[Video]: add H3 artifact demo / 新增 H3 产物演示页

packages/app/cypress/e2e/navigation.cy.ts

@@ -321,3 +321,32 @@ describe('TPUv7 launch banner', { testIsolation: true }, () => {
     cy.get('[data-testid="tco-basis-toggle"]').should('not.exist');
   });
 });
+
+describe('H3 video artifact viewer', () => {
+  it('uses the shared unlock for navigation and keeps an empty viewer free of sample results', () => {
+    cy.viewport(1440, 1000);
+    cy.visit('/video', {
+      onBeforeLoad(win) {
+        win.localStorage.removeItem('inferencex-feature-gate');
+      },
+    });
+    cy.get('[data-testid="video-benchmark"]').should('contain', 'Open a real CI result to begin');
+    cy.get('video[data-role]').should('not.exist');
+    cy.get('[data-testid="tab-trigger-hidden"]').should('not.exist');
+    cy.get('body').type('{upArrow}{upArrow}{downArrow}{downArrow}');
+    cy.get('[data-testid="tab-trigger-hidden"]').click();
+    cy.contains('a', 'Video').should('have.attr', 'href', '/video');
+    cy.get('head meta[name="robots"]').should('have.attr', 'content', 'noindex, nofollow');
+  });
+  it('shows a recoverable load error and the Chinese empty state', () => {
+    cy.visit('/video');
+    cy.get('[data-testid="video-benchmark"]').contains('summary', 'Manifest URL').click();
+    cy.get('input[aria-label="Manifest URL"]').type('https://example.com/wrong.json');
+    cy.contains('button', 'Load manifest').click();
+    cy.get('[role="alert"]').should('contain', 'Could not load this bundle');
+    cy.get('video[data-role]').should('not.exist');
+    cy.visit('/zh/video');
+    cy.get('[data-testid="video-benchmark"]').should('contain', '打开真实 CI 结果开始查看');
+    cy.get('head meta[name="robots"]').should('have.attr', 'content', 'noindex, nofollow');
+  });
+});

packages/app/src/app/(dashboard)/video/page.tsx

@@ -0,0 +1,11 @@
+import type { Metadata } from 'next';
+import VideoBenchmark from '@/components/video-benchmark/VideoBenchmark';
+import { tabMetadata } from '@/lib/tab-meta';
+
+export const metadata: Metadata = {
+  ...tabMetadata('video'),
+  robots: { index: false, follow: false },
+};
+export default function VideoPage() {
+  return <VideoBenchmark />;
+}

packages/app/src/app/zh/(dashboard)/video/page.tsx

@@ -0,0 +1,11 @@
+import type { Metadata } from 'next';
+import VideoBenchmark from '@/components/video-benchmark/VideoBenchmark';
+import { tabMetadataZh } from '@/lib/tab-meta-zh';
+
+export const metadata: Metadata = {
+  ...tabMetadataZh('video'),
+  robots: { index: false, follow: false },
+};
+export default function VideoPage() {
+  return <VideoBenchmark />;
+}

packages/app/src/components/tab-nav.tsx

@@ -56,6 +56,7 @@ const TAB_LABELS_EN: Record<DashboardRouteKey, string> = {
   'ai-chart': 'AI Chart',
   'gpu-metrics': 'PowerX',
   'current-inferencex-image': 'Images',
+  video: 'Video',
   feedback: 'Feedback',
 };
 

packages/app/src/components/video-benchmark/bundle.test.ts

@@ -0,0 +1,170 @@
+import { readFile } from 'node:fs/promises';
+import { join } from 'node:path';
+import { describe, expect, it } from 'vitest';
+import {
+  at,
+  estimateEconomics,
+  httpReader,
+  loadBundle,
+  sampledPower,
+  safePath,
+  sha256,
+  type Json,
+} from './bundle';
+
+// Synthetic contract tests only. These values are never presented as benchmark evidence.
+async function fixture(extra: Record<string, string> = {}) {
+  const files = new Map(
+    Object.entries({
+      'manifest.json': JSON.stringify({
+        schema_version: 1,
+        run_id: '123',
+        run_attempt: '1',
+        ci: { repository: 'SemiAnalysisAI/InferenceX' },
+        git_commit: 'a'.repeat(40),
+        evidence: {},
+      }),
+      'ci.json': JSON.stringify({ phase: 'failed', exit_code: 2 }),
+      ...extra,
+    }).map(([name, value]) => [name, new Blob([value])]),
+  );
+  const sums = [];
+  for (const [name, blob] of files) sums.push(`${await sha256(blob)}  ${name}`);
+  files.set('SHA256SUMS', new Blob([`${sums.join('\n')}\n`]));
+  return {
+    files,
+    read: async (name: string) => {
+      const file = files.get(name);
+      if (!file) throw new Error(`Missing: ${name}`);
+      return await Promise.resolve(file);
+    },
+  };
+}
+describe('H3 artifact contract', () => {
+  it('loads a partial failed attempt without inventing report/media or measurements', async () => {
+    const f = await fixture();
+    const b = await loadBundle(f.read);
+    expect(at(b.ci, 'phase')).toBe('failed');
+    expect(b.report).toBeNull();
+    expect(b.files.size).toBe(3);
+  });
+  it('rejects modified bytes and missing files', async () => {
+    const f = await fixture();
+    f.files.set('ci.json', new Blob(['tampered']));
+    await expect(loadBundle(f.read)).rejects.toThrow('SHA256 mismatch: ci.json');
+    f.files.delete('ci.json');
+    await expect(loadBundle(f.read)).rejects.toThrow('Missing: ci.json');
+  });
+  it('rejects inconsistent manifest evidence even when the inventory is self-consistent', async () => {
+    const f = await fixture({
+      'manifest.json': JSON.stringify({
+        schema_version: 1,
+        run_id: '123',
+        run_attempt: '1',
+        ci: { repository: 'SemiAnalysisAI/InferenceX' },
+        git_commit: 'a'.repeat(40),
+        evidence: { 'ci.json': 'b'.repeat(64) },
+      }),
+    });
+    await expect(loadBundle(f.read)).rejects.toThrow('Manifest evidence mismatch');
+  });
+  it('rejects fixture reports and unknown manifest versions', async () => {
+    const synthetic = await fixture({ 'report/evidence.json': '{"bundle_type":"synthetic"}' });
+    await expect(loadBundle(synthetic.read)).rejects.toThrow('Unsupported report');
+    const unknown = await fixture({ 'manifest.json': '{"schema_version":2}' });
+    await expect(loadBundle(unknown.read)).rejects.toThrow('Unsupported H3');
+  });
+  it.each([
+    '../secret',
+    '/absolute',
+    'a/../b',
+    'a/%2e%2e/b',
+    'https://other/x',
+    String.raw`a\b`,
+    'a?x=1',
+    'a\n',
+  ])('rejects unsafe paths: %s', (path) => expect(() => safePath(path)).toThrow());
+  it('permits only explicit HTTPS or loopback artifact directories without embedded credentials', () => {
+    expect(() => httpReader('https://example.com/run/manifest.json')).not.toThrow();
+    expect(() => httpReader('http://127.0.0.1:8769/manifest.json')).not.toThrow();
+    for (const url of [
+      'http://example.com/manifest.json',
+      'https://user:secret@example.com/manifest.json',
+      'https://example.com/manifest.json?token=secret',
+      'file:///manifest.json',
+    ])
+      expect(() => httpReader(url)).toThrow();
+  });
+});
+function sample(t: number, watts: Json, phase = 'measurement'): Json {
+  return {
+    monotonic_seconds: t,
+    phase,
+    at: `time-${t}`,
+    gpus: [{ uuid: 'GPU-1', power_watts: watts }],
+    owned_compute_apps: [{ gpu_uuid: 'GPU-1' }],
+    unowned_compute_apps: [],
+  };
+}
+describe('sampled board power', () => {
+  it('integrates unequal intervals, excludes startup and does not extrapolate endpoints', () => {
+    const value = sampledPower(
+      [sample(0, 999, 'startup'), sample(1, 100), sample(2, 200), sample(4, 100)],
+      ['GPU-1'],
+      3,
+    );
+    expect(value).toMatchObject({
+      watts: 150,
+      joules: 450,
+      seconds: 3,
+      sampleCount: 3,
+      start: 'time-1',
+      end: 'time-4',
+    });
+  });
+  it('keeps missing power, missing GPUs, insufficient samples, gaps and duplicate timestamps unavailable', () => {
+    for (const samples of [
+      [sample(1, 1)],
+      [sample(1, 100), sample(2, null)],
+      [sample(1, 100), sample(5, 100)],
+      [sample(1, 100), sample(1, 100)],
+    ])
+      expect(sampledPower(samples, ['GPU-1'], 3)).toBeNull();
+    expect(sampledPower([sample(1, 100), sample(2, 100)], ['GPU-2'], 3)).toBeNull();
+    expect(sampledPower([sample(1, 0), sample(2, 0)], ['GPU-1'], 3)?.joules).toBe(0);
+  });
+});
+describe('explicit economics assumptions', () => {
+  it('charges billed GPUs while displaying both denominators and preserves a negative profit', () => {
+    const value = estimateEconomics(1 / 60, 4, 8, 0.2, 3);
+    expect(value).toEqual({
+      revenue: 12,
+      revenuePerParticipating: 3,
+      revenuePerBilled: 1.5,
+      profitPerParticipating: -3,
+      profitPerBilled: -1.5,
+    });
+    expect(estimateEconomics(1 / 60, 4, 8, 0.2, null)?.profitPerBilled).toBeNull();
+  });
+  it('rejects absent/invalid assumptions without converting missing data to zero', () => {
+    expect(estimateEconomics(null, 4, 8, 1, 1)).toBeNull();
+    expect(estimateEconomics(1, 4, 2, 1, 1)).toBeNull();
+    expect(estimateEconomics(1, 4, 8, Number.NaN, 1)).toBeNull();
+  });
+});
+
+// Opt-in verification of original downloaded backend bytes; no model/GPU execution.
+it.skipIf(!process.env.H3_ARTIFACT_DIR)(
+  'opens the real CI artifact with both roles and native audio',
+  async () => {
+    const b = await loadBundle(
+      async (path) => new Blob([await readFile(join(process.env.H3_ARTIFACT_DIR!, path))]),
+    );
+    expect(at(b.ci, 'phase')).toBe('complete');
+    expect(at(b.ci, 'regression_status')).toBe('inconclusive');
+    for (const role of ['baseline', 'candidate']) {
+      expect(at(b.report, 'roles', role, 'observations', 0, 'media', 'audio', 'channels')).toBe(2);
+      expect(at(b.report, 'roles', role, 'summary', 'valid')).toBe(1);
+    }
+  },
+);