문제: 기능 하나를 더하면(리뷰 16 실험 E) 기능 밖 테스트 3개가 깨진다.
① installed-context-routing은 context 목록이 정확히
["reference-context"]이고 첫 context가 참조 context라고 단정한다.
② product-feature-switch는 정렬 첫 기능(INSTALLED_PRODUCT_FEATURE_IDS[0])을
끄고 참조 라우트를 본다. 새 기능이 앞에 정렬되면 깨진다.
③ release-manifest fixture가 contractSet.packages: []로 고정이다.
EXTERNAL_PACKAGE를 쓰는 첫 기능에서 반드시 깨진다.
또 시나리오 하네스의 operation ID 타입이 테스트 계약 3개 리터럴이고
시나리오 응답 함수가 /api/test-entities 핸들러 안에 묶여 있다. 새 기능
테스트가 자기 operation에 실패 시나리오를 쓰려면 카탈로그를 고쳐야 한다
(안 고치면 TS2345, 실행하면 catalog.ts:469 TypeError).
방법: 리뷰의 (b) 최소 개선.
① "목록에 reference-context가 있다"로 바꾸고, 라우트 검사는 id로 찾은
context에서 한다.
② REFERENCE_FEATURE_ID를 쓴다.
③ fixture의 packages와 setDigest를 EXPECTED_CONTRACT_SET_PACKAGES에서
만든다.
하네스: HttpScenarioOperationId를 string으로 넓혔다. 시나리오 사실(명령
여부, 응답 상한, 성공 status)은 httpScenarioOperationFacts가 설치된 V3
descriptor에서 읽는다. 시나리오 응답은 tests/mocks/handlers/
http-scenarios.ts의 createHttpScenarioHandler로 옮겼다. 경로와 method도
descriptor에서 만든다. test-entities 핸들러는 이 함수를 쓰고 카탈로그
확인(assertOperationScenario)을 그대로 거친다. 카탈로그 기대값 57개는
계속 테스트 계약 3개에만 있고, 그 타입도 3개로 둔다. 카탈로그에 없는
operation을 assertOperationScenario에 넘기면 TypeError 대신 그 사실을
말하는 오류가 난다. test-entities 핸들러는 요청의 이름·id를 되돌려 주던
응답을 카탈로그 payload로 고정했다(기대값과 같은 값이다).
새 기능 테스트의 MSW 시나리오: server.use(createHttpScenarioHandler(
<자기 설치 계약>, { scenario: "server-terminal-503", payload }))
버린 대안: (a) 현상 유지는 기능마다 4곳을 고친다. (c) 하네스를 기능별로
나누면 변경이 크고 test-evidence.json의 증거 경로가 바뀐다. 기능
operation을 카탈로그에 더하는 방식은 기능 추가를 공유 파일 수정으로
만든다.
test-evidence.json은 바꾸지 않았다(sourceContracts 토큰이 test-entities.ts에
그대로 있다). 제거 드릴의 플랫폼 보존 목록에 http-scenarios.ts를 더했고,
테스트 전략 문서 11절에 사용법을 적었다.
위험·한계:
- 앱 전체를 렌더하는 테스트에서 response-too-large는 끝나지 않는다.
실행기가 MSW가 넘긴 본문을 cancel하는 promise가 끝나지 않는 기존
제약이다. 카탈로그 통합 테스트는 프록시 fetcher로 우회한다.
- 같은 실험에서 기능 밖 테스트가 더 깨진다. 이 발견 범위 밖이라 두었다.
http-operation-v2-v3-parity는 스키마 ID "NoRequest"를 센티널로 고정한다.
스키마 ID는 전역이라 두 번째 기능은 다른 이름을 써야 하고, 그러면 이
테스트가 깨진다. manual-a11y-evidence·documentation-readiness(라우트별
a11y 기록, 16-01)와 second-api-feature-registries(무효화 그래프 추가가
breaking, 16-01)도 깨진다.
- test:sample-removal, test:http-scenario-evidence는 pnpm이 필요해 돌리지
않았다.
검증:
RED: 참조 기능을 토큰 경계로 order로 바꾼 복사본(order-context,
EXTERNAL_PACKAGE 출처 포함, 커밋 안 함)을 넣고 tests/features
tests/component tests/runtime-schema를 돌렸다. 5파일 8개 실패. 기능 밖은
① 2개(["order-context","reference-context"]), ② 2개(order-feature를 꺼서
참조 링크가 남음), ③ 1개(ReleaseManifestError)다. 나머지 3개는 복사한
테스트다. 하네스 probe는 TS2345 2건, 실행 시 TypeError(catalog.ts:469).
새 통합 테스트는 기준 하네스에서 모듈 없음으로 실패하고, 기준
assertOperationScenario는 미등록 operation에 TypeError를 낸다.
GREEN: 같은 복사본으로 같은 폴더와 카탈로그 통합 테스트를 돌렸다. 78파일
중 실패는 출처를 바꾼 복사본 provenance 1파일뿐이고 ①~③은 통과한다.
order 테스트가 createHttpScenarioHandler(LIST_ORDER_RESOURCES_CONTRACT)로
content-type-mismatch·schema-mismatch·server-terminal-503을 화면에서
확인했다(카탈로그 수정 없음).
복사본을 지운 뒤: tests/features·component·runtime-schema·unit·
integration(security-followup-archive 제외)·contract 250파일 3111개 통과,
카탈로그 57/57, check-test-evidence --scenario-only·--source-only PASS,
tsc app·test·node 통과, verify-documentation-readiness PASS_SCOPED,
check-ci-contract PASS, 바꾼 파일 eslint 통과.
Refs: docs/reviews/2026-11-01-maturity-review/16-structure-patterns.md 16-02
462 lines
15 KiB
TypeScript
462 lines
15 KiB
TypeScript
import { mkdir, rm } from "node:fs/promises";
|
|
import path from "node:path";
|
|
|
|
import { afterAll, beforeAll, describe, expect, it } from "vitest";
|
|
|
|
import { readBoundedBytes } from "../../src/adapters/http/bounded-body-reader.ts";
|
|
import {
|
|
createContractHttpExecutor,
|
|
type HttpExecutionObservation,
|
|
} from "../../src/adapters/http/http-execution-v3.ts";
|
|
import {
|
|
type InstalledHttpContract,
|
|
} from "../../src/contracts/external-contract-runtime.ts";
|
|
import type { CacheScopeSnapshot } from "../../src/contracts/server-state-scope.ts";
|
|
import {
|
|
computeHttpScenarioCatalogDigest,
|
|
httpScenarioReceiptSchema,
|
|
type AttemptStatus,
|
|
type BodyDisposition,
|
|
type HttpScenarioAssertionGroups,
|
|
type RetryReason,
|
|
} from "../../scripts/lib/http-scenario-evidence.ts";
|
|
import { writeValidatedJsonArtifact } from "../../scripts/lib/validated-json-artifact.ts";
|
|
import { TEST_LIST_HTTP_CONTRACT } from "../helpers/external-contract-fixture.ts";
|
|
import { createHttpScenarioHandler } from "../mocks/handlers/http-scenarios.ts";
|
|
import { createTestEntityScenarioHandlers } from "../mocks/handlers/test-entities.ts";
|
|
import { createStrictMockServer } from "../mocks/server.ts";
|
|
import {
|
|
assertOperationScenario,
|
|
HTTP_SCENARIO_CONTRACTS,
|
|
HTTP_SCENARIO_EXECUTION_IDS,
|
|
HTTP_SCENARIO_EXPECTATIONS,
|
|
HTTP_SCENARIO_RECEIPT_SCHEMA_VERSION,
|
|
OPERATION_SCENARIO_CATALOG,
|
|
type HttpScenarioExpectation,
|
|
} from "../mocks/scenarios/catalog.ts";
|
|
|
|
/** A low-cardinality route identity, as a feature binding passes it. */
|
|
const ROUTE_ID = "TEST_ENTITY_ROUTE";
|
|
|
|
const RECEIPT_PATH = path.resolve(
|
|
"artifacts/tests/http-scenario-executions.json",
|
|
);
|
|
const mockApi = createStrictMockServer();
|
|
|
|
beforeAll(mockApi.listen);
|
|
afterAll(mockApi.close);
|
|
|
|
type AttemptTrace = {
|
|
status: AttemptStatus;
|
|
media: string | null;
|
|
pulledBytes: number;
|
|
appliedCeiling: number | null;
|
|
completed: boolean;
|
|
cancelled: boolean;
|
|
};
|
|
|
|
type CatalogOperationId = HttpScenarioExpectation["operationId"];
|
|
|
|
function operationFor(operationId: CatalogOperationId) {
|
|
return HTTP_SCENARIO_CONTRACTS[operationId] as InstalledHttpContract<
|
|
unknown,
|
|
unknown,
|
|
unknown
|
|
>;
|
|
}
|
|
|
|
function inputFor(operationId: CatalogOperationId): unknown {
|
|
if (operationId === "TEST_LIST_ENTITIES") return { limit: 20 };
|
|
if (operationId === "TEST_GET_ENTITY") return { entityId: "entity-1" };
|
|
return { name: "Created" };
|
|
}
|
|
|
|
function normalizeMedia(response: Response): string | null {
|
|
const value = response.headers.get("content-type");
|
|
return value?.split(";", 1)[0]?.trim().toLowerCase() || null;
|
|
}
|
|
|
|
function tracingFetcher(attempts: AttemptTrace[]): typeof fetch {
|
|
return async (input, init) => {
|
|
const trace: AttemptTrace = {
|
|
status: "PENDING_ABORT",
|
|
media: null,
|
|
pulledBytes: 0,
|
|
appliedCeiling: null,
|
|
completed: false,
|
|
cancelled: false,
|
|
};
|
|
attempts.push(trace);
|
|
let response: Response;
|
|
try {
|
|
response = await fetch(input, init);
|
|
} catch (error) {
|
|
trace.status = init?.signal?.aborted
|
|
? "PENDING_ABORT"
|
|
: "NETWORK_REJECTION";
|
|
throw error;
|
|
}
|
|
trace.status = response.status;
|
|
trace.media = normalizeMedia(response);
|
|
if (!response.body) return response;
|
|
|
|
const reader = response.body.getReader();
|
|
const proxy = new ReadableStream<Uint8Array>(
|
|
{
|
|
async pull(controller) {
|
|
try {
|
|
const chunk = await reader.read();
|
|
if (chunk.done) {
|
|
trace.completed = true;
|
|
controller.close();
|
|
return;
|
|
}
|
|
trace.pulledBytes += chunk.value.byteLength;
|
|
controller.enqueue(chunk.value);
|
|
} catch (error) {
|
|
controller.error(error);
|
|
}
|
|
},
|
|
cancel(reason) {
|
|
trace.cancelled = true;
|
|
// MSW-transferred oversized bodies leave the original cancel promise
|
|
// permanently pending. Record and request cancellation without making
|
|
// the proxy response less responsive than the real executor contract.
|
|
void reader.cancel(reason).catch(() => {});
|
|
},
|
|
},
|
|
{ highWaterMark: 0 },
|
|
);
|
|
return new Response(proxy, {
|
|
status: response.status,
|
|
statusText: response.statusText,
|
|
headers: response.headers,
|
|
});
|
|
};
|
|
}
|
|
|
|
function outcomeDetail(outcome: Readonly<Record<string, unknown>>): string | null {
|
|
if (outcome.kind === "CONTRACT_VIOLATION") {
|
|
return String((outcome.violation as Readonly<{ kind: string }>).kind);
|
|
}
|
|
if (outcome.kind === "TRANSPORT_FAILURE") {
|
|
return String((outcome.failure as Readonly<{ kind: string }>).kind);
|
|
}
|
|
if (outcome.kind === "PROBLEM") {
|
|
return String((outcome.metadata as Readonly<{ status: number }>).status);
|
|
}
|
|
return null;
|
|
}
|
|
|
|
function retryReason(status: AttemptStatus): RetryReason {
|
|
if (status === "NETWORK_REJECTION") return "NETWORK_FAILURE";
|
|
if (status === 429) return "HTTP_429";
|
|
if (status === 503) return "HTTP_503";
|
|
throw new Error(`Sleep followed a non-retryable attempt: ${String(status)}`);
|
|
}
|
|
|
|
function bodyDisposition(
|
|
trace: AttemptTrace,
|
|
outcome: Readonly<Record<string, unknown>>,
|
|
): BodyDisposition {
|
|
if (trace.status === "NETWORK_REJECTION" || trace.status === "PENDING_ABORT") {
|
|
return "NO_RESPONSE";
|
|
}
|
|
if (
|
|
outcome.kind === "CONTRACT_VIOLATION" &&
|
|
outcomeDetail(outcome) === "RESPONSE_TOO_LARGE" &&
|
|
trace.cancelled
|
|
) {
|
|
return "REJECTED_LIMIT";
|
|
}
|
|
if (trace.completed) return "FULLY_READ_WITHIN_BOUND";
|
|
if (trace.cancelled) return "CANCELLED_WITHOUT_READ";
|
|
throw new Error(`Response body was neither consumed nor cancelled: ${trace.status}`);
|
|
}
|
|
|
|
async function waitForHandler(
|
|
ready: Promise<void>,
|
|
executionId: string,
|
|
): Promise<void> {
|
|
let watchdog: ReturnType<typeof setTimeout> | undefined;
|
|
try {
|
|
await Promise.race([
|
|
ready,
|
|
new Promise<never>((_resolve, reject) => {
|
|
watchdog = setTimeout(
|
|
() => reject(new Error(`Handler readiness timed out: ${executionId}`)),
|
|
2_000,
|
|
);
|
|
}),
|
|
]);
|
|
} finally {
|
|
if (watchdog !== undefined) clearTimeout(watchdog);
|
|
}
|
|
}
|
|
|
|
async function executeScenario(
|
|
entry: HttpScenarioExpectation,
|
|
): Promise<HttpScenarioAssertionGroups> {
|
|
const physicalAttempts: AttemptTrace[] = [];
|
|
const sleeps: RetryReason[] = [];
|
|
const observations: HttpExecutionObservation[] = [];
|
|
const caller = new AbortController();
|
|
const scopeLifetime = new AbortController();
|
|
let scopeCurrent = true;
|
|
let handlerReady!: () => void;
|
|
const ready = new Promise<void>((resolve) => {
|
|
handlerReady = resolve;
|
|
});
|
|
const scope: CacheScopeSnapshot = Object.freeze({
|
|
generation: 1,
|
|
fingerprint: "scenario-scope-1",
|
|
identities: Object.freeze({}) as CacheScopeSnapshot["identities"],
|
|
signal: scopeLifetime.signal,
|
|
isCurrent: () => scopeCurrent,
|
|
});
|
|
const baseOperation = operationFor(entry.operationId);
|
|
const operation =
|
|
entry.testDeadlineOverrideMs === null
|
|
? baseOperation
|
|
: Object.freeze({
|
|
...baseOperation,
|
|
frontend: Object.freeze({
|
|
...baseOperation.frontend,
|
|
totalDeadlineMs: entry.testDeadlineOverrideMs,
|
|
}),
|
|
});
|
|
const fetcher = tracingFetcher(physicalAttempts);
|
|
const executor = createContractHttpExecutor({
|
|
baseUrl: "https://api.test",
|
|
maxRetryAttempts: 2,
|
|
attachCredentials: () => ({ kind: "READY", headers: {} }),
|
|
fetcher,
|
|
random: () => 0,
|
|
readBoundedResponseBytes: async (response, maximumBytes) => {
|
|
const trace = physicalAttempts.at(-1);
|
|
if (!trace) throw new Error("Body reader ran before a physical attempt");
|
|
if (trace.appliedCeiling !== null) {
|
|
throw new Error("Body reader ran more than once for one attempt");
|
|
}
|
|
trace.appliedCeiling = maximumBytes;
|
|
return readBoundedBytes(response, maximumBytes);
|
|
},
|
|
sleep: async () => {
|
|
const prior = physicalAttempts.at(-1);
|
|
if (!prior) throw new Error("Retry sleep occurred before an attempt");
|
|
sleeps.push(retryReason(prior.status));
|
|
},
|
|
observe: (observation) => observations.push(observation),
|
|
});
|
|
|
|
mockApi.server.use(
|
|
...createTestEntityScenarioHandlers({
|
|
scenarios: { [entry.operationId]: entry.scenarioId },
|
|
onAttempt(attempt) {
|
|
if (
|
|
attempt.operationId === entry.operationId &&
|
|
attempt.scenarioId === entry.scenarioId
|
|
) {
|
|
handlerReady();
|
|
}
|
|
},
|
|
}),
|
|
);
|
|
|
|
try {
|
|
const execution = executor.execute(operation, inputFor(entry.operationId), {
|
|
routeId: ROUTE_ID,
|
|
scope,
|
|
signal: caller.signal,
|
|
...(entry.operationId === "TEST_CREATE_ENTITY"
|
|
? {
|
|
intent: Object.freeze({
|
|
intentId: `intent-${entry.scenarioId}`,
|
|
operationId: entry.operationId,
|
|
canonicalInputIdentity: "scenario-input",
|
|
idempotencyKey: `scenario-${entry.scenarioId}`,
|
|
createdAtMonotonicMs: 1,
|
|
}),
|
|
}
|
|
: {}),
|
|
});
|
|
|
|
if (entry.scenarioId === "timeout" || entry.scenarioId === "aborted") {
|
|
await waitForHandler(ready, entry.executionId);
|
|
if (entry.scenarioId === "aborted") {
|
|
if (entry.operationId === "TEST_LIST_ENTITIES") {
|
|
caller.abort();
|
|
} else {
|
|
scopeCurrent = false;
|
|
scopeLifetime.abort();
|
|
}
|
|
}
|
|
}
|
|
const outcome = (await execution) as unknown as Readonly<Record<string, unknown>>;
|
|
expect(observations, `${entry.executionId} observer count`).toHaveLength(1);
|
|
const observation = observations[0]!;
|
|
const observedSignal = scopeLifetime.signal.aborted ? "ABORTED" : "ACTIVE";
|
|
const cancellationOwner =
|
|
observation.terminalReason === "TIMEOUT"
|
|
? "DEADLINE"
|
|
: caller.signal.aborted
|
|
? "CALLER"
|
|
: scopeLifetime.signal.aborted
|
|
? "SCOPE_FENCE"
|
|
: "NONE";
|
|
const observed: HttpScenarioAssertionGroups = Object.freeze({
|
|
status: Object.freeze({
|
|
attempts: Object.freeze(physicalAttempts.map((attempt) => attempt.status)),
|
|
final: physicalAttempts.at(-1)!.status,
|
|
}),
|
|
outcome: Object.freeze({
|
|
kind: String(outcome.kind),
|
|
detail: outcomeDetail(outcome),
|
|
}),
|
|
effect: Object.freeze({
|
|
outcome: String(outcome.effect),
|
|
observer: observation.terminalReason,
|
|
}),
|
|
retry: Object.freeze({ count: sleeps.length, reasons: Object.freeze(sleeps) }),
|
|
fetch: Object.freeze({
|
|
count: physicalAttempts.length,
|
|
observerAttempts: observation.attemptCount,
|
|
agrees: physicalAttempts.length === observation.attemptCount,
|
|
}),
|
|
media: Object.freeze({
|
|
attempts: Object.freeze(physicalAttempts.map((attempt) => attempt.media)),
|
|
final: physicalAttempts.at(-1)?.media ?? null,
|
|
}),
|
|
body: Object.freeze({
|
|
attempts: Object.freeze(
|
|
physicalAttempts.map((attempt) =>
|
|
Object.freeze({
|
|
disposition: bodyDisposition(attempt, outcome),
|
|
pulledBytes: attempt.pulledBytes,
|
|
ceiling: attempt.appliedCeiling ?? 0,
|
|
}),
|
|
),
|
|
),
|
|
}),
|
|
scope: Object.freeze({
|
|
start: "CURRENT",
|
|
end: scopeCurrent ? "CURRENT" : "STALE",
|
|
signal: observedSignal,
|
|
cancellationOwner,
|
|
}),
|
|
});
|
|
return observed;
|
|
} finally {
|
|
caller.abort();
|
|
scopeLifetime.abort();
|
|
mockApi.reset();
|
|
}
|
|
}
|
|
|
|
describe("HTTP scenario catalog execution evidence", () => {
|
|
it("declares exactly the 57 behaviorally distinct operation scenarios", () => {
|
|
const executionIds = Object.entries(OPERATION_SCENARIO_CATALOG).flatMap(
|
|
([operationId, scenarioIds]) =>
|
|
scenarioIds.map((scenarioId) => `${operationId}::${scenarioId}`),
|
|
);
|
|
|
|
expect(OPERATION_SCENARIO_CATALOG.TEST_LIST_ENTITIES).toHaveLength(20);
|
|
expect(OPERATION_SCENARIO_CATALOG.TEST_GET_ENTITY).toHaveLength(19);
|
|
expect(OPERATION_SCENARIO_CATALOG.TEST_CREATE_ENTITY).toHaveLength(18);
|
|
expect(executionIds).toHaveLength(57);
|
|
expect(new Set(executionIds)).toHaveLength(57);
|
|
expect(executionIds).toEqual(HTTP_SCENARIO_EXECUTION_IDS);
|
|
});
|
|
|
|
it("executes every declared scenario and publishes one complete receipt", async () => {
|
|
await rm(RECEIPT_PATH, { force: true });
|
|
const rows: Array<Readonly<{
|
|
executionId: string;
|
|
expected: HttpScenarioAssertionGroups;
|
|
observed: HttpScenarioAssertionGroups;
|
|
testDeadlineOverrideMs: number | null;
|
|
}>> = [];
|
|
|
|
for (const entry of HTTP_SCENARIO_EXPECTATIONS) {
|
|
const observed = await executeScenario(entry);
|
|
expect(observed, entry.executionId).toEqual(entry.expected);
|
|
rows.push(
|
|
Object.freeze({
|
|
executionId: entry.executionId,
|
|
expected: entry.expected,
|
|
observed,
|
|
testDeadlineOverrideMs: entry.testDeadlineOverrideMs,
|
|
}),
|
|
);
|
|
}
|
|
|
|
const sortedRows = [...rows].sort((left, right) =>
|
|
left.executionId.localeCompare(right.executionId),
|
|
);
|
|
await mkdir(path.dirname(RECEIPT_PATH), { recursive: true });
|
|
await writeValidatedJsonArtifact({
|
|
path: RECEIPT_PATH,
|
|
schema: httpScenarioReceiptSchema,
|
|
value: {
|
|
schemaVersion: HTTP_SCENARIO_RECEIPT_SCHEMA_VERSION,
|
|
catalogDigest: computeHttpScenarioCatalogDigest(
|
|
HTTP_SCENARIO_RECEIPT_SCHEMA_VERSION,
|
|
HTTP_SCENARIO_EXPECTATIONS,
|
|
),
|
|
catalogTotal: 57,
|
|
executedIds: sortedRows.map((row) => row.executionId),
|
|
rows: sortedRows,
|
|
},
|
|
});
|
|
}, 30_000);
|
|
});
|
|
|
|
describe("HTTP scenario handler for an operation outside the catalog", () => {
|
|
// 16-02. A feature test passes its own installed contract; the catalog has
|
|
// no entry for it. The ceiling is larger than the catalog's, so the
|
|
// oversized body shows it was read from this descriptor.
|
|
const featureList = Object.freeze({
|
|
...TEST_LIST_HTTP_CONTRACT,
|
|
contract: Object.freeze({
|
|
...TEST_LIST_HTTP_CONTRACT.contract,
|
|
operationId: "FEATURE_LIST_ITEMS",
|
|
pathTemplate: "/api/feature-groups/{groupId}/items",
|
|
}),
|
|
frontend: Object.freeze({
|
|
...TEST_LIST_HTTP_CONTRACT.frontend,
|
|
responseByteLimit: 65_536,
|
|
}),
|
|
});
|
|
|
|
it("answers on the descriptor's route with the descriptor's ceiling", async () => {
|
|
const attempts: unknown[] = [];
|
|
mockApi.server.use(
|
|
createHttpScenarioHandler(featureList, {
|
|
scenario: "response-too-large",
|
|
payload: [],
|
|
onAttempt: (attempt) => attempts.push(attempt),
|
|
}),
|
|
);
|
|
try {
|
|
const response = await fetch("https://api.test/api/feature-groups/g-1/items");
|
|
expect(response.status).toBe(200);
|
|
expect((await response.arrayBuffer()).byteLength).toBe(65_537);
|
|
expect(attempts).toEqual([
|
|
{
|
|
operationId: "FEATURE_LIST_ITEMS",
|
|
scenarioId: "response-too-large",
|
|
attempt: 1,
|
|
},
|
|
]);
|
|
} finally {
|
|
mockApi.reset();
|
|
}
|
|
});
|
|
|
|
it("keeps the catalog's evidence to its own operations", () => {
|
|
expect(() =>
|
|
assertOperationScenario("FEATURE_LIST_ITEMS", "success"),
|
|
).toThrow("FEATURE_LIST_ITEMS is not a catalog operation");
|
|
});
|
|
});
|