What causes "diet run 不存在或缺少 input" in datawhalechina/hello-agents?

Replay only runs created by the current version, which persist input as a dict.

datawhalechina/hello-agents · error · ValueError

diet run 不存在或缺少 input

Error message

diet run 不存在或缺少 input

What it means

ValueError raised by replay_diet_run when the stored row is missing (run id unknown) OR row['input'] is not a dict — i.e. the persisted input payload needed to re-execute the pipeline is absent/malformed. The route layer converts it to HTTP 400.

Source

Thrown at Co-creation-projects/Shawnxyxy-HealthRecordAgent/backend/service/diet_recommend_service.py:33

    async def run(
        self,
        user_id: str,
        context: Dict[str, Any],
        *,
        replayed_from_run_id: str | None = None,
    ) -> Dict[str, Any]:
        pipeline = DietMultiAgentPipeline()
        return await pipeline.run(
            user_id, context, replayed_from_run_id=replayed_from_run_id
        )


async def replay_diet_run(original_run_id: str) -> Dict[str, Any]:
    """阶段 3：用历史 run 的 input 重跑流水线（新 run_id；溯源 replayed_from）。"""
    row = get_diet_run(original_run_id.strip())
    if not row or not isinstance(row.get("input"), dict):
        raise ValueError("diet run 不存在或缺少 input")
    svc = DietRecommendService()
    return await svc.run(
        row["user_id"],
        row["input"],
        replayed_from_run_id=original_run_id.strip(),
    )

View on GitHub (pinned to 606a07d341)

Solutions

Replay only runs created by the current version, which persist input as a dict.
If rows exist with stringified input, migrate them: parse the JSON string into a dict before replay.
Create a new run via /diet/recommend with the same context if the original input is unrecoverable.

Example fix

# before
await replay_diet_run(old_run_id)  # ValueError: input column is NULL

# after
row = get_diet_run(old_run_id)
inp = row['input'] if isinstance(row.get('input'), dict) else json.loads(row['input'])
if inp is None:
    raise ValueError('original input not persisted; create a new run instead')

Defensive patterns

Strategy: validation

Validate before calling

row = get_diet_run(run_id.strip())
if not row or not isinstance(row.get("input"), dict):
    raise ValueError("cannot replay: run missing or input not persisted")

Type guard

def is_replayable_run(row) -> bool:
    return (
        isinstance(row, dict)
        and isinstance(row.get("input"), dict)
        and bool(row.get("user_id"))
    )

Try / catch

try:
    new_run = await replay_diet_run(rid)
except ValueError as e:
    # route layer maps this to HTTP 400; distinguish missing-run vs missing-input
    log.warning("replay rejected: %s", e)
    raise

Prevention

Persist the pipeline input as a JSON dict on every run from day one.
Migrate legacy rows: parse stringified inputs into dicts before replay.
Surface both failure modes (row missing, input malformed) distinctly in callers.

When it happens

Trigger: POST /diet/runs/{run_id}/replay where the run row was deleted, or where the input column is NULL/JSON string instead of a dict (runs created before input persistence was added).

Common situations: Schema drift: old runs persisted before the 'input' column existed; storage writing input as a serialized string; deletion/cleanup jobs removing runs.

Related errors

AI-assisted analysis of datawhalechina/hello-agents@606a07d341 (2026-08-14). Data as JSON: /api/errors/6f4a0c3a294d2214. Report an issue: GitHub.