From 2ef0cf4a4eea57d66689a76326208237cc10bd56 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=88=98=E8=88=AA=E5=AE=87?= <3364451258@qq.com>
Date: Sun, 5 Jul 2026 01:32:04 +0800
Subject: [PATCH] fix: add path traversal protection to ingest endpoint

---
 src/server/app.py | 12 ++++++++++++
 tests/test_api.py | 20 ++++++++++++++++++++
 2 files changed, 32 insertions(+)

diff --git a/src/server/app.py b/src/server/app.py
index 8fc075b..f77b78a 100644
--- a/src/server/app.py
+++ b/src/server/app.py
@@ -12,6 +12,16 @@ from src.core.ingest import DocumentIngestor
 from src.core.search import Searcher
 
 
+def _is_safe_path(path_str: str) -> bool:
+    """检查路径是否安全: 仅允许相对路径且不含 .. 穿越."""
+    normalized = os.path.normpath(path_str)
+    if os.path.isabs(normalized):
+        return False
+    if ".." in normalized.split(os.sep):
+        return False
+    return True
+
+
 # -- 请求模型 --
 class IngestRequest(BaseModel):
     file_path: str | None = None
@@ -109,6 +119,8 @@ def ingest_document(req: IngestRequest):
     ingestor = _get_ingestor()
     try:
         if req.file_path:
+            if not _is_safe_path(req.file_path):
+                raise HTTPException(status_code=400, detail="不允许的路径")
             path = Path(req.file_path)
             if not path.exists():
                 raise HTTPException(status_code=404, detail=f"文件不存在: {req.file_path}")
diff --git a/tests/test_api.py b/tests/test_api.py
index 824a601..023c92a 100644
--- a/tests/test_api.py
+++ b/tests/test_api.py
@@ -90,3 +90,23 @@ class TestIngestEndpoint:
         assert response.status_code == 200
         data = response.json()
         assert len(data["results"]) > 0
+
+
+class TestSecurity:
+    """安全测试."""
+
+    def test_ingest_rejects_path_traversal(self, client):
+        """拒绝路径遍历攻击."""
+        response = client.post(
+            "/api/v1/ingest",
+            json={"file_path": "../../../etc/passwd"},
+        )
+        assert response.status_code in (400, 403)
+
+    def test_ingest_rejects_absolute_path(self, client):
+        """拒绝绝对路径."""
+        response = client.post(
+            "/api/v1/ingest",
+            json={"file_path": "C:\\Windows\\System32\\config\\SAM"},
+        )
+        assert response.status_code in (400, 403)