sailfork / t3code-github / .repos/ alchemy-effect/ packages/ aws/ src/ ApplicationAutoScaling/ internal.ts

import type * as aas from "@distilled.cloud/aws/application-auto-scaling";
import * as Effect from "effect/Effect";
import * as Schedule from "effect/Schedule";

/**
 * Known Application Auto Scaling service namespaces, used by `list()` to
 * enumerate account-wide state. `DescribeScalableTargets` (and the policy /
 * scheduled-action describes) require a `ServiceNamespace` filter, so an
 * account-wide enumeration is the union of the per-namespace enumerations.
 * The wire type is open (`string & {}`) but AWS only ever returns members of
 * this set; new namespaces are appended when AWS ships them.
 */
export const SERVICE_NAMESPACES: readonly aas.ServiceNamespace[] = [
  "ecs",
  "elasticmapreduce",
  "ec2",
  "appstream",
  "dynamodb",
  "rds",
  "sagemaker",
  "custom-resource",
  "comprehend",
  "lambda",
  "cassandra",
  "kafka",
  "elasticache",
  "neptune",
  "workspaces",
];

/**
 * `RegisterScalableTarget` is eventually consistent: a `PutScalingPolicy` /
 * `PutScheduledAction` issued immediately after registration can fail with
 * `ObjectNotFoundException` until the target propagates. Retry that typed tag
 * on a short bounded schedule (~20s total).
 *
 * NOTE: the explicit `Effect.Effect<A, E, R>` return annotation is
 * load-bearing — an inlined `Effect.retry` leaks `Retry.Return`'s conditional
 * type into declaration emit and widens the provider layer to
 * `Layer<..., never, unknown>` for downstream consumers.
 */
export const retryWhileTargetPropagates = <A, E extends { _tag: string }, R>(
  self: Effect.Effect<A, E, R>,
): Effect.Effect<A, E, R> =>
  Effect.retry(self, {
    while: (e) => e._tag === "ObjectNotFoundException",
    schedule: Schedule.max([Schedule.fixed("2 seconds"), Schedule.recurs(10)]),
  });