69 lines
2.2 KiB
Python

from __future__ import annotations
from typing import Any, Callable
HelperMap = dict[str, Callable[..., Any]]
def run(
run_id: str,
node_id: str,
output_dir: Any,
node_data: dict[str, Any],
inputs: list[dict[str, Any]],
helpers: HelperMap,
) -> list[dict[str, Any]]:
params = helpers["hyperparams"](node_data)
payload = helpers["first_payload_with_keys"](inputs, "packageCount", "items")
if payload is None:
raise RuntimeError("抽检采样需要上游成品数据输入")
rules = helpers["normalize_sampling_rules"](helpers["parse_json_param"](params, "RULES_JSON", [], list))
sampled_items = []
summaries = []
for rule in rules:
rule_id = rule["id"]
field_path = rule["field"]
operator = rule["operator"]
value = rule.get("value")
percent = rule["samplePercent"]
matched = []
sampled = []
for item in payload.get("items", []):
if not isinstance(item, dict):
continue
if not helpers["compare_values"](helpers["get_field_value"](item, field_path), operator, value):
continue
matched.append(item)
package_name = str(item.get("packageName", "")).strip() or str(item.get("clipId", "")).strip()
if helpers["stable_sample_hit"](f"{rule_id}:{package_name}", percent):
sampled.append(item)
sampled_items.append({**item, "sampleRuleId": rule_id})
summaries.append(
{
"id": rule_id,
"field": field_path,
"operator": operator,
"value": value,
"samplePercent": percent,
"matchedCount": len(matched),
"sampledCount": len(sampled),
}
)
manifest = {
"datasetType": "x-platform-audit-samples",
"rules": summaries,
"sampledCount": len(sampled_items),
"items": sampled_items,
}
return [
helpers["write_json_artifact"](
run_id,
node_id,
output_dir,
"audit-samples.json",
manifest,
output=helpers["output_binding"](node_data),
)
]