From e7b915592c8d68dede0e87082d4a32b2bc51a643 Mon Sep 17 00:00:00 2001
From: nankingjing <1079826437@qq.com>
Date: Fri, 3 Jul 2026 13:11:15 +0800
Subject: [PATCH] fix(state): avoid double-encoding reasoning columns on fork
 round-trip

Fixes #57240

get_messages() leaves reasoning_details/codex_* columns as stored TEXT.
replace_messages() must not json.dumps() those strings again or forked
sessions silently lose reasoning replay on resume.

Add _encode_json_column() and use it in append_message and
_insert_message_rows so pre-serialized TEXT passes through unchanged.
---
 hermes_state.py            | 41 +++++++++++++++++++-------------------
 tests/test_hermes_state.py | 28 ++++++++++++++++++++++++++
 2 files changed, 48 insertions(+), 21 deletions(-)

diff --git a/hermes_state.py b/hermes_state.py
index ac1a126675322..d94d7ca5b3992 100644
--- a/hermes_state.py
+++ b/hermes_state.py
@@ -3092,6 +3092,20 @@ def _decode_content(cls, content: Any) -> Any:
                 return content
         return content
 
+    @classmethod
+    def _encode_json_column(cls, value: Any) -> Optional[str]:
+        """Serialize a JSON column value for storage.
+
+        Accepts either a live structure (list/dict) or the column's own
+        already-serialized TEXT from ``get_messages()`` — the latter must
+        not be double-encoded on ``replace_messages()`` round-trips (#57240).
+        """
+        if not value:
+            return None
+        if isinstance(value, str):
+            return value
+        return json.dumps(value)
+
     def append_message(
         self,
         session_id: str,
@@ -3124,18 +3138,9 @@ def append_message(
         message by its platform-side identifier.
         """
         # Serialize structured fields to JSON before entering the write txn
-        reasoning_details_json = (
-            json.dumps(reasoning_details)
-            if reasoning_details else None
-        )
-        codex_items_json = (
-            json.dumps(codex_reasoning_items)
-            if codex_reasoning_items else None
-        )
-        codex_message_items_json = (
-            json.dumps(codex_message_items)
-            if codex_message_items else None
-        )
+        reasoning_details_json = self._encode_json_column(reasoning_details)
+        codex_items_json = self._encode_json_column(codex_reasoning_items)
+        codex_message_items_json = self._encode_json_column(codex_message_items)
         tool_calls_json = json.dumps(tool_calls) if tool_calls else None
         # Multimodal content (list of parts) must be JSON-encoded: sqlite3
         # cannot bind list/dict parameters directly.
@@ -3232,15 +3237,9 @@ def _insert_message_rows(self, conn, session_id: str, messages: List[Dict[str, A
             codex_message_items = (
                 msg.get("codex_message_items") if role == "assistant" else None
             )
-            reasoning_details_json = (
-                json.dumps(reasoning_details) if reasoning_details else None
-            )
-            codex_items_json = (
-                json.dumps(codex_reasoning_items) if codex_reasoning_items else None
-            )
-            codex_message_items_json = (
-                json.dumps(codex_message_items) if codex_message_items else None
-            )
+            reasoning_details_json = self._encode_json_column(reasoning_details)
+            codex_items_json = self._encode_json_column(codex_reasoning_items)
+            codex_message_items_json = self._encode_json_column(codex_message_items)
             tool_calls_json = json.dumps(tool_calls) if tool_calls else None
             # Accept either `platform_message_id` (new explicit name) or
             # `message_id` (yuanbao's existing convention on message dicts).
diff --git a/tests/test_hermes_state.py b/tests/test_hermes_state.py
index eb884129bc56f..d90668d4faa82 100644
--- a/tests/test_hermes_state.py
+++ b/tests/test_hermes_state.py
@@ -1,5 +1,6 @@
 """Tests for hermes_state.py — SessionDB SQLite CRUD, FTS5 search, export."""
 
+import json
 import sqlite3
 import time
 import pytest
@@ -847,6 +848,33 @@ def test_replace_messages_handles_multimodal_content(self, db):
         assert msgs[0]["content"] == content
         assert msgs[1]["content"] == "I see a screenshot."
 
+    def test_get_messages_roundtrip_preserves_reasoning_columns(self, db):
+        """get_messages → replace_messages must not double-encode reasoning JSON (#57240)."""
+        reasoning = [{"type": "thinking", "thinking": "step one"}]
+        codex_items = [{"type": "reasoning", "encrypted_content": "enc_abc"}]
+        db.create_session(session_id="parent", source="api_server")
+        db.append_message(
+            "parent",
+            role="assistant",
+            content="answer",
+            reasoning_details=reasoning,
+            codex_reasoning_items=codex_items,
+        )
+
+        # get_messages leaves reasoning columns as stored TEXT — that is the
+        # fork/branch copy path (POST /api/sessions/{id}/fork).
+        loaded = db.get_messages("parent")
+        assert json.loads(loaded[0]["reasoning_details"]) == reasoning
+        assert json.loads(loaded[0]["codex_reasoning_items"]) == codex_items
+
+        fork_id = "parent-fork"
+        db.create_session(session_id=fork_id, source="api_server", parent_session_id="parent")
+        db.replace_messages(fork_id, db.get_messages("parent"))
+
+        fork_conv = db.get_messages_as_conversation(fork_id)
+        assert fork_conv[0]["reasoning_details"] == reasoning
+        assert fork_conv[0]["codex_reasoning_items"] == codex_items
+
     def test_get_messages_as_conversation(self, db):
         db.create_session(session_id="s1", source="cli")
         db.append_message("s1", role="user", content="Hello")