@velum-labs/routekit-eval-setup 1.3.2 → 1.5.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/adapters/authoring-responses-request.d.ts +5 -0
- package/dist/adapters/authoring-responses-request.js +38 -0
- package/dist/adapters/evaluation-evidence-freshness.d.ts +7 -0
- package/dist/adapters/evaluation-evidence-freshness.js +76 -0
- package/dist/adapters/git-task-history.d.ts +67 -0
- package/dist/adapters/git-task-history.js +171 -0
- package/dist/adapters/integrated-repository-history.d.ts +21 -0
- package/dist/adapters/integrated-repository-history.js +175 -0
- package/dist/adapters/repository-command-diagnostic.d.ts +8 -0
- package/dist/adapters/repository-command-diagnostic.js +46 -0
- package/dist/adapters/repository-command-evidence.d.ts +13 -0
- package/dist/adapters/repository-command-evidence.js +102 -0
- package/dist/adapters/repository-command-runner.d.ts +238 -0
- package/dist/adapters/repository-command-runner.js +1483 -0
- package/dist/adapters/repository-import-context.d.ts +47 -0
- package/dist/adapters/repository-import-context.js +469 -0
- package/dist/adapters/repository-node-test-reporter.d.ts +3 -0
- package/dist/adapters/repository-node-test-reporter.js +27 -0
- package/dist/adapters/repository-review-evidence.d.ts +39 -0
- package/dist/adapters/repository-review-evidence.js +632 -0
- package/dist/adapters/repository-seed-selection.d.ts +7 -0
- package/dist/adapters/repository-seed-selection.js +79 -0
- package/dist/adapters/repository-solution-edits.d.ts +49 -0
- package/dist/adapters/repository-solution-edits.js +136 -0
- package/dist/adapters/repository-vitest-phase-adapter.d.ts +8 -0
- package/dist/adapters/repository-vitest-phase-adapter.js +310 -0
- package/dist/adapters/repository-vitest-reporter.d.ts +24 -0
- package/dist/adapters/repository-vitest-reporter.js +314 -0
- package/dist/adapters/strict-authoring-schema.d.ts +5 -0
- package/dist/adapters/strict-authoring-schema.js +158 -0
- package/dist/adapters/test-discovery.d.ts +30 -0
- package/dist/adapters/test-discovery.js +124 -0
- package/dist/adapters/typescript-repository-index.d.ts +51 -0
- package/dist/adapters/typescript-repository-index.js +226 -0
- package/dist/agentic-capabilities-protocol.d.ts +1373 -0
- package/dist/agentic-capabilities-protocol.js +786 -0
- package/dist/case-checkpoint-store.d.ts +29 -0
- package/dist/case-checkpoint-store.js +133 -0
- package/dist/case-pipeline-protocol-v2.d.ts +184 -0
- package/dist/case-pipeline-protocol-v2.js +193 -0
- package/dist/case-pipeline-protocol.d.ts +2626 -0
- package/dist/case-pipeline-protocol.js +371 -0
- package/dist/effect-api.d.ts +74 -10
- package/dist/effect-api.js +56 -6
- package/dist/errors.d.ts +31 -0
- package/dist/errors.js +10 -0
- package/dist/eval-capability-execution-envelope.d.ts +64 -0
- package/dist/eval-capability-execution-envelope.js +98 -0
- package/dist/eval-capability-policy.d.ts +90 -0
- package/dist/eval-capability-policy.js +107 -0
- package/dist/eval-event-log.d.ts +140 -0
- package/dist/eval-event-log.js +220 -0
- package/dist/evaluation-authoring-policy.d.ts +18 -0
- package/dist/evaluation-authoring-policy.js +19 -0
- package/dist/evaluation-authoring-validation.d.ts +22 -0
- package/dist/evaluation-authoring-validation.js +72 -0
- package/dist/evaluation-evidence.d.ts +20 -0
- package/dist/evaluation-evidence.js +319 -0
- package/dist/evaluation-grader-calibration-protocol.d.ts +108 -0
- package/dist/evaluation-grader-calibration-protocol.js +80 -0
- package/dist/evaluation-grader-calibration.d.ts +18 -0
- package/dist/evaluation-grader-calibration.js +334 -0
- package/dist/evaluation-grading-policy.d.ts +24 -0
- package/dist/evaluation-grading-policy.js +54 -0
- package/dist/evaluation-proposal-policy.d.ts +4 -0
- package/dist/evaluation-proposal-policy.js +91 -0
- package/dist/evaluation-source-retrieval.d.ts +68 -0
- package/dist/evaluation-source-retrieval.js +513 -0
- package/dist/evaluation-structure-policy.d.ts +29 -0
- package/dist/evaluation-structure-policy.js +138 -0
- package/dist/index.d.ts +124 -17
- package/dist/index.js +69 -11
- package/dist/inspection.js +2 -3
- package/dist/project-artifacts.d.ts +7 -2
- package/dist/project-artifacts.js +49 -136
- package/dist/project-authoring.d.ts +66 -5
- package/dist/project-authoring.js +783 -109
- package/dist/project-contracts.d.ts +419 -84
- package/dist/project-contracts.js +160 -52
- package/dist/project-store.js +2 -1
- package/dist/project-workflow.d.ts +5 -4
- package/dist/project-workflow.js +154 -35
- package/dist/repository-adversary-protocol.d.ts +64 -0
- package/dist/repository-adversary-protocol.js +105 -0
- package/dist/repository-behavior-protocol.d.ts +188 -0
- package/dist/repository-behavior-protocol.js +202 -0
- package/dist/repository-benchmark-protocol.d.ts +487 -0
- package/dist/repository-benchmark-protocol.js +96 -0
- package/dist/repository-execution-protocol.d.ts +150 -0
- package/dist/repository-execution-protocol.js +38 -0
- package/dist/repository-fixture-instructions.d.ts +3 -0
- package/dist/repository-fixture-instructions.js +91 -0
- package/dist/repository-fixture-protocol.d.ts +79 -0
- package/dist/repository-fixture-protocol.js +79 -0
- package/dist/repository-foundry-plan-protocol.d.ts +118 -0
- package/dist/repository-foundry-plan-protocol.js +296 -0
- package/dist/repository-foundry-progress-protocol.d.ts +52 -0
- package/dist/repository-foundry-progress-protocol.js +52 -0
- package/dist/repository-improvement-protocol.d.ts +100 -0
- package/dist/repository-improvement-protocol.js +106 -0
- package/dist/repository-language-model-protocol.d.ts +43 -0
- package/dist/repository-language-model-protocol.js +146 -0
- package/dist/repository-oracle-coverage-protocol.d.ts +18 -0
- package/dist/repository-oracle-coverage-protocol.js +39 -0
- package/dist/repository-oracle-execution-binding.d.ts +27 -0
- package/dist/repository-oracle-execution-binding.js +59 -0
- package/dist/repository-oracle-protocol.d.ts +230 -0
- package/dist/repository-oracle-protocol.js +156 -0
- package/dist/repository-oracle-scope-policy.d.ts +22 -0
- package/dist/repository-oracle-scope-policy.js +92 -0
- package/dist/repository-quality-policy.d.ts +15 -0
- package/dist/repository-quality-policy.js +357 -0
- package/dist/repository-routing-benchmark-protocol.d.ts +176 -0
- package/dist/repository-routing-benchmark-protocol.js +103 -0
- package/dist/repository-routing-model-protocol.d.ts +36 -0
- package/dist/repository-routing-model-protocol.js +89 -0
- package/dist/repository-routing-plan-protocol.d.ts +112 -0
- package/dist/repository-routing-plan-protocol.js +58 -0
- package/dist/repository-routing-quality-policy.d.ts +9 -0
- package/dist/repository-routing-quality-policy.js +191 -0
- package/dist/repository-seed-qualification-progress-protocol.d.ts +205 -0
- package/dist/repository-seed-qualification-progress-protocol.js +28 -0
- package/dist/repository-semantic-calibration-protocol.d.ts +768 -0
- package/dist/repository-semantic-calibration-protocol.js +276 -0
- package/dist/repository-semantic-calibration.d.ts +163 -0
- package/dist/repository-semantic-calibration.js +581 -0
- package/dist/repository-specification-contract-facts-protocol.d.ts +224 -0
- package/dist/repository-specification-contract-facts-protocol.js +276 -0
- package/dist/repository-specification-critique-protocol.d.ts +189 -0
- package/dist/repository-specification-critique-protocol.js +103 -0
- package/dist/repository-task-family-protocol.d.ts +24 -0
- package/dist/repository-task-family-protocol.js +37 -0
- package/dist/repository-task-seed-protocol.d.ts +384 -0
- package/dist/repository-task-seed-protocol.js +236 -0
- package/dist/repository-trajectory-protocol.d.ts +20 -0
- package/dist/repository-trajectory-protocol.js +42 -0
- package/dist/service.js +1 -1
- package/dist/services/adversary/service.d.ts +64 -0
- package/dist/services/adversary/service.js +330 -0
- package/dist/services/benchmark-compiler/service.d.ts +450 -0
- package/dist/services/benchmark-compiler/service.js +9 -0
- package/dist/services/budgeted-model/service.d.ts +118 -0
- package/dist/services/budgeted-model/service.js +460 -0
- package/dist/services/case-authoring/service.d.ts +163 -0
- package/dist/services/case-authoring/service.js +1456 -0
- package/dist/services/case-finalization/service.d.ts +283 -0
- package/dist/services/case-finalization/service.js +370 -0
- package/dist/services/case-generation/service.d.ts +619 -0
- package/dist/services/case-generation/service.js +2628 -0
- package/dist/services/case-pipeline/service.d.ts +31 -0
- package/dist/services/case-pipeline/service.js +485 -0
- package/dist/services/case-pipeline-v2/service.d.ts +70 -0
- package/dist/services/case-pipeline-v2/service.js +477 -0
- package/dist/services/command-observability/service.d.ts +13 -0
- package/dist/services/command-observability/service.js +3 -0
- package/dist/services/dimension-labeling/service.d.ts +77 -0
- package/dist/services/dimension-labeling/service.js +188 -0
- package/dist/services/eval-candidate/service.d.ts +208 -0
- package/dist/services/eval-candidate/service.js +64 -0
- package/dist/services/eval-capabilities/service.d.ts +183 -0
- package/dist/services/eval-capabilities/service.js +1433 -0
- package/dist/services/eval-environment/service.d.ts +173 -0
- package/dist/services/eval-environment/service.js +127 -0
- package/dist/services/evidence-reconstruction/service.d.ts +36 -0
- package/dist/services/evidence-reconstruction/service.js +145 -0
- package/dist/services/fixture-builder/service.d.ts +62 -0
- package/dist/services/fixture-builder/service.js +36 -0
- package/dist/services/fixture-validation/service.d.ts +75 -0
- package/dist/services/fixture-validation/service.js +295 -0
- package/dist/services/foundry/service.d.ts +831 -0
- package/dist/services/foundry/service.js +442 -0
- package/dist/services/foundry-progress/service.d.ts +62 -0
- package/dist/services/foundry-progress/service.js +149 -0
- package/dist/services/foundry-v2/service.d.ts +54 -0
- package/dist/services/foundry-v2/service.js +28 -0
- package/dist/services/grounded-authoring/service.d.ts +126 -0
- package/dist/services/grounded-authoring/service.js +822 -0
- package/dist/services/historical-case/service.d.ts +722 -0
- package/dist/services/historical-case/service.js +177 -0
- package/dist/services/improvement-loop/service.d.ts +59 -0
- package/dist/services/improvement-loop/service.js +176 -0
- package/dist/services/language-model/service.d.ts +52 -0
- package/dist/services/language-model/service.js +194 -0
- package/dist/services/oracle-builder/service.d.ts +146 -0
- package/dist/services/oracle-builder/service.js +513 -0
- package/dist/services/oracle-coverage/service.d.ts +28 -0
- package/dist/services/oracle-coverage/service.js +50 -0
- package/dist/services/oracle-coverage-witness/service.d.ts +130 -0
- package/dist/services/oracle-coverage-witness/service.js +538 -0
- package/dist/services/pipeline-challenge/service.d.ts +551 -0
- package/dist/services/pipeline-challenge/service.js +427 -0
- package/dist/services/pipeline-controls/service.d.ts +130 -0
- package/dist/services/pipeline-controls/service.js +483 -0
- package/dist/services/pipeline-oracle/service.d.ts +8 -0
- package/dist/services/pipeline-oracle/service.js +256 -0
- package/dist/services/pipeline-seed/service.d.ts +298 -0
- package/dist/services/pipeline-seed/service.js +428 -0
- package/dist/services/pipeline-spec/service.d.ts +103 -0
- package/dist/services/pipeline-spec/service.js +619 -0
- package/dist/services/pipeline-tournament/service.d.ts +258 -0
- package/dist/services/pipeline-tournament/service.js +476 -0
- package/dist/services/quality-gate/service.d.ts +233 -0
- package/dist/services/quality-gate/service.js +136 -0
- package/dist/services/repository-bundle/service.d.ts +33 -0
- package/dist/services/repository-bundle/service.js +114 -0
- package/dist/services/repository-model/service.d.ts +105 -0
- package/dist/services/repository-model/service.js +250 -0
- package/dist/services/repository-public-artifact/service.d.ts +133 -0
- package/dist/services/repository-public-artifact/service.js +330 -0
- package/dist/services/routing-benchmark/service.d.ts +362 -0
- package/dist/services/routing-benchmark/service.js +96 -0
- package/dist/services/specification-critic/service.d.ts +92 -0
- package/dist/services/specification-critic/service.js +172 -0
- package/dist/services/task-family/service.d.ts +40 -0
- package/dist/services/task-family/service.js +55 -0
- package/dist/services/task-seed/service.d.ts +906 -0
- package/dist/services/task-seed/service.js +1406 -0
- package/dist/services/task-specification/service.d.ts +27 -0
- package/dist/services/task-specification/service.js +40 -0
- package/dist/services/trajectory-policy/service.d.ts +110 -0
- package/dist/services/trajectory-policy/service.js +216 -0
- package/dist/test/agentic-capabilities-protocol.test.d.ts +1 -0
- package/dist/test/agentic-capabilities-protocol.test.js +570 -0
- package/dist/test/agentic-capabilities.test.d.ts +1 -0
- package/dist/test/agentic-capabilities.test.js +1461 -0
- package/dist/test/agentic-environment.test.d.ts +1 -0
- package/dist/test/agentic-environment.test.js +213 -0
- package/dist/test/case-pipeline-foundation.test.d.ts +1 -0
- package/dist/test/case-pipeline-foundation.test.js +535 -0
- package/dist/test/case-pipeline-protocol-v2.test.d.ts +1 -0
- package/dist/test/case-pipeline-protocol-v2.test.js +124 -0
- package/dist/test/case-pipeline-v2.test.d.ts +1 -0
- package/dist/test/case-pipeline-v2.test.js +286 -0
- package/dist/test/case-pipeline.test.d.ts +1 -0
- package/dist/test/case-pipeline.test.js +851 -0
- package/dist/test/eval-capability-policy.test.d.ts +1 -0
- package/dist/test/eval-capability-policy.test.js +50 -0
- package/dist/test/eval-event-log.test.d.ts +1 -0
- package/dist/test/eval-event-log.test.js +125 -0
- package/dist/test/evaluation-evidence-freshness.test.d.ts +1 -0
- package/dist/test/evaluation-evidence-freshness.test.js +44 -0
- package/dist/test/evaluation-evidence.test.d.ts +1 -0
- package/dist/test/evaluation-evidence.test.js +230 -0
- package/dist/test/evaluation-grader-calibration.test.d.ts +1 -0
- package/dist/test/evaluation-grader-calibration.test.js +373 -0
- package/dist/test/evaluation-proposal-digest.test.d.ts +1 -0
- package/dist/test/evaluation-proposal-digest.test.js +187 -0
- package/dist/test/evaluation-source-retrieval.test.d.ts +1 -0
- package/dist/test/evaluation-source-retrieval.test.js +237 -0
- package/dist/test/evaluation-structure-policy.test.d.ts +1 -0
- package/dist/test/evaluation-structure-policy.test.js +196 -0
- package/dist/test/fixtures/repository-resource-panel.d.ts +39 -0
- package/dist/test/fixtures/repository-resource-panel.js +111 -0
- package/dist/test/fixtures/vitest-boundary-panel.d.ts +84 -0
- package/dist/test/fixtures/vitest-boundary-panel.js +120 -0
- package/dist/test/fixtures/vitest-phase-panel.d.ts +135 -0
- package/dist/test/fixtures/vitest-phase-panel.js +213 -0
- package/dist/test/fixtures/vitest-reporter-results.d.ts +76 -0
- package/dist/test/fixtures/vitest-reporter-results.js +94 -0
- package/dist/test/grounded-authoring.test.d.ts +1 -0
- package/dist/test/grounded-authoring.test.js +565 -0
- package/dist/test/integrated-repository-history.test.d.ts +1 -0
- package/dist/test/integrated-repository-history.test.js +227 -0
- package/dist/test/project-authoring.test.js +593 -43
- package/dist/test/project-workflow.test.js +419 -40
- package/dist/test/repository-authoring-artifacts.test.d.ts +1 -0
- package/dist/test/repository-authoring-artifacts.test.js +185 -0
- package/dist/test/repository-bundle.test.d.ts +1 -0
- package/dist/test/repository-bundle.test.js +52 -0
- package/dist/test/repository-case-generation.test.d.ts +1 -0
- package/dist/test/repository-case-generation.test.js +3465 -0
- package/dist/test/repository-command-diagnostic.test.d.ts +1 -0
- package/dist/test/repository-command-diagnostic.test.js +55 -0
- package/dist/test/repository-command-signals.test.d.ts +1 -0
- package/dist/test/repository-command-signals.test.js +124 -0
- package/dist/test/repository-fixture-scope-coverage.test.d.ts +1 -0
- package/dist/test/repository-fixture-scope-coverage.test.js +127 -0
- package/dist/test/repository-fixture-validation.test.d.ts +1 -0
- package/dist/test/repository-fixture-validation.test.js +362 -0
- package/dist/test/repository-foundry-progress.test.d.ts +1 -0
- package/dist/test/repository-foundry-progress.test.js +110 -0
- package/dist/test/repository-foundry-quality.test.d.ts +1 -0
- package/dist/test/repository-foundry-quality.test.js +1138 -0
- package/dist/test/repository-import-context.test.d.ts +1 -0
- package/dist/test/repository-import-context.test.js +354 -0
- package/dist/test/repository-model-authoring.test.d.ts +1 -0
- package/dist/test/repository-model-authoring.test.js +544 -0
- package/dist/test/repository-model.test.d.ts +1 -0
- package/dist/test/repository-model.test.js +2195 -0
- package/dist/test/repository-node-test-reporter.test.d.ts +1 -0
- package/dist/test/repository-node-test-reporter.test.js +104 -0
- package/dist/test/repository-oracle-concurrency.test.d.ts +1 -0
- package/dist/test/repository-oracle-concurrency.test.js +542 -0
- package/dist/test/repository-oracle-coverage-witness.test.d.ts +1 -0
- package/dist/test/repository-oracle-coverage-witness.test.js +511 -0
- package/dist/test/repository-oracle-coverage.test.d.ts +1 -0
- package/dist/test/repository-oracle-coverage.test.js +168 -0
- package/dist/test/repository-oracle-evidence.test.d.ts +1 -0
- package/dist/test/repository-oracle-evidence.test.js +185 -0
- package/dist/test/repository-oracle-plan.test.d.ts +1 -0
- package/dist/test/repository-oracle-plan.test.js +176 -0
- package/dist/test/repository-overlay-isolation.test.d.ts +1 -0
- package/dist/test/repository-overlay-isolation.test.js +85 -0
- package/dist/test/repository-preparation-cache.test.d.ts +1 -0
- package/dist/test/repository-preparation-cache.test.js +414 -0
- package/dist/test/repository-public-artifact.test.d.ts +1 -0
- package/dist/test/repository-public-artifact.test.js +273 -0
- package/dist/test/repository-qualification-diagnostics.test.d.ts +1 -0
- package/dist/test/repository-qualification-diagnostics.test.js +524 -0
- package/dist/test/repository-reference-authoring.test.d.ts +1 -0
- package/dist/test/repository-reference-authoring.test.js +1633 -0
- package/dist/test/repository-review-evidence-v2.test.d.ts +1 -0
- package/dist/test/repository-review-evidence-v2.test.js +183 -0
- package/dist/test/repository-review-evidence.test.d.ts +1 -0
- package/dist/test/repository-review-evidence.test.js +124 -0
- package/dist/test/repository-seed-exclusions.test.d.ts +1 -0
- package/dist/test/repository-seed-exclusions.test.js +96 -0
- package/dist/test/repository-seed-selection.test.d.ts +1 -0
- package/dist/test/repository-seed-selection.test.js +504 -0
- package/dist/test/repository-semantic-calibration.test.d.ts +1 -0
- package/dist/test/repository-semantic-calibration.test.js +688 -0
- package/dist/test/repository-solution-edits.test.d.ts +1 -0
- package/dist/test/repository-solution-edits.test.js +377 -0
- package/dist/test/repository-specification-budget.test.d.ts +1 -0
- package/dist/test/repository-specification-budget.test.js +171 -0
- package/dist/test/repository-specification-contract-checkpoint.test.d.ts +1 -0
- package/dist/test/repository-specification-contract-checkpoint.test.js +228 -0
- package/dist/test/repository-specification-contract-facts.test.d.ts +1 -0
- package/dist/test/repository-specification-contract-facts.test.js +177 -0
- package/dist/test/repository-trajectory-authoring.test.d.ts +1 -0
- package/dist/test/repository-trajectory-authoring.test.js +176 -0
- package/dist/test/repository-valid-control-plan.test.d.ts +1 -0
- package/dist/test/repository-valid-control-plan.test.js +45 -0
- package/dist/test/repository-vitest-phase.test.d.ts +1 -0
- package/dist/test/repository-vitest-phase.test.js +848 -0
- package/dist/test/repository-vitest-reporter.test.d.ts +1 -0
- package/dist/test/repository-vitest-reporter.test.js +158 -0
- package/dist/test/repository-workspace-build.test.d.ts +1 -0
- package/dist/test/repository-workspace-build.test.js +160 -0
- package/dist/test/strict-authoring-schema.test.d.ts +1 -0
- package/dist/test/strict-authoring-schema.test.js +169 -0
- package/package.json +48 -6
|
@@ -0,0 +1 @@
|
|
|
1
|
+
export {};
|
|
@@ -0,0 +1,524 @@
|
|
|
1
|
+
import assert from "node:assert/strict";
|
|
2
|
+
import { execFile } from "node:child_process";
|
|
3
|
+
import { access, mkdtemp, realpath, rm, writeFile } from "node:fs/promises";
|
|
4
|
+
import { tmpdir } from "node:os";
|
|
5
|
+
import { join } from "node:path";
|
|
6
|
+
import { test } from "node:test";
|
|
7
|
+
import { promisify } from "node:util";
|
|
8
|
+
import { Effect } from "effect";
|
|
9
|
+
import { repositoryCommandTestEvidenceV1 } from "../adapters/repository-command-evidence.js";
|
|
10
|
+
import { RepositoryCommandExecutionPort, replayRepositoryCommandsV1 } from "../adapters/repository-command-runner.js";
|
|
11
|
+
import { RepositoryFoundryError } from "../errors.js";
|
|
12
|
+
import { RepositoryFoundryProgressLive } from "../services/foundry-progress/service.js";
|
|
13
|
+
import { qualifyHistoricalTaskSeedAcrossBatchesResumableV1, qualifyHistoricalTaskSeedAcrossBatchesV1 } from "../services/task-seed/service.js";
|
|
14
|
+
const execute = promisify(execFile);
|
|
15
|
+
const CANARY = "private-qualification-command-evidence";
|
|
16
|
+
async function fixture(t) {
|
|
17
|
+
const root = await realpath(await mkdtemp(join(tmpdir(), "qualification-diagnostics-")));
|
|
18
|
+
t.after(() => rm(root, { recursive: true, force: true }));
|
|
19
|
+
const git = (...args) => execute("git", ["-C", root, ...args]);
|
|
20
|
+
await git("init", "-q");
|
|
21
|
+
const testSource = (assertion) => `import test from "node:test";
|
|
22
|
+
import assert from "node:assert/strict";
|
|
23
|
+
import { existsSync, writeFileSync } from "node:fs";
|
|
24
|
+
import { value } from "./subject.mjs";
|
|
25
|
+
test("public contract", () => {
|
|
26
|
+
console.log(${JSON.stringify(CANARY)});
|
|
27
|
+
console.log("diagnostic-cwd=" + process.cwd());
|
|
28
|
+
assert.equal(existsSync("candidate-ran"), false);
|
|
29
|
+
writeFileSync("candidate-ran", "each repetition must be isolated");
|
|
30
|
+
${assertion}
|
|
31
|
+
});
|
|
32
|
+
`;
|
|
33
|
+
await writeFile(join(root, "baseline.test.mjs"), testSource("assert.equal(1, 1);"));
|
|
34
|
+
const commit = async (value) => {
|
|
35
|
+
await writeFile(join(root, "subject.mjs"), `export const value = ${value};\n`);
|
|
36
|
+
await writeFile(join(root, "case.test.mjs"), testSource(`assert.equal(value, ${value});`));
|
|
37
|
+
await git("add", ".");
|
|
38
|
+
await git("-c", "core.hooksPath=/dev/null", "-c", "commit.gpgSign=false", "-c", "user.name=Qualification Diagnostics", "-c", "user.email=fixture@example.test", "commit", "-qm", "fixture");
|
|
39
|
+
return (await git("rev-parse", "HEAD")).stdout.trim();
|
|
40
|
+
};
|
|
41
|
+
const initialCommit = await commit(0);
|
|
42
|
+
const referenceCommit = await commit(1);
|
|
43
|
+
const recipe = (id, file) => ({
|
|
44
|
+
id,
|
|
45
|
+
kind: "test",
|
|
46
|
+
cwd: ".",
|
|
47
|
+
executable: "node",
|
|
48
|
+
args: ["--test", file],
|
|
49
|
+
timeoutMs: 5000
|
|
50
|
+
});
|
|
51
|
+
const seed = {
|
|
52
|
+
version: 1,
|
|
53
|
+
id: "diagnostic-seed",
|
|
54
|
+
status: "candidate",
|
|
55
|
+
source: { kind: "historical-fix", changeEpisodeId: "diagnostic-change" },
|
|
56
|
+
summary: "Return the reference value.",
|
|
57
|
+
capabilityEvidence: [{ kind: "test", id: "case", path: "case.test.mjs" }],
|
|
58
|
+
initialState: { commit: initialCommit },
|
|
59
|
+
referenceCommit,
|
|
60
|
+
targetBehavior: [{ id: "behavior", description: "Return one.", critical: true, evidence: [] }],
|
|
61
|
+
baselineObservations: [],
|
|
62
|
+
preChangeObservations: [],
|
|
63
|
+
postChangeObservations: [],
|
|
64
|
+
candidateTestIds: ["case"],
|
|
65
|
+
environment: {
|
|
66
|
+
protectedControlPaths: [],
|
|
67
|
+
trustedPreparationRecipes: [
|
|
68
|
+
{
|
|
69
|
+
id: "prepare",
|
|
70
|
+
kind: "custom",
|
|
71
|
+
cwd: ".",
|
|
72
|
+
executable: "node",
|
|
73
|
+
args: ["-e", `process.stdout.write(${JSON.stringify(CANARY)});`],
|
|
74
|
+
timeoutMs: 5000
|
|
75
|
+
}
|
|
76
|
+
],
|
|
77
|
+
candidateValidationRecipes: [
|
|
78
|
+
{
|
|
79
|
+
id: "validate",
|
|
80
|
+
kind: "check",
|
|
81
|
+
cwd: ".",
|
|
82
|
+
executable: "node",
|
|
83
|
+
args: ["--check", "subject.mjs"],
|
|
84
|
+
timeoutMs: 5000
|
|
85
|
+
}
|
|
86
|
+
],
|
|
87
|
+
baselineGradeRecipes: [recipe("baseline", "baseline.test.mjs")],
|
|
88
|
+
gradeRecipes: [recipe("grade", "case.test.mjs")]
|
|
89
|
+
},
|
|
90
|
+
confidence: "medium",
|
|
91
|
+
risks: ["review-required"],
|
|
92
|
+
rejectionReasons: []
|
|
93
|
+
};
|
|
94
|
+
return { root, seed };
|
|
95
|
+
}
|
|
96
|
+
function diagnosticRun(data) {
|
|
97
|
+
const artifacts = [];
|
|
98
|
+
const events = [];
|
|
99
|
+
const effect = qualifyHistoricalTaskSeedAcrossBatchesV1({
|
|
100
|
+
repositoryRoot: data.root,
|
|
101
|
+
seed: data.seed,
|
|
102
|
+
batches: 2,
|
|
103
|
+
repetitionsPerBatch: 2
|
|
104
|
+
}).pipe(Effect.provide(RepositoryFoundryProgressLive((event) => Effect.sync(() => {
|
|
105
|
+
events.push(event);
|
|
106
|
+
}), (artifact) => Effect.promise(async () => {
|
|
107
|
+
const report = artifact.value;
|
|
108
|
+
// Observe at recorder invocation, not after the complete qualification:
|
|
109
|
+
// the phase must already have released every candidate checkout.
|
|
110
|
+
for (const result of report.results) {
|
|
111
|
+
const checkout = /diagnostic-cwd=([^\r\n]+)/u.exec(result.stdout)?.[1];
|
|
112
|
+
if (checkout !== undefined)
|
|
113
|
+
await assert.rejects(access(checkout), { code: "ENOENT" });
|
|
114
|
+
}
|
|
115
|
+
artifacts.push(artifact);
|
|
116
|
+
}))));
|
|
117
|
+
return { effect, artifacts, events };
|
|
118
|
+
}
|
|
119
|
+
test("qualification records complete raw phase rows after cleanup without changing batches, repetitions, or model accounting", async (t) => {
|
|
120
|
+
const data = await fixture(t);
|
|
121
|
+
const run = diagnosticRun(data);
|
|
122
|
+
const qualification = await Effect.runPromise(run.effect);
|
|
123
|
+
assert.equal(qualification.seed.status, "qualified");
|
|
124
|
+
assert.equal(qualification.stability.stable, true);
|
|
125
|
+
assert.equal(qualification.stability.batches.length, 2);
|
|
126
|
+
assert.equal(qualification.seed.baselineObservations.length, 4);
|
|
127
|
+
assert.equal(qualification.seed.preChangeObservations.length, 4);
|
|
128
|
+
assert.equal(qualification.seed.postChangeObservations.length, 4);
|
|
129
|
+
assert.equal(run.artifacts.length, 6);
|
|
130
|
+
assert.equal(new Set(run.artifacts.map((artifact) => artifact.operationId)).size, 6);
|
|
131
|
+
assert.deepEqual(run.artifacts.map((artifact) => artifact.value.batch), [1, 1, 1, 2, 2, 2]);
|
|
132
|
+
for (const batch of [1, 2]) {
|
|
133
|
+
const reports = run.artifacts
|
|
134
|
+
.map((artifact) => artifact.value)
|
|
135
|
+
.filter((report) => report.batch === batch);
|
|
136
|
+
assert.deepEqual(reports.map((report) => report.phase).sort(), [
|
|
137
|
+
"baseline",
|
|
138
|
+
"post-change",
|
|
139
|
+
"pre-change"
|
|
140
|
+
]);
|
|
141
|
+
}
|
|
142
|
+
for (const artifact of run.artifacts) {
|
|
143
|
+
const report = artifact.value;
|
|
144
|
+
assert.equal(artifact.response.text, JSON.stringify(report));
|
|
145
|
+
assert.equal(artifact.role, "seed-qualifier");
|
|
146
|
+
assert.equal(artifact.model, "local-replay");
|
|
147
|
+
assert.equal(artifact.request.maximumOutputTokens, 0);
|
|
148
|
+
assert.equal(report.diagnosticOnly, true);
|
|
149
|
+
assert.equal(report.admissionEvidence, false);
|
|
150
|
+
assert.equal(report.seedId, data.seed.id);
|
|
151
|
+
assert.equal(report.referenceCommit, data.seed.referenceCommit);
|
|
152
|
+
assert.deepEqual(report.results.map((result) => result.repetition), [1, 2]);
|
|
153
|
+
for (const result of report.results) {
|
|
154
|
+
assert.equal(result.commit, report.phase === "post-change" ? data.seed.referenceCommit : data.seed.initialState.commit);
|
|
155
|
+
assert.equal(result.gradeExecuted, true);
|
|
156
|
+
assert.equal(result.stage, "grade");
|
|
157
|
+
assert.equal(result.exitCode, report.phase === "pre-change" ? 1 : 0);
|
|
158
|
+
assert.equal(repositoryCommandTestEvidenceV1(result, result.recipeId).outcome, report.phase === "pre-change" ? "fail" : "pass");
|
|
159
|
+
assert.equal(result.timedOut, false);
|
|
160
|
+
assert.equal(result.outputTruncated, false);
|
|
161
|
+
assert.ok(result.stdout.includes(CANARY));
|
|
162
|
+
assert.equal(result.trustedPreparationResults.length, 1);
|
|
163
|
+
assert.equal(result.trustedPreparationResults[0].stdout, CANARY);
|
|
164
|
+
assert.equal(result.trustedPreparationResults[0].stage, "trusted-preparation");
|
|
165
|
+
assert.equal(result.candidateValidationResults.length, 1);
|
|
166
|
+
assert.equal(result.candidateValidationResults[0].stage, "candidate-validation");
|
|
167
|
+
assert.ok(result.isolationTimings);
|
|
168
|
+
for (const duration of Object.values(result.isolationTimings))
|
|
169
|
+
assert.ok(duration >= 0);
|
|
170
|
+
assert.ok(report.wallMs >= result.durationMs);
|
|
171
|
+
}
|
|
172
|
+
}
|
|
173
|
+
assert.equal(run.events.length, 6);
|
|
174
|
+
assert.ok(run.events.every((event) => event.kind === "heartbeat" && event.modelCalls === 0 && event.role === "seed-qualifier"));
|
|
175
|
+
assert.doesNotMatch(JSON.stringify(run.events), /private-qualification|stdout|stderr|diagnostic-cwd|case\.test|AssertionError/u);
|
|
176
|
+
});
|
|
177
|
+
test("interrupted cross-batch qualification resumes after the last durable repetition", async (t) => {
|
|
178
|
+
const data = await fixture(t);
|
|
179
|
+
let retained;
|
|
180
|
+
const checkpointed = [];
|
|
181
|
+
const checkpoint = (progress, interruptAfterTwo) => Effect.gen(function* () {
|
|
182
|
+
retained = structuredClone(progress);
|
|
183
|
+
const unit = progress.activeUnits.at(-1);
|
|
184
|
+
if (unit !== undefined) {
|
|
185
|
+
checkpointed.push(`${String(unit.batch)}:${unit.phase}:${String(unit.repetition)}`);
|
|
186
|
+
}
|
|
187
|
+
if (interruptAfterTwo && progress.activeUnits.length === 2) {
|
|
188
|
+
return yield* new RepositoryFoundryError({
|
|
189
|
+
operation: "extract-task-seeds",
|
|
190
|
+
detail: "injected interruption after two durable repetitions"
|
|
191
|
+
});
|
|
192
|
+
}
|
|
193
|
+
});
|
|
194
|
+
await assert.rejects(Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
195
|
+
repositoryRoot: data.root,
|
|
196
|
+
seed: data.seed,
|
|
197
|
+
bindingDigest: "binding-v1",
|
|
198
|
+
batches: 2,
|
|
199
|
+
repetitionsPerBatch: 2,
|
|
200
|
+
checkpoint: (progress) => checkpoint(progress, true)
|
|
201
|
+
})), /injected interruption/u);
|
|
202
|
+
assert.ok(retained);
|
|
203
|
+
assert.deepEqual(checkpointed, ["1:baseline:1", "1:baseline:2"]);
|
|
204
|
+
const resumed = await Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
205
|
+
repositoryRoot: data.root,
|
|
206
|
+
seed: data.seed,
|
|
207
|
+
bindingDigest: "binding-v1",
|
|
208
|
+
progress: retained,
|
|
209
|
+
batches: 2,
|
|
210
|
+
repetitionsPerBatch: 2,
|
|
211
|
+
checkpoint: (progress) => checkpoint(progress, false)
|
|
212
|
+
}));
|
|
213
|
+
assert.equal(resumed.qualification.seed.status, "qualified");
|
|
214
|
+
assert.equal(resumed.progress.completedBatches.length, 2);
|
|
215
|
+
assert.deepEqual(resumed.progress.activeUnits, []);
|
|
216
|
+
assert.equal(checkpointed.length, 12);
|
|
217
|
+
assert.equal(new Set(checkpointed).size, 12);
|
|
218
|
+
assert.equal(checkpointed[2], "1:pre-change:1");
|
|
219
|
+
});
|
|
220
|
+
test("an incomplete grading unit blocks qualification durably before later repetitions or batches", async (t) => {
|
|
221
|
+
const data = await fixture(t);
|
|
222
|
+
data.seed = {
|
|
223
|
+
...data.seed,
|
|
224
|
+
environment: {
|
|
225
|
+
...data.seed.environment,
|
|
226
|
+
baselineGradeRecipes: [
|
|
227
|
+
...data.seed.environment.baselineGradeRecipes,
|
|
228
|
+
{ ...data.seed.environment.baselineGradeRecipes[0], id: "must-not-start-after-timeout" }
|
|
229
|
+
]
|
|
230
|
+
}
|
|
231
|
+
};
|
|
232
|
+
let retained;
|
|
233
|
+
let commands = 0;
|
|
234
|
+
let grades = 0;
|
|
235
|
+
const run = (progress) => Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
236
|
+
repositoryRoot: data.root, seed: data.seed, bindingDigest: "binding-v1",
|
|
237
|
+
...(progress === undefined ? {} : { progress }),
|
|
238
|
+
checkpoint: (value) => Effect.sync(() => { retained = structuredClone(value); })
|
|
239
|
+
}).pipe(Effect.provideService(RepositoryCommandExecutionPort, {
|
|
240
|
+
run: (input) => Effect.sync(() => {
|
|
241
|
+
commands += 1;
|
|
242
|
+
if (input.args.includes("--test")) {
|
|
243
|
+
grades += 1;
|
|
244
|
+
return { exitCode: null, timedOut: true, stdout: CANARY, stderr: "" };
|
|
245
|
+
}
|
|
246
|
+
return { exitCode: 0, stdout: "", stderr: "" };
|
|
247
|
+
})
|
|
248
|
+
})));
|
|
249
|
+
await assert.rejects(run(), /operational failure prevented grading.*stage=grade.*timeout=true/u);
|
|
250
|
+
assert.equal(grades, 1, "do not start the second repetition after a grader timeout");
|
|
251
|
+
assert.equal(retained?.activeUnits.length, 1, "retain the completed failure for diagnostics");
|
|
252
|
+
const completedCommands = commands;
|
|
253
|
+
for (const result of [
|
|
254
|
+
{ timedOut: true, exitCode: null, outputTruncated: false },
|
|
255
|
+
{ timedOut: false, exitCode: null, outputTruncated: false },
|
|
256
|
+
{ timedOut: false, exitCode: 0, outputTruncated: true }
|
|
257
|
+
]) {
|
|
258
|
+
const progress = structuredClone(retained);
|
|
259
|
+
const first = progress.activeUnits[0];
|
|
260
|
+
const changed = {
|
|
261
|
+
...progress,
|
|
262
|
+
activeUnits: [{ ...first, results: first.results.map((row) => ({ ...row, ...result })) }]
|
|
263
|
+
};
|
|
264
|
+
const failure = await Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
265
|
+
repositoryRoot: data.root, seed: data.seed, bindingDigest: "binding-v1",
|
|
266
|
+
progress: changed, checkpoint: () => Effect.void
|
|
267
|
+
}).pipe(Effect.result));
|
|
268
|
+
assert.equal(failure._tag, "Failure");
|
|
269
|
+
if (failure._tag === "Failure")
|
|
270
|
+
assert.equal(failure.failure.classification, "environment_blocked");
|
|
271
|
+
await assert.rejects(run(changed), /operational failure prevented grading/u);
|
|
272
|
+
assert.equal(commands, completedCommands, "unchanged failed execution is never replayed");
|
|
273
|
+
}
|
|
274
|
+
});
|
|
275
|
+
for (const { timedOut, continueAfterIncompleteGrade } of [false, true].flatMap((timedOut) => [false, true].map((continueAfterIncompleteGrade) => ({ timedOut, continueAfterIncompleteGrade })))) {
|
|
276
|
+
test(`incomplete replay preserves independent repetitions and explicit diagnostic continuation (timeout=${String(timedOut)}; continue=${String(continueAfterIncompleteGrade)})`, async (t) => {
|
|
277
|
+
const data = await fixture(t);
|
|
278
|
+
const checkouts = [];
|
|
279
|
+
const results = await Effect.runPromise(replayRepositoryCommandsV1({
|
|
280
|
+
repositoryRoot: data.root,
|
|
281
|
+
commit: data.seed.initialState.commit,
|
|
282
|
+
recipes: [
|
|
283
|
+
data.seed.environment.gradeRecipes[0],
|
|
284
|
+
{ ...data.seed.environment.gradeRecipes[0], id: "queued-grade" }
|
|
285
|
+
],
|
|
286
|
+
repetitions: 3,
|
|
287
|
+
...(continueAfterIncompleteGrade ? { continueAfterIncompleteGrade } : {})
|
|
288
|
+
}).pipe(Effect.provideService(RepositoryCommandExecutionPort, {
|
|
289
|
+
run: (input) => Effect.sync(() => {
|
|
290
|
+
checkouts.push(input.cwd);
|
|
291
|
+
return { exitCode: null, timedOut, stdout: CANARY, stderr: "" };
|
|
292
|
+
})
|
|
293
|
+
})));
|
|
294
|
+
const expectedCommands = continueAfterIncompleteGrade ? 6 : 3;
|
|
295
|
+
assert.equal(checkouts.length, expectedCommands);
|
|
296
|
+
assert.equal(new Set(checkouts).size, 3);
|
|
297
|
+
assert.equal(results.length, expectedCommands);
|
|
298
|
+
assert.equal(results[0].timedOut, timedOut);
|
|
299
|
+
assert.equal(results[0].stdout, CANARY);
|
|
300
|
+
for (const checkout of checkouts)
|
|
301
|
+
await assert.rejects(access(checkout), { code: "ENOENT" });
|
|
302
|
+
});
|
|
303
|
+
}
|
|
304
|
+
test("failed trusted preparation retains completed raw rows and skips queued phases and later batches", async (t) => {
|
|
305
|
+
const data = await fixture(t);
|
|
306
|
+
data.seed = {
|
|
307
|
+
...data.seed,
|
|
308
|
+
environment: {
|
|
309
|
+
...data.seed.environment,
|
|
310
|
+
trustedPreparationRecipes: [
|
|
311
|
+
{
|
|
312
|
+
...data.seed.environment.trustedPreparationRecipes[0],
|
|
313
|
+
args: [
|
|
314
|
+
"-e",
|
|
315
|
+
`process.stdout.write(${JSON.stringify(CANARY)}); process.stderr.write("private preparation failure"); process.exit(7);`
|
|
316
|
+
]
|
|
317
|
+
}
|
|
318
|
+
]
|
|
319
|
+
}
|
|
320
|
+
};
|
|
321
|
+
const run = diagnosticRun(data);
|
|
322
|
+
await assert.rejects(Effect.runPromise(run.effect), /operational failure prevented grading.*stage=trusted-preparation/u);
|
|
323
|
+
assert.equal(run.artifacts.length, 1, "later phases and batches never start");
|
|
324
|
+
assert.deepEqual(run.artifacts.map((artifact) => artifact.value.phase), [
|
|
325
|
+
"baseline"
|
|
326
|
+
]);
|
|
327
|
+
assert.ok(run.artifacts.every((artifact) => artifact.value.batch === 1));
|
|
328
|
+
for (const artifact of run.artifacts) {
|
|
329
|
+
const report = artifact.value;
|
|
330
|
+
assert.equal(report.results.length, 2);
|
|
331
|
+
for (const result of report.results) {
|
|
332
|
+
assert.equal(result.recipeId, "prepare");
|
|
333
|
+
assert.equal(result.stage, "trusted-preparation");
|
|
334
|
+
assert.equal(result.gradeExecuted, false);
|
|
335
|
+
assert.equal(result.exitCode, 7);
|
|
336
|
+
assert.equal(repositoryCommandTestEvidenceV1(result, result.recipeId).outcome, "unknown");
|
|
337
|
+
assert.equal(result.stdout, CANARY);
|
|
338
|
+
assert.equal(result.stderr, "private preparation failure");
|
|
339
|
+
assert.equal(result.trustedPreparationResults.length, 1);
|
|
340
|
+
assert.equal(result.trustedPreparationResults[0].exitCode, 7);
|
|
341
|
+
assert.equal(result.candidateValidationResults.length, 0);
|
|
342
|
+
assert.ok(result.isolationTimings);
|
|
343
|
+
assert.ok(report.wallMs >= result.durationMs);
|
|
344
|
+
}
|
|
345
|
+
}
|
|
346
|
+
assert.equal(run.events.length, 2);
|
|
347
|
+
assert.ok(run.events.every((event) => event.modelCalls === 0));
|
|
348
|
+
assert.doesNotMatch(JSON.stringify(run.events), /unstable|flaky/u);
|
|
349
|
+
assert.doesNotMatch(JSON.stringify(run.events), /private preparation failure|private-qualification|stdout|stderr/u);
|
|
350
|
+
});
|
|
351
|
+
for (const timedOut of [false, true]) {
|
|
352
|
+
test(`candidate-validation ${timedOut ? "timeout" : "failure"} finishes cleanup before skipping later phases`, {
|
|
353
|
+
timeout: 30000
|
|
354
|
+
}, async (t) => {
|
|
355
|
+
const data = await fixture(t);
|
|
356
|
+
data.seed = {
|
|
357
|
+
...data.seed,
|
|
358
|
+
environment: {
|
|
359
|
+
...data.seed.environment,
|
|
360
|
+
candidateValidationRecipes: [
|
|
361
|
+
{
|
|
362
|
+
...data.seed.environment.candidateValidationRecipes[0],
|
|
363
|
+
args: [
|
|
364
|
+
"-e",
|
|
365
|
+
`console.log(${JSON.stringify(CANARY)}); console.log("diagnostic-cwd=" + process.cwd()); process.stderr.write("private validation failure"); ${timedOut ? "setInterval(() => {}, 1000);" : "process.exit(23);"}`
|
|
366
|
+
],
|
|
367
|
+
timeoutMs: timedOut ? 500 : 5000
|
|
368
|
+
}
|
|
369
|
+
]
|
|
370
|
+
}
|
|
371
|
+
};
|
|
372
|
+
const run = diagnosticRun(data);
|
|
373
|
+
await assert.rejects(Effect.runPromise(run.effect), /operational failure prevented grading.*stage=candidate-validation/u);
|
|
374
|
+
assert.equal(run.artifacts.length, 1);
|
|
375
|
+
assert.deepEqual(run.artifacts.map((artifact) => artifact.value.phase), [
|
|
376
|
+
"baseline"
|
|
377
|
+
]);
|
|
378
|
+
for (const artifact of run.artifacts) {
|
|
379
|
+
const report = artifact.value;
|
|
380
|
+
assert.equal(report.batch, 1);
|
|
381
|
+
assert.deepEqual(report.results.map((result) => result.repetition), [1, 2]);
|
|
382
|
+
for (const result of report.results) {
|
|
383
|
+
assert.equal(result.gradeExecuted, false);
|
|
384
|
+
assert.equal(result.stage, "candidate-validation");
|
|
385
|
+
assert.equal(result.exitCode, timedOut ? null : 23);
|
|
386
|
+
assert.equal(result.timedOut, timedOut);
|
|
387
|
+
assert.equal(repositoryCommandTestEvidenceV1(result, result.recipeId).outcome, "unknown");
|
|
388
|
+
assert.equal(result.trustedPreparationResults[0].exitCode, 0);
|
|
389
|
+
assert.equal(result.candidateValidationResults[0].stderr, "private validation failure");
|
|
390
|
+
assert.match(result.stdout, /diagnostic-cwd=/u);
|
|
391
|
+
assert.ok(result.isolationTimings);
|
|
392
|
+
}
|
|
393
|
+
}
|
|
394
|
+
assert.equal(run.events.length, 2);
|
|
395
|
+
assert.doesNotMatch(JSON.stringify(run.events), /private validation failure|private-qualification|stdout|stderr|unstable|flaky/u);
|
|
396
|
+
});
|
|
397
|
+
}
|
|
398
|
+
test("unknown grade execution preserves diagnostics and stops queued qualification phases and batches", async (t) => {
|
|
399
|
+
const data = await fixture(t);
|
|
400
|
+
data.seed = {
|
|
401
|
+
...data.seed,
|
|
402
|
+
environment: {
|
|
403
|
+
...data.seed.environment,
|
|
404
|
+
gradeRecipes: [{ ...data.seed.environment.gradeRecipes[0], args: ["-e", "process.exit(1)"] }]
|
|
405
|
+
}
|
|
406
|
+
};
|
|
407
|
+
const run = diagnosticRun(data);
|
|
408
|
+
await assert.rejects(Effect.runPromise(run.effect), /operational failure prevented grading/u);
|
|
409
|
+
assert.equal(run.artifacts.length, 2);
|
|
410
|
+
assert.deepEqual(run.artifacts.map((artifact) => artifact.value.phase), ["baseline", "pre-change"]);
|
|
411
|
+
for (const artifact of run.artifacts) {
|
|
412
|
+
const report = artifact.value;
|
|
413
|
+
assert.equal(report.results.length, 2);
|
|
414
|
+
for (const result of report.results) {
|
|
415
|
+
assert.equal(result.gradeExecuted, true);
|
|
416
|
+
assert.equal(result.stage, "grade");
|
|
417
|
+
assert.equal(repositoryCommandTestEvidenceV1(result, result.recipeId).outcome, report.phase === "baseline" ? "pass" : "unknown");
|
|
418
|
+
}
|
|
419
|
+
}
|
|
420
|
+
assert.equal(run.events.length, 3);
|
|
421
|
+
});
|
|
422
|
+
test("unattributable grading is retained once and cannot be retried without changing the execution binding", async (t) => {
|
|
423
|
+
const data = await fixture(t);
|
|
424
|
+
let retained;
|
|
425
|
+
let grades = 0;
|
|
426
|
+
const run = (progress) => Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
427
|
+
repositoryRoot: data.root,
|
|
428
|
+
seed: data.seed,
|
|
429
|
+
bindingDigest: "unknown-evidence-v1",
|
|
430
|
+
...(progress === undefined ? {} : { progress }),
|
|
431
|
+
checkpoint: (value) => Effect.sync(() => { retained = structuredClone(value); })
|
|
432
|
+
}).pipe(Effect.provideService(RepositoryCommandExecutionPort, {
|
|
433
|
+
run: (input) => Effect.sync(() => {
|
|
434
|
+
if (input.args.includes("--test")) {
|
|
435
|
+
grades += 1;
|
|
436
|
+
return { exitCode: 1, stdout: "", stderr: "test setup failed without assertion evidence" };
|
|
437
|
+
}
|
|
438
|
+
return { exitCode: 0, stdout: "", stderr: "" };
|
|
439
|
+
})
|
|
440
|
+
})));
|
|
441
|
+
await assert.rejects(run(), /operational failure prevented grading/u);
|
|
442
|
+
assert.equal(grades, 1);
|
|
443
|
+
assert.equal(retained?.activeUnits.length, 1);
|
|
444
|
+
await assert.rejects(run(retained), /operational failure prevented grading/u);
|
|
445
|
+
assert.equal(grades, 1);
|
|
446
|
+
});
|
|
447
|
+
test("pre-change compilation blocks resumable qualification once and preserves the exact failed unit", async (t) => {
|
|
448
|
+
const data = await fixture(t);
|
|
449
|
+
const seed = {
|
|
450
|
+
...data.seed,
|
|
451
|
+
environment: {
|
|
452
|
+
...data.seed.environment,
|
|
453
|
+
candidateValidationRecipes: [{
|
|
454
|
+
...data.seed.environment.candidateValidationRecipes[0],
|
|
455
|
+
args: ["-e", [
|
|
456
|
+
"const fs = require('node:fs');",
|
|
457
|
+
"const reviewed = fs.readFileSync('case.test.mjs', 'utf8').includes('assert.equal(value, 1);');",
|
|
458
|
+
"const initial = fs.readFileSync('subject.mjs', 'utf8').includes('value = 0');",
|
|
459
|
+
"if (reviewed && initial) process.exit(2);"
|
|
460
|
+
].join(" ")]
|
|
461
|
+
}]
|
|
462
|
+
}
|
|
463
|
+
};
|
|
464
|
+
let retained;
|
|
465
|
+
const effect = (progress) => qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
466
|
+
repositoryRoot: data.root,
|
|
467
|
+
seed,
|
|
468
|
+
bindingDigest: "compile-only-seed-v1",
|
|
469
|
+
...(progress === undefined ? {} : { progress }),
|
|
470
|
+
checkpoint: (value) => Effect.sync(() => { retained = structuredClone(value); })
|
|
471
|
+
});
|
|
472
|
+
await assert.rejects(Effect.runPromise(effect()), /operational failure prevented grading.*phase=pre-change; stage=candidate-validation/u);
|
|
473
|
+
assert.ok(retained);
|
|
474
|
+
assert.equal(retained.completedBatches.length, 0);
|
|
475
|
+
assert.deepEqual(retained.activeUnits.map(({ phase }) => phase), ["baseline", "baseline", "pre-change"]);
|
|
476
|
+
const failed = retained.activeUnits.at(-1).results[0];
|
|
477
|
+
assert.equal(failed.gradeExecuted, false);
|
|
478
|
+
assert.equal(failed.exitCode, 2);
|
|
479
|
+
assert.equal(repositoryCommandTestEvidenceV1(failed, failed.recipeId).outcome, "unknown");
|
|
480
|
+
let resumedCommands = 0;
|
|
481
|
+
await assert.rejects(Effect.runPromise(effect(retained).pipe(Effect.provideService(RepositoryCommandExecutionPort, {
|
|
482
|
+
run: () => Effect.sync(() => {
|
|
483
|
+
resumedCommands += 1;
|
|
484
|
+
return { exitCode: 0, stdout: "", stderr: "" };
|
|
485
|
+
})
|
|
486
|
+
}))), /operational failure prevented grading.*phase=pre-change/u);
|
|
487
|
+
assert.equal(resumedCommands, 0, "resuming must not repeat the failed build or start later phases");
|
|
488
|
+
});
|
|
489
|
+
test("legacy completed compile-only batches cannot bypass the executed evidence contract", async (t) => {
|
|
490
|
+
const data = await fixture(t);
|
|
491
|
+
const observation = (subjectId, outcome, detail) => ({
|
|
492
|
+
id: subjectId, kind: "test-command", subjectId, outcome, detail
|
|
493
|
+
});
|
|
494
|
+
const legacySeed = {
|
|
495
|
+
...data.seed,
|
|
496
|
+
status: "qualified",
|
|
497
|
+
baselineObservations: [observation("baseline", "pass", "stage=grade; gradeExecuted=true")],
|
|
498
|
+
preChangeObservations: [observation("grade", "fail", "stage=candidate-validation; gradeExecuted=false; exit=2")],
|
|
499
|
+
postChangeObservations: [observation("grade", "pass", "stage=grade; gradeExecuted=true")]
|
|
500
|
+
};
|
|
501
|
+
let commands = 0;
|
|
502
|
+
let checkpoints = 0;
|
|
503
|
+
await assert.rejects(Effect.runPromise(qualifyHistoricalTaskSeedAcrossBatchesResumableV1({
|
|
504
|
+
repositoryRoot: data.root,
|
|
505
|
+
seed: data.seed,
|
|
506
|
+
bindingDigest: "legacy-completed-v1",
|
|
507
|
+
progress: {
|
|
508
|
+
version: 1,
|
|
509
|
+
bindingDigest: "legacy-completed-v1",
|
|
510
|
+
batchesRequired: 2,
|
|
511
|
+
repetitionsPerBatch: 2,
|
|
512
|
+
completedBatches: [{ batch: 1, seed: legacySeed }],
|
|
513
|
+
activeUnits: []
|
|
514
|
+
},
|
|
515
|
+
checkpoint: () => Effect.sync(() => { checkpoints += 1; })
|
|
516
|
+
}).pipe(Effect.provideService(RepositoryCommandExecutionPort, {
|
|
517
|
+
run: () => Effect.sync(() => {
|
|
518
|
+
commands += 1;
|
|
519
|
+
return { exitCode: 0, stdout: "", stderr: "" };
|
|
520
|
+
})
|
|
521
|
+
}))), /retained qualification does not satisfy the executed behavioral evidence contract/u);
|
|
522
|
+
assert.equal(commands, 0);
|
|
523
|
+
assert.equal(checkpoints, 0);
|
|
524
|
+
});
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
export {};
|