fix(opencode): handle duplicate message ID in append_message_to_session

[Leoyzen]

Summary

Fix the ValueError: Duplicate message ID regression introduced by the D14 _pending_message_ids mechanism (PR #237 on main, PR #171 on feat/dynamic-team-mode).

Bug

In the sync path (POST /message), the REST handler pre-stores the assistant message at message_routes.py:392 before calling route_message. The event bridge then tries to store the same message again with the same canonical ID (via _pending_message_ids), causing ValueError: Duplicate message ID from the storage provider — repeated for every PartDeltaEvent (850+ errors per request in production logs).

Root Cause

REST handler (message_routes.py:392)     Event bridge (opencode_event_bridge.py:449)
├─ append_message_to_session(ID=A) ✅    ├─ _pending_message_ids.pop() → ID=A
└─ route_message(message_id=A)           └─ append_message_to_session(ID=A) 💥 Duplicate

The D14 fix made the event bridge reuse the REST handler's assistant_msg_id but didn't account for the REST handler already having stored the message.

Fix

append_message_to_session now catches ValueError for duplicate message IDs and treats the write as idempotent (skip + debug log). The in-memory dict also skips duplicate appends.

Test Infrastructure Fixes

The bug went undetected because the conftest mocks didn't match real behavior:

Mock	Before	After
_mock_append_message	Unconditional append	Check duplicate IDs like real MemoryProvider
_mock_route_message	Dropped message_id in **kwargs	Pass message_id to _pending_message_ids

New Tests (test_duplicate_message_id.py)

• test_mock_append_message_raises_on_duplicate — verifies mock catches duplicates (meta-test)
• test_duplicate_assistant_message_does_not_raise — verifies fix handles double-write gracefully
• test_different_messages_both_stored — non-duplicate messages still work
• test_user_message_then_assistant_message_no_error — mixed messages OK
• test_route_message_sets_pending_message_ids — verifies mock passes message_id through

Test Results

tests/servers/opencode_server/test_duplicate_message_id.py: 5 passed
tests/servers/opencode_server/ (full suite): 674 passed, 0 failed

Closes #229

[gemini-code-assist]

Code Review

This pull request introduces graceful handling of duplicate message IDs in append_message_to_session by catching ValueError and treating duplicate writes as idempotent, preventing errors in the sync path where the REST handler pre-stores assistant messages. It also updates the test mocks and adds unit tests to verify this behavior. The reviewer suggested removing defensive None checks (via getattr) when checking for duplicate messages in memory, as the message ID fields are non-optional and guaranteed to be initialized.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

Avoid adding defensive None checks (guards) for fields or attributes that are typed as non-optional (non-nullable) and guaranteed to be initialized, to maintain consistency with codebase conventions and avoid redundant code.

Since msg.info and msg.info.id are non-optional and guaranteed to be initialized on MessageWithParts, we can access them directly instead of using getattr with default values.

Suggested change

	msg_id = getattr(msg.info, "id", None)
	already_in_memory = msg_id is not None and any(
	getattr(m.info, "id", None) == msg_id for m in messages[session_id]
	)
	msg_id = msg.info.id
	already_in_memory = any(
	m.info.id == msg_id for m in messages[session_id]
	)

References

1. Avoid adding defensive None checks (guards) for fields or attributes that are typed as non-optional (non-nullable) and guaranteed to be initialized, to maintain consistency with codebase conventions and avoid redundant code.