feat(api): add web search endpoint

2026-06-09 16:13:24 +08:00
parent 441979f581
commit 1712ecd4c7
7 changed files with 256 additions and 35 deletions
@@ -0,0 +1,115 @@
+import asyncio
+import importlib.util
+from pathlib import Path
+
+import httpx
+import pytest
+
+
+def _load_web_search_module():
+    module_path = (
+        Path(__file__).resolve().parents[2] / "app" / "services" / "web_search.py"
+    )
+    spec = importlib.util.spec_from_file_location("tests_web_search_under_test", module_path)
+    module = importlib.util.module_from_spec(spec)
+    assert spec and spec.loader
+    spec.loader.exec_module(module)
+    return module
+
+
+web_search = _load_web_search_module()
+
+
+class FakeClient:
+    def __init__(self, response):
+        self.response = response
+        self.calls = []
+
+    async def post(self, url, *, headers, json):
+        self.calls.append({"url": url, "headers": headers, "json": json})
+        return self.response
+
+
+def test_search_bocha_web_posts_expected_payload(monkeypatch):
+    monkeypatch.setattr(web_search.settings, "BOCHA_API_KEY", "sk-test")
+    monkeypatch.setattr(
+        web_search.settings,
+        "BOCHA_WEB_SEARCH_URL",
+        "https://api.bochaai.com/v1/web-search",
+    )
+    response = httpx.Response(
+        200,
+        json={"data": {"webPages": {"value": []}}},
+        request=httpx.Request("POST", "https://api.bochaai.com/v1/web-search"),
+    )
+    client = FakeClient(response)
+
+    result = asyncio.run(
+        web_search.search_bocha_web(
+            web_search.WebSearchRequest(
+                query="天津水务",
+                freshness="oneWeek",
+                summary=True,
+                count=5,
+                include=["example.com", "news.example.com"],
+                exclude=["spam.example.com"],
+            ),
+            client=client,
+        )
+    )
+
+    assert result == {"data": {"webPages": {"value": []}}}
+    assert client.calls == [
+        {
+            "url": "https://api.bochaai.com/v1/web-search",
+            "headers": {
+                "Authorization": "Bearer sk-test",
+                "Content-Type": "application/json",
+            },
+            "json": {
+                "query": "天津水务",
+                "freshness": "oneWeek",
+                "summary": True,
+                "count": 5,
+                "include": "example.com,news.example.com",
+                "exclude": "spam.example.com",
+            },
+        }
+    ]
+
+
+def test_search_bocha_web_requires_api_key(monkeypatch):
+    monkeypatch.setattr(web_search.settings, "BOCHA_API_KEY", "")
+
+    with pytest.raises(web_search.BochaSearchConfigError):
+        asyncio.run(
+            web_search.search_bocha_web(
+                web_search.WebSearchRequest(query="天津水务"),
+                client=FakeClient(httpx.Response(200, json={})),
+            )
+        )
+
+
+def test_search_bocha_web_surfaces_upstream_error(monkeypatch):
+    monkeypatch.setattr(web_search.settings, "BOCHA_API_KEY", "sk-test")
+    response = httpx.Response(
+        401,
+        json={"error": "invalid api key"},
+        request=httpx.Request("POST", "https://api.bochaai.com/v1/web-search"),
+    )
+
+    with pytest.raises(web_search.BochaSearchAPIError) as exc_info:
+        asyncio.run(
+            web_search.search_bocha_web(
+                web_search.WebSearchRequest(query="天津水务"),
+                client=FakeClient(response),
+            )
+        )
+
+    assert exc_info.value.status_code == 401
+    assert exc_info.value.detail == {"error": "invalid api key"}
+
+
+def test_web_search_request_validates_count_range():
+    with pytest.raises(ValueError):
+        web_search.WebSearchRequest(query="天津水务", count=51)