From 381091df71c23a46754be806ee53507a082fbcf9 Mon Sep 17 00:00:00 2001
From: bohdansolovie <153934212+bohdansolovie@users.noreply.github.com>
Date: Thu, 11 Jun 2026 15:52:59 +0800
Subject: [PATCH] fix(dialog): guard async_ask() against empty or invalid
 kb_ids (#15530)

Fixes #15529 .

### Problem

`async_ask()` accessed `kbs[0]` without verifying that
`KnowledgebaseService.get_by_ids()` returned any knowledge bases. Empty
or stale `kb_ids` raised `IndexError`, which surfaced as HTTP 500 on
search/bot SSE endpoints.

### Fix

- Add an early guard when `kbs` is empty, yielding a final SSE error
event (consistent with `gen_mindmap()` in the same module).
- Add regression tests for empty `kb_ids` and deleted/invalid KB IDs.

### Test plan

- [ ] `pytest
test/unit_test/api/db/services/test_dialog_service_final_answer.py -k
"async_ask_empty or async_ask_stale"`
- [ ] Manual: `POST /api/v1/searchbots/ask` with invalid `kb_ids`
returns SSE error, not HTTP 500

---------

Co-authored-by: Wang Qi <wangq8@outlook.com>
---
 api/db/services/dialog_service.py             |  8 ++++
 .../test_dialog_service_final_answer.py       | 47 +++++++++++++++++++
 2 files changed, 55 insertions(+)

diff --git a/api/db/services/dialog_service.py b/api/db/services/dialog_service.py
index b4af21b587..72e8c1cf1d 100644
--- a/api/db/services/dialog_service.py
+++ b/api/db/services/dialog_service.py
@@ -1600,6 +1600,14 @@ async def async_ask(question, kb_ids, tenant_id, chat_llm_name=None, search_conf
     include_reference_metadata, metadata_fields = _resolve_reference_metadata(search_config)
 
     kbs = KnowledgebaseService.get_by_ids(kb_ids)
+    if not kbs:
+        if not kb_ids:
+            error = "**ERROR**: No KB selected"
+        else:
+            error = "**ERROR**: The selected KB is not valid"
+        yield {"answer": error, "reference": {}, "final": True}
+        return
+
     embedding_list = list(set([kb.embd_id for kb in kbs]))
 
     is_knowledge_graph = all([kb.parser_id == ParserType.KG for kb in kbs])
diff --git a/test/unit_test/api/db/services/test_dialog_service_final_answer.py b/test/unit_test/api/db/services/test_dialog_service_final_answer.py
index 44b2bf570a..7fefb323a9 100644
--- a/test/unit_test/api/db/services/test_dialog_service_final_answer.py
+++ b/test/unit_test/api/db/services/test_dialog_service_final_answer.py
@@ -292,6 +292,53 @@ def test_async_ask_delta_events_carry_incremental_text_only(monkeypatch):
     )
 
 
+@pytest.mark.p2
+def test_async_ask_empty_kb_ids_yields_error_final_event(monkeypatch):
+    """
+    When kb_ids is empty, async_ask() must not crash with IndexError on kbs[0].
+    """
+    monkeypatch.setattr(
+        dialog_service.KnowledgebaseService, "get_by_ids", lambda _ids: []
+    )
+
+    events = _collect(
+        dialog_service.async_ask(
+            question="What is RAGFlow?",
+            kb_ids=[],
+            tenant_id="tenant-1",
+        )
+    )
+
+    assert len(events) == 1
+    final = events[0]
+    assert final.get("final") is True
+    assert "No KB selected" in final["answer"]
+    assert final["reference"] == {}
+
+
+@pytest.mark.p2
+def test_async_ask_stale_kb_ids_yields_error_final_event(monkeypatch):
+    """Provided kb_ids that do not resolve to any KB should report invalid selection."""
+    monkeypatch.setattr(
+        dialog_service.KnowledgebaseService,
+        "get_by_ids",
+        lambda ids: [] if ids == ["deleted-kb"] else [_KB],
+    )
+
+    events = _collect(
+        dialog_service.async_ask(
+            question="What is RAGFlow?",
+            kb_ids=["deleted-kb"],
+            tenant_id="tenant-1",
+        )
+    )
+
+    assert len(events) == 1
+    assert events[0].get("final") is True
+    assert "not valid" in events[0]["answer"]
+    assert events[0]["reference"] == {}
+
+
 # ---------------------------------------------------------------------------
 # Tests for async_chat  (production code path)
 # ---------------------------------------------------------------------------