Files
姚章浩 693cd952cd feat(asset): 清理被截断的结尾图片引用,修复整单 502
仓库首次入库(此前仅存在于本地工作区与测试服构建产物中,未上传)。

改动(src/knowledge_adapter/asset_content_processor.py):
- 新增 _TRUNCATED_TRAILING_MARKDOWN_RESOURCE_IMAGE 规则,清除 chunk 末尾被截断的
  ![](resource://xxxx 残片(WeKnora 分块在图片引用中间截断时产生,旧逻辑无法匹配完整引用,
  残留 resource:// 触发 ContentProcessingError,导致整单 502)
- 回归测试 tests/test_asset_content_processor.py::
  test_truncated_trailing_markdown_resource_image_is_removed

验证证据:
- 测试服 120.76.217.51,镜像 knowledge-adapter-service:reviewed-test-aeab4b5fca2d
  (release aeab4b5fca2d8661):KB d015d05c-25c8-4d63-b2ab-e8590d3c2dfb 5 个查询全部
  http=200、records=5、残留 resource:// = 0;KB b9dacd62 无退化;容器日志无 error/exception
- 本地:pytest 141 项全通过、ruff 全通过
- 该文件与 258c31e0fe07fd08 源码的差异经逐文件 md5 比对,为唯一变更文件
2026-09-16 10:01:27 +08:00

128 lines
4.1 KiB
Python

from uuid import UUID
import httpx
from knowledge_adapter.app import create_app
from knowledge_adapter.asset_redirect import (
AssetNotFoundError,
AssetRedirectUnavailableError,
)
ASSET_ID = UUID("c7f61f89-19ce-45b1-ac60-dbeed5e2a491")
SIGNED_URL = (
"https://private-assets.oss-cn-shenzhen.aliyuncs.com/"
"weknora/10008/assets/image.png?x-oss-signature=redacted"
)
class RecordingRedirectService:
def __init__(
self,
url: str = SIGNED_URL,
*,
error: Exception | None = None,
) -> None:
self.url = url
self.error = error
self.calls: list[UUID] = []
async def get_redirect_url(self, asset_id: UUID) -> str:
self.calls.append(asset_id)
if self.error is not None:
raise self.error
return self.url
async def test_asset_route_returns_anonymous_302() -> None:
service = RecordingRedirectService()
app = create_app(asset_redirect_service=service)
transport = httpx.ASGITransport(app=app)
async with httpx.AsyncClient(
transport=transport,
base_url="http://test",
follow_redirects=False,
) as client:
response = await client.get(f"/i/{ASSET_ID}")
assert response.status_code == 302
assert response.headers["location"] == SIGNED_URL
assert response.headers["cache-control"] == "no-store"
assert response.headers["referrer-policy"] == "no-referrer"
assert response.content == b""
assert service.calls == [ASSET_ID]
async def test_asset_route_returns_empty_404_for_unknown_asset() -> None:
service = RecordingRedirectService(error=AssetNotFoundError("asset not found"))
app = create_app(asset_redirect_service=service)
transport = httpx.ASGITransport(app=app, raise_app_exceptions=False)
async with httpx.AsyncClient(transport=transport, base_url="http://test") as client:
response = await client.get(f"/i/{ASSET_ID}")
assert response.status_code == 404
assert response.content == b""
assert "asset" not in response.text.lower()
assert str(ASSET_ID) not in response.text
async def test_asset_route_rejects_malformed_uuid_before_service() -> None:
service = RecordingRedirectService()
app = create_app(asset_redirect_service=service)
transport = httpx.ASGITransport(app=app)
async with httpx.AsyncClient(transport=transport, base_url="http://test") as client:
response = await client.get("/i/not-a-uuid")
assert response.status_code == 404
assert response.content == b""
assert service.calls == []
async def test_asset_route_returns_empty_503_when_redirect_is_unavailable() -> None:
service = RecordingRedirectService(
error=AssetRedirectUnavailableError(
"metadata http://100.100.100.200?token=secret object_key=private/image.png"
)
)
app = create_app(asset_redirect_service=service)
transport = httpx.ASGITransport(app=app, raise_app_exceptions=False)
async with httpx.AsyncClient(transport=transport, base_url="http://test") as client:
response = await client.get(f"/i/{ASSET_ID}")
assert response.status_code == 503
assert response.content == b""
assert "100.100.100.200" not in response.text
assert "object_key" not in response.text
assert "secret" not in response.text
async def test_asset_route_does_not_accept_bucket_object_key_or_url_overrides() -> None:
service = RecordingRedirectService()
app = create_app(asset_redirect_service=service)
transport = httpx.ASGITransport(app=app)
async with httpx.AsyncClient(
transport=transport,
base_url="http://test",
follow_redirects=False,
) as client:
response = await client.get(
f"/i/{ASSET_ID}",
params={
"bucket": "attacker-bucket",
"object_key": "other/secret.png",
"url": "https://attacker.example/file",
},
headers={
"X-OSS-Bucket": "attacker-bucket",
"X-Object-Key": "other/secret.png",
},
)
assert response.status_code == 302
assert response.headers["location"] == SIGNED_URL
assert service.calls == [ASSET_ID]