Files
MediaCrawler/tests/test_monitor_api.py
T
butubb 06718a1351
Deploy VitePress site to Pages / build (push) Canceled after 0s
Deploy VitePress site to Pages / Deploy (push) Canceled after 0s
feat(monitor): 抖音接入博主监控
上游爬虫本身不缺抖音能力(三模式、四项指标、二级评论都与小红书对等、指标还是同名同列),
缺的全在监控层的适配。这次把「平台之间不一样」的管子集中到一个新模块,再把散落的
xhs 硬编码接上去。

* 新增 api/monitor/adapters.py:产物目录名、jsonl 字段别名、目标链接形态与正则、
  通知链接模板。不放进 platforms.py 是因为那个模块被 describe_all() 整个序列化进
  /api/config/platforms 交给前端,塞进正则和目录名会让爬虫内部细节漏进 API 载荷。
  代价是两个注册表可能漂移,用一条测试钉住「声明接通就必须有适配器」。
* 两个必须知道的坑,都在这版里处理掉了:
  1) 抖音的平台 id 是 dy,而 store 把产物写在 douyin/ 下(store/douyin/_store_impl.py:47)。
     不改就是 ingest 一个文件都读不到 —— 不报错,只是 0 条,然后被冒充成「疑似登录失效」。
  2) 抖音的作品没有 note_id(叫 aweme_id)、评论也用 aweme_id 指作品。ingest 第一步是
     `if not note_id: continue`,不映射就逐条全丢。
  另外抖音顶层评论的 parent_comment_id 是字符串 "0",归一成空串,免得前端多出悬空的父节点。
* 顺带把「东西抓到了、只是没落在期望目录里」单独识别出来。这类故障的现象和登录失效
  一模一样,按登录失效报会把人指去查完全错误的方向。
* 修两个既有 bug(今天只有小红书所以无害,加抖音就踩响):
  - service.py update_task 换目标时漏传 task.platform,回落到默认小红书
  - scheduler.py 取 cookie 没传 platform,抖音任务会读着小红书那份 cookie 不动
* 行为变更(已与用户确认):cookie 闸门改成「没 cookie 且没开 CDP」才跳过。
  CDP 模式下登录态来自被接管的浏览器,粘不粘 cookie 由不得它决定;不放行的话,
  选了「接管已有 Chrome」却没粘 cookie 的用户会看到任务永远不触发,而且不报错。
  副作用是开启了 CDP 的小红书任务也不再被该闸门拦住 —— 语义上是对的。
* 目标输入框的示例链接与措辞改由能力矩阵提供(notes_label 抖音说「作品」、小红书说
  「笔记」;「建议只填纯 ID」是小红书专属劝告,抖音链接不带令牌,不再显示)。

测试 +22 条(858 通过),其中最关键的是「抖音作品/评论不被静默丢弃」与「产物目录名
不等于平台 id」两条 —— 都是把最难查的失败模式钉死在回归网里。

注意:抖音这条路的**端到端尚未验证**,需要一份可用的抖音登录态(CDP 那台 Chrome 里
登录,或导出一份 cookie)。单测覆盖的是解析与入库,真实抓取还没跑过。
2026-10-10 14:55:28 +08:00

320 lines
13 KiB
Python

# -*- coding: utf-8 -*-
# Copyright (c) 2025 [email protected]
#
# This file is part of MediaCrawler project.
# Repository: https://github.com/NanmiCoder/MediaCrawler/blob/main/tests/test_monitor_api.py
# GitHub: https://github.com/NanmiCoder
# Licensed under NON-COMMERCIAL LEARNING LICENSE 1.1
#
# 声明:本代码仅供学习和研究目的使用。使用者应遵守以下原则:
# 1. 不得用于任何商业用途。
# 2. 使用时应遵守目标平台的使用条款和robots.txt规则。
# 3. 不得进行大规模爬取或对平台造成运营干扰。
# 4. 应合理控制请求频率,避免给目标平台带来不必要的负担。
# 5. 不得用于任何非法或不当的用途。
#
# 详细许可条款请参阅项目根目录下的LICENSE文件。
# 使用本代码即表示您同意遵守上述原则和LICENSE中的所有条款。
"""API-level tests for the monitoring endpoints.
Run against an ASGI transport with a temporary database, so no server, network
or login is required. Lifespan is deliberately not exercised: it would start the
scheduler, and these tests only cover routing, validation and persistence.
"""
import httpx
import pytest
import pytest_asyncio
from api.main import app
from api.monitor import db as monitor_db
from api.monitor.service import TargetParseError, parse_target_input
CREATOR_URL = (
"https://www.xiaohongshu.com/user/profile/5f58bd990000000001003753"
"?xsec_token=ABYVg1evluJZZzpMX-VWzchxQ1qSNVW3r-jOEnKqMcgZw=&xsec_source=pc_search"
)
NOTE_URL = "https://www.xiaohongshu.com/explore/6aa3d827000000002802c5c8?xsec_token=TOKEN&xsec_source=pc_search"
@pytest_asyncio.fixture
async def client(tmp_path):
monitor_db.set_sqlite_path(tmp_path / "monitor.db")
await monitor_db.init_db()
transport = httpx.ASGITransport(app=app)
async with httpx.AsyncClient(transport=transport, base_url="http://test") as http_client:
yield http_client
await monitor_db.dispose_engine()
class TestParseTargetInput:
def test_full_url_splits_id_from_token(self):
"""The id is the stable key; the token is a refreshable credential."""
parsed = parse_target_input(CREATOR_URL, "creator")
assert parsed["external_id"] == "5f58bd990000000001003753"
assert parsed["xsec_token"].startswith("ABYVg1evluJZZzpMX")
assert parsed["xsec_source"] == "pc_search"
def test_bare_id_is_accepted(self):
parsed = parse_target_input("5f58bd990000000001003753", "creator")
assert parsed["external_id"] == "5f58bd990000000001003753"
assert parsed["xsec_token"] == ""
def test_note_url_without_token_still_parses(self):
parsed = parse_target_input(
"https://www.xiaohongshu.com/explore/6aa3d827000000002802c5c8", "note"
)
assert parsed["external_id"] == "6aa3d827000000002802c5c8"
assert parsed["xsec_token"] == ""
def test_creator_url_rejected_in_note_mode(self):
with pytest.raises(TargetParseError):
parse_target_input(CREATOR_URL, "note")
def test_garbage_is_rejected(self):
with pytest.raises(TargetParseError):
parse_target_input("not a url at all !!", "creator")
# --- 抖音 -------------------------------------------------------------
# 链接形态由平台决定,所以每一个都要显式带上 "dy"。
def test_douyin_creator_url(self):
parsed = parse_target_input(
"https://www.douyin.com/user/MS4wLjABAAAATJPY7LAlaa5X-c8uNdWkvz0jUGgpw4eeXIwu_8BhvqE"
"?from_tab_name=main",
"creator",
"dy",
)
assert (
parsed["external_id"]
== "MS4wLjABAAAATJPY7LAlaa5X-c8uNdWkvz0jUGgpw4eeXIwu_8BhvqE"
)
def test_douyin_video_url(self):
parsed = parse_target_input(
"https://www.douyin.com/video/7525082444551310602", "note", "dy"
)
assert parsed["external_id"] == "7525082444551310602"
def test_douyin_modal_id_url(self):
"""在别人主页或搜索结果里点开视频,拿到的就是带 modal_id 的链接。"""
parsed = parse_target_input(
"https://www.douyin.com/root/search/python?aid=b733a3b0&modal_id=7471165520058862848",
"note",
"dy",
)
assert parsed["external_id"] == "7471165520058862848"
def test_douyin_bare_sec_uid_is_accepted(self):
sec_uid = "MS4wLjABAAAATJPY7LAlaa5X-c8uNdWkvz0jUGgpw4eeXIwu_8BhvqE"
parsed = parse_target_input(sec_uid, "creator", "dy")
assert parsed["external_id"] == sec_uid
def test_douyin_bare_sec_uid_beyond_the_xhs_length_cap(self):
"""裸 id 的长度上限必须按平台分开。
小红书那条规则封顶 64 字符,而 sec_user_id 长过 64 是常态(实测样本 55,
但字段本身是变长的)。共用一条规则的话,长一点的 sec_uid 会被直接拒掉 ——
对用户来说就是「粘贴了一个完全正确的链接却报无法识别」。
"""
sec_uid = "MS4wLjABAAAA" + "aB3dEf6hIj9lMn2pQr5tUv8xYz1" * 3
assert len(sec_uid) > 64
parsed = parse_target_input(sec_uid, "creator", "dy")
assert parsed["external_id"] == sec_uid
# 同一条 id 拿小红书规则来解析会被拒 —— 这正是两条规则必须分开的原因。
with pytest.raises(TargetParseError):
parse_target_input(sec_uid, "creator", "xhs")
def test_douyin_bare_video_id(self):
parsed = parse_target_input("7525082444551310602", "note", "dy")
assert parsed["external_id"] == "7525082444551310602"
# 抖音不需要 xsec_token —— 和小红书不同,裸链接就能用。
assert parsed["xsec_token"] == ""
def test_douyin_short_link_is_rejected_with_a_reason(self):
"""短链要联网跳一次才知道指向谁。明确拒绝好过存一个永远抓不到东西的目标。"""
with pytest.raises(TargetParseError) as excinfo:
parse_target_input("https://v.douyin.com/drIPtQ_WPWY/", "note", "dy")
assert "短链" in str(excinfo.value)
def test_a_douyin_link_is_not_parsed_with_xhs_rules(self):
with pytest.raises(TargetParseError):
parse_target_input(
"https://www.douyin.com/video/7525082444551310602", "note", "xhs"
)
def test_an_xhs_link_is_not_parsed_for_douyin(self):
with pytest.raises(TargetParseError):
parse_target_input(NOTE_URL, "note", "dy")
def test_a_platform_without_an_adapter_is_rejected(self):
with pytest.raises(TargetParseError):
parse_target_input("whatever", "creator", "bili")
class TestTargetReplacement:
@pytest.mark.asyncio
async def test_replacing_targets_uses_the_tasks_own_platform(self, client):
"""改目标必须按任务**自己**的平台解析。
``update_task`` 原先漏传了 platform,解析回落到默认的小红书。只有小红书时
行为恰好正确,接上抖音就会拿小红书的正则去解析抖音链接 —— 建任务时对、
改任务时错,是最难注意到的那种不一致。
"""
sec_uid = "MS4wLjABAAAATJPY7LAlaa5X-c8uNdWkvz0jUGgpw4eeXIwu_8BhvqE"
created = await client.post(
"/api/monitor/tasks",
json={"name": "抖音", "mode": "creator", "platform": "dy", "targets": [sec_uid]},
)
assert created.status_code == 201
task_id = created.json()["id"]
updated = await client.patch(
f"/api/monitor/tasks/{task_id}",
json={"targets": [f"https://www.douyin.com/user/{sec_uid}"]},
)
assert updated.status_code == 200
tasks = (
await client.get("/api/monitor/tasks", params={"platform": "dy"})
).json()["tasks"]
target = tasks[0]["targets"][0]
assert target["external_id"] == sec_uid
assert target["raw_value"].startswith("https://www.douyin.com/user/")
class TestTaskCrud:
@pytest.mark.asyncio
async def test_create_and_list_task(self, client):
response = await client.post(
"/api/monitor/tasks",
json={
"name": "网文作者监控",
"mode": "creator",
"interval_minutes": 120,
"targets": [CREATOR_URL, "5f58bd990000000001003754"],
},
)
assert response.status_code == 201
task_id = response.json()["id"]
listing = await client.get("/api/monitor/tasks")
assert listing.status_code == 200
tasks = listing.json()["tasks"]
assert len(tasks) == 1
assert tasks[0]["id"] == task_id
assert tasks[0]["target_count"] == 2
# next_run_at is persisted so the schedule survives a restart.
assert tasks[0]["next_run_at"] is not None
@pytest.mark.asyncio
async def test_duplicate_targets_are_deduplicated(self, client):
response = await client.post(
"/api/monitor/tasks",
json={
"name": "dedup",
"mode": "creator",
"targets": [CREATOR_URL, CREATOR_URL],
},
)
assert response.status_code == 201
listing = await client.get("/api/monitor/tasks")
assert listing.json()["tasks"][0]["target_count"] == 1
@pytest.mark.asyncio
async def test_invalid_target_returns_400(self, client):
response = await client.post(
"/api/monitor/tasks",
json={"name": "bad", "mode": "creator", "targets": ["!!! nonsense !!!"]},
)
assert response.status_code == 400
@pytest.mark.asyncio
async def test_interval_floor_is_enforced(self, client):
"""A tight poll loop is the pattern that triggers platform rate limits."""
response = await client.post(
"/api/monitor/tasks",
json={"name": "too fast", "mode": "creator", "interval_minutes": 1, "targets": [CREATOR_URL]},
)
assert response.status_code == 422
@pytest.mark.asyncio
async def test_update_and_delete(self, client):
created = await client.post(
"/api/monitor/tasks",
json={"name": "t", "mode": "note", "targets": [NOTE_URL]},
)
task_id = created.json()["id"]
patched = await client.patch(f"/api/monitor/tasks/{task_id}", json={"enabled": False})
assert patched.status_code == 200
listing = await client.get("/api/monitor/tasks")
assert listing.json()["tasks"][0]["enabled"] is False
deleted = await client.delete(f"/api/monitor/tasks/{task_id}")
assert deleted.status_code == 200
assert (await client.get("/api/monitor/tasks")).json()["tasks"] == []
@pytest.mark.asyncio
async def test_run_now_on_missing_task_is_404(self, client):
response = await client.post("/api/monitor/tasks/9999/run")
assert response.status_code == 404
@pytest.mark.asyncio
async def test_run_history_starts_empty(self, client):
created = await client.post(
"/api/monitor/tasks",
json={"name": "t", "mode": "creator", "targets": [CREATOR_URL]},
)
task_id = created.json()["id"]
runs = await client.get(f"/api/monitor/tasks/{task_id}/runs")
assert runs.status_code == 200
assert runs.json()["runs"] == []
class TestCookieEndpoints:
@pytest.mark.asyncio
async def test_cookie_value_is_never_returned(self, client):
"""The GET must expose health only, never the credential."""
secret = "web_session=SUPERSECRETVALUE; a1=abc123"
saved = await client.post("/api/monitor/cookie", json={"cookie": secret})
assert saved.status_code == 200
status_response = await client.get("/api/monitor/cookie")
assert status_response.status_code == 200
body = status_response.json()
assert body["present"] is True
assert body["length"] == len(secret)
assert "SUPERSECRETVALUE" not in status_response.text
@pytest.mark.asyncio
async def test_cookie_initially_absent_and_clearable(self, client):
assert (await client.get("/api/monitor/cookie")).json()["present"] is False
await client.post("/api/monitor/cookie", json={"cookie": "web_session=x"})
assert (await client.get("/api/monitor/cookie")).json()["present"] is True
await client.delete("/api/monitor/cookie")
assert (await client.get("/api/monitor/cookie")).json()["present"] is False
class TestDashboardQueries:
@pytest.mark.asyncio
async def test_empty_dashboard_shapes(self, client):
assert (await client.get("/api/monitor/notes")).json()["notes"] == []
assert (await client.get("/api/monitor/comments")).json()["comments"] == []
assert (await client.get("/api/monitor/events")).json()["events"] == []
overview = (await client.get("/api/monitor/overview")).json()
assert overview["tasks"] == 0
assert overview["notes"] == 0