|
||||
|
||||
# 确保能找到 ORACLE.yaml
|
||||
_THIS_DIR = Path(__file__).resolve().parent.parent
|
||||
if str(_THIS_DIR) not in sys.path:
|
||||
sys.path.insert(0, str(_THIS_DIR))
|
||||
|
||||
ORACLE_PATH = _THIS_DIR.parent.parent / "ORACLE.yaml"
|
||||
|
||||
|
||||
def _get_machines():
|
||||
"""从 ORACLE.yaml 读取所有机器信息（排除本机）"""
|
||||
import yaml
|
||||
import os, socket, subprocess, re
|
||||
|
||||
if not ORACLE_PATH.exists():
|
||||
return []
|
||||
|
||||
with open(ORACLE_PATH) as f:
|
||||
oracle = yaml.safe_load(f)
|
||||
|
||||
# 获取本机所有 IP（包括 Tailscale）
|
||||
try:
|
||||
output = subprocess.run(['ifconfig'], capture_output=True, text=True, timeout=5)
|
||||
local_ips = set(re.findall(r'inet (\d+\.\d+\.\d+\.\d+)', output.stdout))
|
||||
except:
|
||||
local_ips = {"127.0.0.1"}
|
||||
local_ips.add("127.0.0.1")
|
||||
|
||||
machines = []
|
||||
for name, info in oracle.get("machines", {}).items():
|
||||
machine_ip = info.get("tailscale_ip", "")
|
||||
if machine_ip in local_ips:
|
||||
continue # 跳过本机
|
||||
machines.append({
|
||||
"name": name,
|
||||
"hostname": info.get("hostname", name),
|
||||
"ip": info.get("tailscale_ip", ""),
|
||||
"port": info.get("dashboard_port", 9988),
|
||||
"user": info.get("ssh_user", ""),
|
||||
})
|
||||
return machines
|
||||
|
||||
|
||||
def fetch_machine_data(machine: dict, path: str, timeout: int = 5) -> dict:
|
||||
"""从指定机器拉取数据（使用 urllib，无外部依赖）"""
|
||||
import urllib.request, urllib.error
|
||||
|
||||
url = f"http://{machine['ip']}:{machine['port']}{path}"
|
||||
try:
|
||||
r = urllib.request.urlopen(url, timeout=timeout)
|
||||
if r.status == 200:
|
||||
return json.loads(r.read().decode())
|
||||
return {"error": f"HTTP {r.status}"}
|
||||
except urllib.error.URLError as e:
|
||||
return {"error": f"连接失败: {e.reason}"}
|
||||
except TimeoutError:
|
||||
return {"error": "超时"}
|
||||
except Exception as e:
|
||||
return {"error": str(e)}
|
||||
|
||||
|
||||
def aggregate_accounts():
|
||||
"""聚合所有机器的账号列表（含本机）"""
|
||||
from plugins.base import HOSTNAME
|
||||
machines = _get_machines()
|
||||
all_accounts = []
|
||||
|
||||
# 先加本机数据（只取本机账号）
|
||||
try:
|
||||
import urllib.request, json as _json
|
||||
local_resp = urllib.request.urlopen("http://localhost:9988/api/matrix/accounts", timeout=5)
|
||||
local_data = _json.loads(local_resp.read())
|
||||
if isinstance(local_data, list):
|
||||
for acct in local_data:
|
||||
if acct.get("owner_machine", "") != HOSTNAME:
|
||||
continue
|
||||
acct["_source_machine"] = HOSTNAME
|
||||
acct["is_local"] = True
|
||||
all_accounts.append(acct)
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# 去重集合: (machine, account_id)
|
||||
seen = set()
|
||||
|
||||
for m in machines:
|
||||
data = fetch_machine_data(m, "/api/matrix/accounts")
|
||||
if isinstance(data, list):
|
||||
for acct in data:
|
||||
# 只取该机器自己的账号（去重）
|
||||
if acct.get("owner_machine", "") != m["name"]:
|
||||
continue
|
||||
key = m["name"] + "|" + acct.get("id", "")
|
||||
if key in seen:
|
||||
continue
|
||||
seen.add(key)
|
||||
acct["_source_machine"] = m["name"]
|
||||
acct["is_local"] = False
|
||||
all_accounts.append(acct)
|
||||
elif isinstance(data, dict) and "error" not in data:
|
||||
accts = data.get("data", data.get("accounts", []))
|
||||
if isinstance(accts, list):
|
||||
for acct in accts:
|
||||
if acct.get("owner_machine", "") != m["name"]:
|
||||
continue
|
||||
key = m["name"] + "|" + acct.get("id", "")
|
||||
if key in seen:
|
||||
continue
|
||||
seen.add(key)
|
||||
acct["_source_machine"] = m["name"]
|
||||
acct["is_local"] = False
|
||||
all_accounts.append(acct)
|
||||
else:
|
||||
all_accounts.append({
|
||||
"_source_machine": m["name"],
|
||||
"_error": data.get("error", "未知错误"),
|
||||
"machine_ip": m.get("ip", ""),
|
||||
"is_local": False,
|
||||
})
|
||||
|
||||
return all_accounts
|
||||
|
||||
|
||||
def aggregate_health():
|
||||
"""聚合所有机器的健康状态（含本机）"""
|
||||
from plugins.base import HOSTNAME
|
||||
machines = _get_machines()
|
||||
results = {}
|
||||
|
||||
# 先加本机健康
|
||||
try:
|
||||
import urllib.request, json as _json
|
||||
local_resp = urllib.request.urlopen("http://localhost:9988/api/health", timeout=3)
|
||||
results[HOSTNAME] = _json.loads(local_resp.read())
|
||||
except Exception as e:
|
||||
results[HOSTNAME] = {"status": "ok", "note": "本机直连"}
|
||||
|
||||
for m in machines:
|
||||
data = fetch_machine_data(m, "/api/health")
|
||||
results[m["name"]] = data
|
||||
|
||||
return results
|
||||
|
||||
|
||||
def aggregate_status():
|
||||
"""聚合所有机器的详细状态（含本机）"""
|
||||
from plugins.base import HOSTNAME
|
||||
machines = _get_machines()
|
||||
results = {}
|
||||
|
||||
# 本机状态
|
||||
try:
|
||||
import urllib.request, json as _json
|
||||
lh = urllib.request.urlopen("http://localhost:9988/api/health", timeout=3)
|
||||
local_health = _json.loads(lh.read())
|
||||
ls = None
|
||||
try:
|
||||
ls_resp = urllib.request.urlopen("http://localhost:9988/api/machine/status", timeout=3)
|
||||
ls = _json.loads(ls_resp.read())
|
||||
except:
|
||||
pass
|
||||
results[HOSTNAME] = {
|
||||
"status": "online",
|
||||
"hostname": HOSTNAME,
|
||||
"ip": "127.0.0.1",
|
||||
"health": local_health,
|
||||
"detail": ls,
|
||||
}
|
||||
except:
|
||||
results[HOSTNAME] = {"status": "online", "hostname": HOSTNAME, "ip": "127.0.0.1"}
|
||||
|
||||
for m in machines:
|
||||
health = fetch_machine_data(m, "/api/health", timeout=3)
|
||||
if "error" in health:
|
||||
results[m["name"]] = {
|
||||
"status": "offline",
|
||||
"error": health["error"],
|
||||
"hostname": m["hostname"],
|
||||
"ip": m["ip"],
|
||||
}
|
||||
continue
|
||||
status = fetch_machine_data(m, "/api/machine/status", timeout=5)
|
||||
results[m["name"]] = {
|
||||
"status": "online",
|
||||
"hostname": m["hostname"],
|
||||
"ip": m["ip"],
|
||||
"health": health,
|
||||
"detail": status if "error" not in status else None,
|
||||
}
|
||||
|
||||
return results
|
||||
Reference in New Issue
Block a user