Rate limiters decide whether each incoming task should be accepted based on how many recent accepted tasks are still inside a fixed time window.

In this question, implement rateLimit(timestamps, maxAllowed, windowLen) for a simplified batch setting. The input is a sorted list of task timestamps, and you should return a boolean array where each entry says whether that task is accepted.

This question is intentionally interview-scoped:

• The timestamps are already sorted in non-decreasing order.
• Only previously accepted tasks count against the limit.
• Rejected tasks do not consume slots in the window.
• You do not need real-time behavior, retries, token buckets, or per-user limits.

Examples

rateLimit([1, 3, 5, 6, 8, 9, 10], 2, 5);
// [true, true, false, true, true, false, false]

rateLimit([1, 6, 7], 1, 5);
// [true, true, false]

In the second example, the task at time 6 is accepted because the accepted task at time 1 is exactly 5 units earlier, so it is outside the window.

rateLimit([2, 2, 2], 2, 3);
// [true, true, false]

Arguments

rateLimit(timestamps, maxAllowed, windowLen)

• timestamps (number[]): Sorted, non-decreasing integer timestamps for each task.
• maxAllowed (number): Maximum number of accepted tasks allowed inside the current sliding window.
• windowLen (number): Window length in the same time unit as timestamps.

Returns

Returns a boolean[] of the same length as timestamps.

Each result is:

• true if the task is accepted.
• false if the task is rejected because the sliding window already contains maxAllowed previously accepted tasks.

Notes

• A task at time t should count previously accepted tasks whose timestamps satisfy t - previous < windowLen.
• A task exactly windowLen units earlier does not count.
• Inputs are guaranteed valid.
• Do not mutate the input array.