Some checks failed
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-school (push) Successful in 42s
CI / test-go-edu-search (push) Successful in 34s
CI / test-python-klausur (push) Failing after 2m51s
CI / test-python-agent-core (push) Successful in 21s
CI / test-nodejs-website (push) Successful in 29s
sed replacement left orphaned hostname references in story page and empty lines in getApiBase functions. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
78 lines
3.2 KiB
TypeScript
78 lines
3.2 KiB
TypeScript
'use client'
|
|
|
|
import type { BQASMetrics } from '../types'
|
|
import { IntentScoresChart } from './IntentScoresChart'
|
|
import { FailedTestsList } from './FailedTestsList'
|
|
|
|
export function GoldenTab({
|
|
goldenMetrics,
|
|
isRunningGolden,
|
|
runGoldenTests,
|
|
}: {
|
|
goldenMetrics: BQASMetrics | null
|
|
isRunningGolden: boolean
|
|
runGoldenTests: () => void
|
|
}) {
|
|
return (
|
|
<div className="space-y-6">
|
|
<div className="bg-white rounded-xl border border-slate-200 p-6">
|
|
<div className="flex items-center justify-between mb-6">
|
|
<div>
|
|
<h3 className="text-lg font-semibold text-slate-900">Golden Test Suite</h3>
|
|
<p className="text-sm text-slate-500">Validierte Referenz-Tests gegen definierte Erwartungen</p>
|
|
</div>
|
|
<button
|
|
onClick={runGoldenTests}
|
|
disabled={isRunningGolden}
|
|
className={`px-4 py-2 rounded-lg text-sm font-medium transition-all ${
|
|
isRunningGolden
|
|
? 'bg-teal-100 text-teal-600 cursor-wait'
|
|
: 'bg-teal-600 text-white hover:bg-teal-700 active:scale-95'
|
|
}`}
|
|
>
|
|
{isRunningGolden ? 'Laeuft...' : 'Tests starten'}
|
|
</button>
|
|
</div>
|
|
|
|
{goldenMetrics && (
|
|
<>
|
|
<div className="grid grid-cols-2 md:grid-cols-5 gap-4 mb-6">
|
|
<div className="text-center p-4 bg-slate-50 rounded-lg">
|
|
<p className="text-2xl font-bold text-slate-900">{goldenMetrics.total_tests}</p>
|
|
<p className="text-xs text-slate-500">Tests</p>
|
|
</div>
|
|
<div className="text-center p-4 bg-emerald-50 rounded-lg">
|
|
<p className="text-2xl font-bold text-emerald-600">{goldenMetrics.passed_tests}</p>
|
|
<p className="text-xs text-slate-500">Bestanden</p>
|
|
</div>
|
|
<div className="text-center p-4 bg-red-50 rounded-lg">
|
|
<p className="text-2xl font-bold text-red-600">{goldenMetrics.failed_tests}</p>
|
|
<p className="text-xs text-slate-500">Fehlgeschlagen</p>
|
|
</div>
|
|
<div className="text-center p-4 bg-blue-50 rounded-lg">
|
|
<p className="text-2xl font-bold text-blue-600">{goldenMetrics.avg_intent_accuracy.toFixed(0)}%</p>
|
|
<p className="text-xs text-slate-500">Intent Accuracy</p>
|
|
</div>
|
|
<div className="text-center p-4 bg-purple-50 rounded-lg">
|
|
<p className="text-2xl font-bold text-purple-600">{goldenMetrics.avg_composite_score.toFixed(2)}</p>
|
|
<p className="text-xs text-slate-500">Composite Score</p>
|
|
</div>
|
|
</div>
|
|
|
|
<div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
|
|
<div>
|
|
<h4 className="font-medium text-slate-900 mb-4">Scores nach Intent</h4>
|
|
<IntentScoresChart scores={goldenMetrics.scores_by_intent} />
|
|
</div>
|
|
<div>
|
|
<h4 className="font-medium text-slate-900 mb-4">Fehlgeschlagene Tests ({goldenMetrics.failed_tests})</h4>
|
|
<FailedTestsList testIds={goldenMetrics.failed_test_ids} />
|
|
</div>
|
|
</div>
|
|
</>
|
|
)}
|
|
</div>
|
|
</div>
|
|
)
|
|
}
|