/**
 * Split a Unix epoch interval at arbitrary `points`, producing consecutive sub-intervals.
 *
 * - `points` need not be sorted — they are sorted internally before splitting.
 * - Points outside `[start, end]` are dropped; they cannot introduce a boundary that isn't
 *   inside the interval.
 * - Points exactly on `start` or `end` are dropped too — they would only produce a
 *   zero-length sub-interval at the edge.
 * - Duplicate points collapse to a single boundary.
 * - Returns `[{ start, end }]` (the whole interval, unsplit) when no valid in-range point remains.
 * - Returns `[]` when `points` is not an array, when any element is not a finite number (or
 *   numeric string), or on invalid input (non-finite/non-integer start/end, `start > end`).
 *
 * @param start Unix epoch value (seconds or milliseconds) — interval start
 * @param end Unix epoch value (seconds or milliseconds) — interval end
 * @param points array of Unix epoch values to split at
 * @returns array of `{ start, end }` records, or `[]` on invalid input
 *
 * @example intervalSplitAtUnix(0, 100000, [50000]) // [{ start: 0, end: 50000 }, { start: 50000, end: 100000 }]
 * @example intervalSplitAtUnix(0, 100000, []) // [{ start: 0, end: 100000 }]
 * @example intervalSplitAtUnix(NaN, 100000, [50000]) // []
 */
export declare function intervalSplitAtUnix(start: number | string, end: number | string, points: Array<number | string>): Array<{
    start: number;
    end: number;
}>;