From 9d78d3ddb1286733869da5bfcc6e6e1e2ee890e7 Mon Sep 17 00:00:00 2001
From: Idriss Sbaaoui <112825897+6ba3i@users.noreply.github.com>
Date: Mon, 2 Mar 2026 10:44:33 +0800
Subject: [PATCH] Tests: fix failling http in CI (#13301)

### What problem does this PR solve?
test_doc_sdk_routes_unit had two flaky/incorrect branch assumptions:

1. parse/stop_parsing production logic gates on doc.run, but tests used
progress, causing branch mismatch and unintended fallthrough into
mutation/DB paths.
2. stop_parsing invalid-state test asserted an outdated message
fragment, making the contract brittle.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
---
 api/apps/sdk/doc.py                           | 10 ++++++-
 .../test_doc_sdk_routes_unit.py               | 27 +++++++++++++++----
 2 files changed, 31 insertions(+), 6 deletions(-)

diff --git a/api/apps/sdk/doc.py b/api/apps/sdk/doc.py
index 3f0295e569..513a01c4c1 100644
--- a/api/apps/sdk/doc.py
+++ b/api/apps/sdk/doc.py
@@ -809,6 +809,10 @@ async def delete(tenant_id, dataset_id):
     return get_result()
 
 
+DOC_STOP_PARSING_INVALID_STATE_MESSAGE = "Can't stop parsing document that has not started or already completed"
+DOC_STOP_PARSING_INVALID_STATE_ERROR_CODE = "DOC_STOP_PARSING_INVALID_STATE"
+
+
 @manager.route("/datasets/<dataset_id>/chunks", methods=["POST"])  # noqa: F821
 @token_required
 async def parse(tenant_id, dataset_id):
@@ -947,7 +951,11 @@ async def stop_parsing(tenant_id, dataset_id):
         if not doc:
             return get_error_data_result(message=f"You don't own the document {id}.")
         if doc[0].run != TaskStatus.RUNNING.value :
-            return get_error_data_result("Can't stop parsing document that has not started or already completed")
+            return construct_json_result(
+                code=RetCode.DATA_ERROR,
+                message=DOC_STOP_PARSING_INVALID_STATE_MESSAGE,
+                data={"error_code": DOC_STOP_PARSING_INVALID_STATE_ERROR_CODE},
+            )
         # Send cancellation signal via Redis to stop background task
         cancel_all_task_of(id)
         info = {"run": "2", "progress": 0, "chunk_num": 0}
diff --git a/test/testcases/test_http_api/test_file_management_within_dataset/test_doc_sdk_routes_unit.py b/test/testcases/test_http_api/test_file_management_within_dataset/test_doc_sdk_routes_unit.py
index b1fe88f931..635d09b555 100644
--- a/test/testcases/test_http_api/test_file_management_within_dataset/test_doc_sdk_routes_unit.py
+++ b/test/testcases/test_http_api/test_file_management_within_dataset/test_doc_sdk_routes_unit.py
@@ -592,7 +592,12 @@ class TestDocRoutesUnit:
         res = _run(module.parse.__wrapped__("tenant-1", "ds-1"))
         assert "don't own the document" in res["message"]
 
-        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(progress=0.5)])
+        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(run=module.TaskStatus.RUNNING.value)])
+        monkeypatch.setattr(
+            module.DocumentService,
+            "update_by_id",
+            lambda *_args, **_kwargs: (_ for _ in ()).throw(AssertionError("update_by_id must not be called for running docs")),
+        )
         res = _run(module.parse.__wrapped__("tenant-1", "ds-1"))
         assert "currently being processed" in res["message"]
 
@@ -631,11 +636,23 @@ class TestDocRoutesUnit:
         res = _run(module.stop_parsing.__wrapped__("tenant-1", "ds-1"))
         assert "don't own the document" in res["message"]
 
-        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(progress=1)])
+        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(run=module.TaskStatus.DONE.value)])
+        monkeypatch.setattr(
+            module,
+            "cancel_all_task_of",
+            lambda *_args, **_kwargs: (_ for _ in ()).throw(AssertionError("cancel_all_task_of must not be called for non-running docs")),
+        )
+        monkeypatch.setattr(
+            module.DocumentService,
+            "update_by_id",
+            lambda *_args, **_kwargs: (_ for _ in ()).throw(AssertionError("update_by_id must not be called for non-running docs")),
+        )
         res = _run(module.stop_parsing.__wrapped__("tenant-1", "ds-1"))
-        assert "progress at 0 or 1" in res["message"]
+        assert res["code"] == module.RetCode.DATA_ERROR
+        assert res["data"]["error_code"] == module.DOC_STOP_PARSING_INVALID_STATE_ERROR_CODE
+        assert res["message"] == module.DOC_STOP_PARSING_INVALID_STATE_MESSAGE
 
-        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(progress=0.3)])
+        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(run=module.TaskStatus.RUNNING.value)])
         monkeypatch.setattr(module, "cancel_all_task_of", lambda *_args, **_kwargs: None)
         monkeypatch.setattr(module.DocumentService, "update_by_id", lambda *_args, **_kwargs: True)
         _patch_docstore(monkeypatch, module, delete=lambda *_args, **_kwargs: None)
@@ -651,7 +668,7 @@ class TestDocRoutesUnit:
         assert "Duplicate document ids" in res["message"]
 
         monkeypatch.setattr(module, "check_duplicate_ids", lambda ids, _kind: (ids, []))
-        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(progress=0.3)])
+        monkeypatch.setattr(module.DocumentService, "query", lambda **_kwargs: [_DummyDoc(run=module.TaskStatus.RUNNING.value)])
         res = _run(module.stop_parsing.__wrapped__("tenant-1", "ds-1"))
         assert res["code"] == 0