hermes-agent/tests/tui_gateway
ethernet 3133af8215
fix(desktop): prevent duplicate messages when verification candidates are persisted (#68149)
The display_history_prefix calculation used by session.resume's
_live_session_payload was display_history[:len(display) - len(raw)].
This assumed the model (repaired) history is always a suffix of the
display history — i.e., repair_message_sequence only removes messages
from the tail. That assumption broke when verification candidates
(finish_reason=verification_required) were persisted to state.db (#65919):

  - repair collapses consecutive assistant messages, removing the
    verification candidate from the MODEL history
  - the candidate stays in the DISPLAY history (it's real persisted content)
  - the length gap (gap = len(display) - len(raw)) counts BOTH ancestor
    messages AND repair-removed tip messages
  - the prefix = display[:gap] grabs the first N display messages, which
    are tip messages (not ancestors) when there are no compression ancestors
  - _live_session_payload concatenates prefix + model_history, duplicating
    the first N messages

On session 20260720_110036_a33889 (8 verification candidates), this
duplicated the first 8 messages in every warm-cache session.activate
response, producing visible duplicate user messages in the desktop.

Fix: add SessionDB.get_ancestor_display_prefix() which returns ONLY
genuine ancestor messages (rows where session_id != tip_session_id),
identified at the row level before _rows_to_conversation strips
session_id. Both resume paths (eager + deferred) now use this instead
of the length-slice heuristic.

Tests:
  - test_get_ancestor_display_prefix_single_session_returns_empty
  - test_get_ancestor_display_prefix_returns_ancestor_only_messages
  - Updated 12 mock DBs across test_protocol.py + test_tui_gateway_server.py
  - 848 passed (run_tests.sh), 0 regressions
2026-07-20 17:14:41 +00:00
..
__init__.py
test_billing_rpc.py feat(billing): /billing terminal billing — interactive TUI + CLI client (#45449) 2026-06-19 01:53:32 +05:30
test_codex_app_server_live_events.py feat(codex): stream live app-server events to TUI/desktop tool cards 2026-07-17 13:44:12 -07:00
test_compaction_status.py fix(desktop): show summarizing indicator during auto-compaction 2026-06-14 02:28:07 -05:00
test_compute_host.py feat(dashboard): isolate turns in compute host (#65895) 2026-07-16 15:24:03 -04:00
test_compute_host_phase1.py feat(dashboard): isolate turns in compute host (#65895) 2026-07-16 15:24:03 -04:00
test_custom_provider_session_persistence.py fix(desktop): never persist or restore a named custom provider as bare "custom" (#48547) 2026-06-18 11:11:51 -07:00
test_delegation_session_lifecycle.py fix(delegation): fail-closed orphan handling + session-scoped delegation lifecycle 2026-07-08 07:06:15 -07:00
test_entry_sys_path.py fix(tui): stop a cwd package named utils/proxy/ui from crashing the gateway child (#51693) 2026-06-23 23:29:45 -07:00
test_fast_session_scope.py fix(desktop): session-scope fast mode, surface profile ownership + pinned model override 2026-07-17 14:31:10 -04:00
test_finalize_session_persist.py fix tui finalize persist drop conversation_history so disconnect saves chat 2026-07-10 07:47:10 -07:00
test_gateway_owned_session_reap.py fix(tui): derive gateway-owned sources from the Platform enum, not a hardcoded list 2026-07-07 22:15:36 -07:00
test_goal_command.py feat(moa): make /moa one-shot only; route preset switching through the model picker 2026-06-27 03:09:09 -07:00
test_image_routing_stale_model.py fix(tui): route images with the live switched model 2026-07-17 09:08:30 -07:00
test_inline_rpc_gil_starvation.py fix(desktop): prevent false runtime-not-ready under gateway load (#66174) 2026-07-20 10:51:27 -04:00
test_interim_assistant_callback.py fix(desktop, ink): don't wipe messages before final message (#65919) 2026-07-20 11:42:29 -04:00
test_iso_certify_seam.py fix(ci): make tests, workflows, and attribution reliable under load (#66373) 2026-07-17 20:55:24 +00:00
test_make_agent_provider.py fix(model): persist /model switch by default across sessions 2026-06-19 07:07:06 -07:00
test_mcp_late_refresh_thread_owner.py fix(mcp): late-refresh must see desktop/dashboard discovery thread owner (#55514) 2026-06-30 02:08:37 -07:00
test_moa_reference_emit.py feat(moa): render reference-model blocks in TUI and desktop, not just CLI (#53855) 2026-06-27 18:46:20 -07:00
test_model_switch_marker_role.py fix(tui): use role=user for model switch marker to avoid HTTP 400 on strict providers (#48338) 2026-06-28 04:34:55 -07:00
test_pet_generate_rpc.py feat(pets): polish generate flow and reduce hatch CPU pressure 2026-06-24 19:08:06 -05:00
test_project_tree.py fix(desktop): mirror Windows path identity in live overlay + WSL spelling 2026-07-12 04:33:05 -04:00
test_projects_rpc.py feat: unify active-project identity in chat status (supersedes #64721) 2026-07-18 23:08:13 -04:00
test_protocol.py fix(desktop): prevent duplicate messages when verification candidates are persisted (#68149) 2026-07-20 17:14:41 +00:00
test_reasoning_config_per_model.py feat(config): support per-model reasoning_effort overrides 2026-07-14 11:46:40 -07:00
test_reasoning_session_scope.py fix(desktop,tui-gateway,zai): stop thinking-off from reverting to medium 2026-07-02 15:23:47 -05:00
test_render.py
test_review_summary_callback.py fix(desktop,tui): surface self-improvement review summary + honor memory_notifications 2026-06-18 13:22:12 -05:00
test_session_platform_resolution.py test: deflake CI and dev-machine flaky tests in bulk (11 tests, 10 files) (#61816) 2026-07-09 20:03:11 -07:00
test_slash_worker_ansi.py fix(tui_gateway): strip ANSI from slash-worker output for desktop chat 2026-07-01 16:28:34 -05:00
test_slash_worker_mcp_discovery.py test(tui): cover profile-local MCP discovery 2026-07-10 18:09:17 +05:30
test_slash_worker_profile_home.py fix(tui): pass profile_home to slash_worker subprocess for profile-local skill discovery (#40677) 2026-07-07 05:14:00 -07:00
test_slash_worker_sys_path.py fix(tui): use shared harden_import_path guard in slash_worker 2026-07-01 05:29:22 -07:00
test_subagent_child_mirror.py feat(desktop): stream subagent activity into watch windows (#47060) 2026-06-16 14:30:11 -04:00
test_undo_command.py test: deflake CI and dev-machine flaky tests in bulk (11 tests, 10 files) (#61816) 2026-07-09 20:03:11 -07:00
test_wait_for_mcp_discovery.py fix(ci): make tests, workflows, and attribution reliable under load (#66373) 2026-07-17 20:55:24 +00:00