v1.91.5.0 feat: balanced free-suite shards, test:ubicloud, and faster PR eval lane (#2989)

* v1.91.5.0 feat: balanced free-suite shards, 16-way Linux runs, and bun run test:ubicloud

* chore: regenerate agents digest for v1.91.5.0

* ci: serial flaky retry and flake ledger for the Windows free lane

* ci: cancel superseded eval runs; relax LLM-judge clarity bar to 3
This commit is contained in:
Garry Tan authored and GitHub committed 2026-09-28 16:00:12 -07:00
1 parent d2a0bbcf4c
commit 080a655791
30 files changed
+2474 -1784

No files matched your search

+9 -2
View File
@@ -81,12 +81,19 @@ describe('paid CI coordination stays off the eval image', () => {
expect(planner.steps.find(step => step.run?.includes('--emit-plan'))?.run).toContain('bun --no-install run');
});
test(`${name}: cancel-in-progress can stop every job (no job-level always())`, () => {
for (const [id, job] of Object.entries(jobs)) {
const condition = String((job as { if?: unknown }).if ?? '');
expect(`${id}: ${condition}`).not.toMatch(/(^|[^!])always\(\)/);
}
});
test(`${name}: executors still require both prerequisites and consume the image`, () => {
const executor = jobs['eval-slices'];
expect(executor.needs).toEqual(['build-image', 'plan-slices']);
expect(JSON.stringify(executor.container)).toContain('needs.build-image.outputs.image-tag');
if (name === 'evals.yml') {
expect(executor.if).toBe("always() && needs.build-image.result == 'success' && needs.plan-slices.result == 'success'");
expect(executor.if).toBe("${{ !cancelled() && needs.build-image.result == 'success' && needs.plan-slices.result == 'success' }}");
} else {
expect(executor.if).toBeUndefined();
}
@@ -99,7 +106,7 @@ describe('paid CI coordination stays off the eval image', () => {
expect(report.container).toBeUndefined();
expect(report.needs).toContain('plan-slices');
expect(report.needs).toContain('eval-slices');
expect(report.if).toBe("always() && needs.plan-slices.result == 'success'");
expect(report.if).toBe("${{ !cancelled() && needs.plan-slices.result == 'success' }}");
expect(JSON.stringify(report.steps)).not.toMatch(/restore-deps|bun install/);
expect(report.steps.find(step => step.run?.includes('--report'))?.run).toContain('bun --no-install run');
if (name === 'evals.yml') expect(report.permissions).toEqual({ contents: 'read' });