Skip to content

Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

fix(queuefs): give SemanticMsg a per-instance creation timestamp (#4757)

now-inga3416bb830b0
diff --git a/openviking/storage/queuefs/semantic_msg.py b/openviking/storage/queuefs/semantic_msg.py
index d61e984ec..5a60c8161 100644
--- a/openviking/storage/queuefs/semantic_msg.py
+++ b/openviking/storage/queuefs/semantic_msg.py
@@ -47,7 +47,7 @@ class SemanticMsg:
uri: str # Directory URI
context_type: str # resource, memory, skill, session
status: str = "pending" # pending/processing/completed
- timestamp: int = int(datetime.now().timestamp())
+ timestamp: int = field(default_factory=lambda: int(datetime.now().timestamp()))
recursive: bool = True # Whether to recursively process subdirectories
account_id: str = "default"
user_id: str = "default"
@@ -102,6 +102,7 @@ class SemanticMsg:
copy_source_uri: str = "",
):
self.id = str(uuid4())
+ self.timestamp = int(datetime.now().timestamp())
self.uri = uri
self.context_type = context_type
self.recursive = recursive
diff --git a/tests/storage/test_semantic_msg_timestamp.py b/tests/storage/test_semantic_msg_timestamp.py
new file mode 100644
index 000000000..52ecb3735
--- /dev/null
+++ b/tests/storage/test_semantic_msg_timestamp.py
@@ -0,0 +1,57 @@
+# Copyright (c) 2026 Beijing Volcano Engine Technology Co., Ltd.
+# SPDX-License-Identifier: AGPL-3.0
+"""SemanticMsg.timestamp must reflect per-instance creation time.
+
+Regression guard: the field default used to be evaluated once at class
+definition (module import), so every message created by a long-running
+process carried the process-start timestamp instead of its own creation
+time.
+"""
+
+from datetime import datetime
+
+from openviking.storage.queuefs.semantic_msg import SemanticMsg
+
+
+def _now_epoch() -> int:
+ return int(datetime.now().timestamp())
+
+
+def test_new_message_timestamp_is_own_creation_time():
+ before = _now_epoch()
+ msg = SemanticMsg(uri="viking://res/a", context_type="resource")
+ after = _now_epoch()
+
+ # Instance-level attribute (falls back to the class attribute only when
+ # unset — the frozen-timestamp regression is exactly that fallback).
+ assert "timestamp" in msg.__dict__
+ assert before <= msg.timestamp <= after
+
+
+def test_messages_do_not_share_a_frozen_timestamp():
+ first = SemanticMsg(uri="viking://res/a", context_type="resource")
+ first.timestamp = 1111111111 # mutate one instance only
+
+ second = SemanticMsg(uri="viking://res/b", context_type="memory")
+ assert second.timestamp != 1111111111
+ assert second.timestamp != first.timestamp
+
+
+def test_to_dict_carries_creation_timestamp():
+ msg = SemanticMsg(uri="viking://res/a", context_type="resource")
+ data = msg.to_dict()
+ assert data["timestamp"] == msg.timestamp
+ assert abs(data["timestamp"] - _now_epoch()) <= 5
+
+
+def test_from_dict_preserves_stored_timestamp():
+ msg = SemanticMsg.from_dict(
+ {"uri": "viking://res/a", "context_type": "resource", "timestamp": 1234567890}
+ )
+ assert msg.timestamp == 1234567890
+
+
+def test_from_dict_without_timestamp_still_gets_fresh_one():
+ msg = SemanticMsg.from_dict({"uri": "viking://res/a", "context_type": "resource"})
+ assert "timestamp" in msg.__dict__
+ assert abs(msg.timestamp - _now_epoch()) <= 5