fix: self-contained tree-inject test + fail-loud on missing question (review)

app/harness/video_split_cli.py

@@ -610,7 +610,15 @@ def _execute_real(config: VideoSplitConfig, fingerprint: str, args: argparse.Nam
     canonical_preds = load_canonical_predictions(harness_db, config.baseline_run_id)
     wrong_ids = select_diagnosable_wrong_ids(canonical_preds)
     questions = load_questions_by_id(questions_dir)
-    video_ids = [questions[qid].video_id for qid in wrong_ids]
+    video_ids: list[str] = []
+    for qid in wrong_ids:
+        q = questions.get(qid)
+        if q is None:
+            raise SystemExit(
+                f"wrong_id {qid!r} 不在 questions_dir 题库中"
+                "（baseline predictions 与题库不匹配，P5 fail loud）"
+            )
+        video_ids.append(q.video_id)
     deps = build_diagnosis_deps(
         harness_db=harness_db,
         store_dir=store_dir,

tests/unit/test_video_split_cli_tree_inject.py

@@ -1,16 +1,35 @@
-"""离线诊断注入：build_diagnosis_deps 按 video_ids 填充非空 tree_data。"""
+"""离线诊断注入：build_diagnosis_deps 按 video_ids 填充非空 tree_data。
 
-from pathlib import Path
+自包含：在 tmp_path 造最小 tree.json，不依赖真实 store/videos/*（CI/checkout 稳健）。
+下面 patch 的三项是 build_diagnosis_deps 的**外部依赖边界**（假 LLM 凭证、LLM HTTP
+客户端、redis 缓存），非实现细节——避免测试触真 .env / 打开 httpx.AsyncClient /
+连 redis，聚焦"给定 store_dir/video_ids → tree_data 被真实树填充"。
+"""
+
+import json
 from unittest.mock import patch
 
 from app.harness.video_split_cli import build_diagnosis_deps
 
 
-def test_build_diagnosis_deps_loads_tree_for_videos():
+def test_build_diagnosis_deps_loads_tree_for_videos(tmp_path):
+    vdir = tmp_path / "videos" / "vX"
+    vdir.mkdir(parents=True)
+    (vdir / "tree.json").write_text(
+        json.dumps(
+            {
+                "metadata": {},
+                "roots": [
+                    {"id": "vX_L1_000", "card": {"scene_summary": "s"}, "time_range": [0, 1]}
+                ],
+            }
+        ),
+        encoding="utf-8",
+    )
+
     with (
         patch("app.harness.video_split_cli._DiagLLMSettings") as settings_cls,
         patch("adapters.llm.GovernedLLMClient"),
-        patch("adapters.telemetry.SQLiteTelemetryRecorder"),
         patch("app.harness.video_split_cli._build_redis_cache", return_value=None),
     ):
         s = settings_cls.return_value
@@ -23,11 +42,12 @@ def test_build_diagnosis_deps_loads_tree_for_videos():
         s.llm_max_retries = 1
         s.llm_retry_base_delay = s.llm_retry_max_delay = 1
         deps = build_diagnosis_deps(
-            harness_db=Path("workspaces/default/harness.db"),
+            harness_db=tmp_path / "h.db",
-            store_dir=Path("store"),
+            store_dir=tmp_path,
-            video_ids=["0RxMZBLeqRI"],
+            video_ids=["vX"],
             concurrency=1,
             expected_model="deepseek-v4-pro",
         )
-    assert "0RxMZBLeqRI" in deps.tree_data
+
-    assert deps.tree_data["0RxMZBLeqRI"]["nodes"]
+    assert "vX" in deps.tree_data
+    assert deps.tree_data["vX"]["nodes"]