feat(monitor): 抖音 Web 接口客户端 —— 绕开爬虫子进程,直接发 HTTP
Deploy VitePress site to Pages / build (push) Canceled after 0s
Deploy VitePress site to Pages / Deploy (push) Canceled after 0s

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:
2026-10-10 17:12:25 +08:00
parent 21ff01b894
commit 2112c1a870
1601 changed files with 367425 additions and 0 deletions
@@ -0,0 +1,130 @@
"""
xhs_scrape.py — 小红书抓取适配器
工具:OpenCLI xiaohongshu → agent-browser
"""
import asyncio, json, logging
from services.adapters import ScrapeAdapter
logger = logging.getLogger("dashboard.scrape.xhs")
class XhsScrapeAdapter(ScrapeAdapter):
platform = "xiaohongshu"
adapter_name = "xhs_scrape"
async def collect_user(self, user_id: str, limit: int = 20) -> list[dict]:
"""抓取指定用户的笔记列表"""
results = []
ok, data, tool = await self._try_tools(2, [
("opencli", lambda: self._opencli_user_notes(user_id, limit)),
])
if ok and data:
for item in (data if isinstance(data, list) else [data]):
results.append(self._to_schema(item))
return results
async def collect_item(self, target: str, depth: str = "light",
tool_level: int = 2) -> dict:
"""抓取单篇笔记详情"""
note_id = self._extract_note_id(target)
if not note_id:
return {"error": f"无法解析笔记ID: {target}"}
ok, data, tool = await self._try_tools(tool_level, [
("opencli", lambda: self._opencli_note(note_id)),
])
if ok and data:
return self._to_schema(data)
return {"error": "所有工具均无法抓取"}
async def collect_comments(self, item_id: str, limit: int = 20) -> list[dict]:
"""抓取笔记评论"""
ok, data, tool = await self._try_tools(2, [
("opencli", lambda: self._opencli_comments(item_id, limit)),
])
if ok and data:
return [
{"author": c.get("nickname", ""), "text": c.get("text", ""),
"likes": c.get("likes", 0)}
for c in (data if isinstance(data, list) else [data])
]
return []
async def collect_search(self, keyword: str, limit: int = 20) -> list[dict]:
"""搜索笔记"""
results = []
ok, data, tool = await self._try_tools(2, [
("opencli", lambda: self._opencli_search(keyword, limit)),
])
if ok and data:
for item in (data if isinstance(data, list) else [data]):
results.append(self._to_schema(item))
return results
# ── OpenCLI 调用 ──
async def _opencli_user_notes(self, user_id: str, limit: int):
return await self._run_opencli([
"xiaohongshu", "user", user_id,
"--limit", str(limit), "-f", "json"
])
async def _opencli_note(self, note_id: str):
return await self._run_opencli([
"xiaohongshu", "note", note_id, "-f", "json"
])
async def _opencli_comments(self, note_id: str, limit: int):
return await self._run_opencli([
"xiaohongshu", "comments", note_id,
"--limit", str(limit), "-f", "json"
])
async def _opencli_search(self, keyword: str, limit: int):
return await self._run_opencli([
"xiaohongshu", "search", keyword,
"--limit", str(limit), "-f", "json"
])
# ── URL 解析 ──
def _extract_note_id(self, target: str) -> str:
import re
m = re.search(r'xiaohongshu\.com/explore/([a-f0-9]+)', target)
if m:
return m.group(1)
m = re.search(r'xiaohongshu\.com/discovery/item/([a-f0-9]+)', target)
if m:
return m.group(1)
if target.isalnum() and len(target) >= 10:
return target
return ""
# ── 统一 Schema ──
def _to_schema(self, raw: dict) -> dict:
return {
"platform": "xiaohongshu",
"item_id": raw.get("note_id", raw.get("id", "")),
"url": raw.get("url", f"https://www.xiaohongshu.com/explore/{raw.get('note_id', '')}"),
"title": raw.get("title", ""),
"author_name": raw.get("author", raw.get("nickname", raw.get("user_name", ""))),
"author_id": raw.get("user_id", raw.get("author_id", "")),
"published_at": raw.get("create_time", raw.get("time", "")),
"text_content": raw.get("desc", raw.get("description", raw.get("text", ""))),
"tags": raw.get("tags", []),
"stats": {
"likes": raw.get("likes", raw.get("liked_count", 0)),
"comments": raw.get("comments", raw.get("comment_count", 0)),
"shares": raw.get("shares", raw.get("share_count", 0)),
"views": raw.get("views", 0),
},
"extra": {
"note_id": raw.get("note_id", ""),
"user_id": raw.get("user_id", ""),
},
"media": [
{"type": "image", "url": u} for u in (raw.get("images") or [])
] if raw.get("images") else [],
"comments": [],
}