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)。
200 lines
6.2 KiB
Python
200 lines
6.2 KiB
Python
"""
|
||
agentos sync — 双机增量同步
|
||
|
||
检查 agent-sync/ 与本地 WorkBuddy 的差异并增量更新:
|
||
1. 技能差异检测(对比版本号和文件变化)
|
||
2. MCP 配置差异检测
|
||
3. 自动化任务差异检测
|
||
4. 增量修复差异项
|
||
"""
|
||
|
||
import sys
|
||
import json
|
||
import shutil
|
||
import hashlib
|
||
from pathlib import Path
|
||
|
||
from .utils import (
|
||
get_sync_root, get_local_root, get_python,
|
||
info, ok, warn, err, run, banner
|
||
)
|
||
from . import skill_mgr
|
||
|
||
|
||
def file_hash(path: Path) -> str:
|
||
"""计算文件 SHA256"""
|
||
h = hashlib.sha256()
|
||
with open(path, "rb") as f:
|
||
for chunk in iter(lambda: f.read(65536), b""):
|
||
h.update(chunk)
|
||
return h.hexdigest()[:16]
|
||
|
||
|
||
def diff_skills(dry_run: bool) -> list:
|
||
"""检测技能差异"""
|
||
sync_root = get_sync_root()
|
||
wb_skills = Path.home() / ".workbuddy" / "skills"
|
||
source_dir = sync_root / "02_skills"
|
||
|
||
changes = []
|
||
source_dirs = {d.name: d for d in skill_mgr.get_skill_dirs(source_dir)}
|
||
wb_dirs = {d.name: d for d in skill_mgr.get_skill_dirs(wb_skills)} if wb_skills.exists() else {}
|
||
|
||
for name, src_dir in source_dirs.items():
|
||
if name not in wb_dirs:
|
||
changes.append({"type": "missing", "name": name, "action": "install"})
|
||
continue
|
||
|
||
src_card = skill_mgr.read_skill_card(src_dir)
|
||
wb_card = skill_mgr.read_skill_card(wb_dirs[name])
|
||
|
||
if src_card["version"] != wb_card["version"]:
|
||
changes.append({
|
||
"type": "version_diff",
|
||
"name": name,
|
||
"action": "update",
|
||
"src_version": src_card["version"],
|
||
"wb_version": wb_card["version"],
|
||
})
|
||
|
||
for name in wb_dirs:
|
||
if name not in source_dirs:
|
||
changes.append({"type": "extra", "name": name, "action": "uninstall"})
|
||
|
||
return changes
|
||
|
||
|
||
def diff_mcp(dry_run: bool) -> list:
|
||
"""检测 MCP 配置差异"""
|
||
sync_root = get_sync_root()
|
||
src = sync_root / "01_core" / "mcp.json"
|
||
dst = Path.home() / ".workbuddy" / "mcp.json"
|
||
|
||
changes = []
|
||
if not src.exists():
|
||
return changes
|
||
|
||
if not dst.exists():
|
||
changes.append({"type": "missing", "name": "mcp.json", "action": "deploy"})
|
||
else:
|
||
src_hash = file_hash(src)
|
||
dst_hash = file_hash(dst)
|
||
if src_hash != dst_hash:
|
||
changes.append({"type": "content_diff", "name": "mcp.json", "action": "update"})
|
||
|
||
return changes
|
||
|
||
|
||
def diff_automations(dry_run: bool) -> list:
|
||
"""检测自动化任务差异"""
|
||
sync_root = get_sync_root()
|
||
automations_file = sync_root / "01_core" / "automations.yaml"
|
||
if not automations_file.exists():
|
||
return []
|
||
return [{"type": "info", "name": "automations.yaml", "action": "需手动在 WorkBuddy 中比对"}]
|
||
|
||
|
||
def apply_changes(changes: list, dry_run: bool, force: bool):
|
||
"""执行差异修复"""
|
||
if dry_run:
|
||
return
|
||
|
||
for ch in changes:
|
||
action = ch["action"]
|
||
|
||
if action == "install":
|
||
info(f" 安装技能: {ch['name']}")
|
||
src_dir = get_sync_root() / "02_skills" / ch["name"]
|
||
skill_mgr.install_all(get_sync_root() / "02_skills")
|
||
|
||
elif action == "update":
|
||
info(f" 更新技能: {ch['name']} ({ch.get('src_version', '?')})")
|
||
skill_mgr.install_all(get_sync_root() / "02_skills")
|
||
|
||
elif action == "uninstall":
|
||
warn(f" agent-sync 中已移除: {ch['name']}(建议手动清理 WorkBuddy)")
|
||
|
||
elif action == "deploy":
|
||
info(f" 部署 MCP 配置")
|
||
src = get_sync_root() / "01_core" / "mcp.json"
|
||
dst = Path.home() / ".workbuddy" / "mcp.json"
|
||
if dst.exists():
|
||
dst.rename(dst.with_suffix(".json.syncbak"))
|
||
shutil.copy2(str(src), str(dst))
|
||
ok(" MCP 配置已更新")
|
||
|
||
elif action == "update" and ch["name"] == "mcp.json":
|
||
info(f" 更新 MCP 配置")
|
||
src = get_sync_root() / "01_core" / "mcp.json"
|
||
dst = Path.home() / ".workbuddy" / "mcp.json"
|
||
dst.rename(dst.with_suffix(".json.syncbak"))
|
||
shutil.copy2(str(src), str(dst))
|
||
ok(" MCP 配置已更新")
|
||
|
||
|
||
def run(args):
|
||
dry_run = args.dry_run
|
||
force = args.force
|
||
|
||
banner()
|
||
info("双机增量同步" + (" [DRY-RUN]" if dry_run else ""))
|
||
print()
|
||
|
||
# 1. 检测技能差异
|
||
print("📦 技能差异:")
|
||
skill_changes = diff_skills(dry_run)
|
||
if not skill_changes:
|
||
ok(" 无差异")
|
||
else:
|
||
for ch in skill_changes:
|
||
action_icon = {"install": "➕", "update": "🔄", "uninstall": "🗑️"}.get(ch["action"], "?")
|
||
if ch["type"] == "missing":
|
||
warn(f" {action_icon} 新增技能: {ch['name']}")
|
||
elif ch["type"] == "version_diff":
|
||
info(f" {action_icon} 版本变更: {ch['name']} ({ch['wb_version']} → {ch['src_version']})")
|
||
elif ch["type"] == "extra":
|
||
warn(f" {action_icon} 多余的: {ch['name']}(源目录已移除)")
|
||
|
||
print()
|
||
|
||
# 2. MCP 差异
|
||
print("🔧 MCP 配置差异:")
|
||
mcp_changes = diff_mcp(dry_run)
|
||
if not mcp_changes:
|
||
ok(" 无差异")
|
||
else:
|
||
for ch in mcp_changes:
|
||
if ch["action"] == "deploy":
|
||
warn(f" 缺失: {ch['name']}(待部署)")
|
||
elif ch["action"] == "update":
|
||
info(f" 内容变更: {ch['name']}(待更新)")
|
||
|
||
print()
|
||
|
||
# 3. 自动化差异
|
||
print("⏰ 自动化任务差异:")
|
||
auto_changes = diff_automations(dry_run)
|
||
if not auto_changes:
|
||
ok(" 无自动化配置文件(无差异检测)")
|
||
else:
|
||
for ch in auto_changes:
|
||
info(f" {ch['name']}: {ch['action']}")
|
||
|
||
print()
|
||
|
||
# 汇总
|
||
all_changes = skill_changes + mcp_changes + [c for c in auto_changes if c["action"] != "需手动在 WorkBuddy 中比对"]
|
||
if not all_changes:
|
||
ok("完全一致,无需同步")
|
||
return
|
||
|
||
print(f"共 {len(all_changes)} 项差异")
|
||
|
||
if dry_run:
|
||
info("添加 --force 参数执行同步")
|
||
else:
|
||
# 执行同步
|
||
info("执行同步...")
|
||
apply_changes(all_changes, dry_run, force)
|
||
ok("同步完成,请重启 WorkBuddy 使更改生效")
|