拆分抓取与交易服务

把需要账号登录态的链路从抓取服务里拆出成独立进程。分界线不是「要不要登录」，而是抓取无状态、幂等、可多开实例，而交易的写操作不可逆、登录态全局唯一、订单监控是常驻轮询——同进程时抓取一扩容就会复制出 N 份登录态与 N 个轮询，同一账号会被并发操作。 - app/shared：配置、错误码、日志、ApiResponse 信封 + Bearer 鉴权 + 异常处理器、导航请求头构造器 - app/scraping：站点常量、会话、解析器与 10 个抓取接口，:31107，可多开 - app/trading：登录态查询/重载与健康检查，:31108，只能单实例 - 依赖方向锁为 scraping→shared、trading→shared，两侧互不 import； tests/test_architecture.py 用 AST 检查 import 并校验两个 app 的路径不串 - 登录态 UA 在 trading 独立持有：与抓取 UA 值相同但变更理由不同，抓取 UA 为绕反爬可随时调整，登录 UA 一改可能触发设备校验使已落盘 cookie 失效 - scripts/login.py 与 AuthSession 共用 auth_site.PROFILES 与 is_logged_in，判据只写一遍 - 同一镜像两个启动命令，交易容器覆盖 command 并设 RAKUTEN_HEALTH_PORT 同时带上此前未提交的ラクマ分类接口与登录态基础设施。验证：239 个离线用例全绿；两个入口真实启动，/health 与鉴权正常。未验证：真实探测登录态（当前开发机无外网，对站点的连接全部超时）。 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-07-27 15:05:01 +08:00
parent 4250388762
commit 104d7fef6b
80 changed files with 2330 additions and 402 deletions
@@ -7,19 +7,20 @@ from pathlib import Path

 import pytest

-from app.core.errors import InvalidRequestError, ScrapeParseError
-from app.models.scrape import (
+from app.shared.errors import InvalidRequestError, ItemNotFoundError, ScrapeParseError
+from app.scraping.models.scrape import (
    RakumaAuthenticity,
    RakumaCondition,
    RakumaSearchRequest,
    RakumaSortOption,
    RakumaTransaction,
 )
-from app.parsers.rakuma.base import parse_int, parse_total_count
-from app.parsers.rakuma.item import parse_item_detail
-from app.parsers.rakuma.search import parse_search
-from app.parsers.rakuma.shop import parse_shop_detail, parse_shop_items
-from app.utils.rakuma_urls import (
+from app.scraping.parsers.rakuma.base import parse_int, parse_total_count
+from app.scraping.parsers.rakuma.category import parse_categories
+from app.scraping.parsers.rakuma.item import parse_item_detail
+from app.scraping.parsers.rakuma.search import parse_search
+from app.scraping.parsers.rakuma.shop import parse_shop_detail, parse_shop_items
+from app.scraping.utils.rakuma_urls import (
    build_item_url,
    build_search_url,
    build_shop_url,
@@ -269,6 +270,67 @@ def test_parse_shop_pages_reject_non_shop_page():
        parse_shop_items("<html><body>x</body></html>", shop_id="S", request_url="u", page=1)


+# ---- 分类树解析 ----
+
+def test_parse_categories_returns_top_level_by_default():
+    data = parse_categories(
+        fixture("rakuma_category.html"), category_id=None, include_descendants=False
+    )
+    assert data.category_id == ""
+    assert len(data.children) == 14  # 站点顶层分类固定 14 个
+    top = data.children[0]
+    assert top.category_id == "10001"
+    assert top.name == "レディース"
+    assert top.parent_id == "0"
+    assert top.is_leaf is False
+    assert top.url == "https://fril.jp/category/10001"
+    assert top.children == []  # 未要求子树时不展开
+
+
+def test_parse_categories_reads_ancestors_and_children():
+    data = parse_categories(
+        fixture("rakuma_category.html"), category_id="786", include_descendants=False
+    )
+    assert data.name == "ゲームソフト/ゲーム機本体"
+    assert [node.category_id for node in data.ancestors] == ["10007"]
+    assert data.full_name == "エンタメ/ホビー / ゲームソフト/ゲーム機本体"
+    assert [node.category_id for node in data.children] == ["787", "788", "789"]
+    assert all(node.is_leaf for node in data.children)
+    assert data.is_leaf is False
+
+
+def test_parse_categories_marks_leaf_without_children():
+    data = parse_categories(
+        fixture("rakuma_category.html"), category_id="788", include_descendants=False
+    )
+    assert data.is_leaf is True
+    assert data.children == []
+    assert data.full_name == "エンタメ/ホビー / ゲームソフト/ゲーム機本体 / 家庭用ゲームソフト"
+
+
+def test_parse_categories_can_expand_full_subtree():
+    """整棵树本来就在一次响应里，展开子树不需要多打请求"""
+    data = parse_categories(
+        fixture("rakuma_category.html"), category_id="10007", include_descendants=True
+    )
+    branch = data.children[0]
+    assert branch.category_id == "786"
+    assert [node.category_id for node in branch.children] == ["787", "788", "789"]
+
+
+def test_parse_categories_rejects_unknown_category():
+    """无效分类要报 404，而不是当成「没有子分类」返回空结果"""
+    with pytest.raises(ItemNotFoundError):
+        parse_categories(
+            fixture("rakuma_category.html"), category_id="99999999", include_descendants=False
+        )
+
+
+def test_parse_categories_rejects_page_without_tree():
+    with pytest.raises(ScrapeParseError):
+        parse_categories("<html><body>home</body></html>", category_id=None, include_descendants=False)
+
+
 # ---- 取值工具 ----

@pytest.mark.parametrize(