timeSeriesGroupArray

Sorts time series by timestamp in ascending order.

timeSeriesGroupArray

Introduced in: v25.8.0

Sorts time series data by timestamp in ascending order.

This function is experimental, enable it by setting allow_experimental_ts_to_grid_aggregate_function=true.

Syntax

timeSeriesGroupArray(timestamp, value)

Arguments

• timestamp — Timestamp of the sample. DateTime or UInt32 or UInt64
• value — Value of the time series corresponding to the timestamp. (U)Int* or Float* or Decimal

Returned value

Returns an array of tuples (timestamp, value) sorted by timestamp in ascending order. If there are multiple values for the same timestamp then the function chooses the greatest of these values. Array(Tuple(T1, T2))

Examples

Basic usage with individual values

Query

WITH
    [110, 120, 130, 140, 140, 100]::Array(UInt32) AS timestamps,
    [1, 6, 8, 17, 19, 5]::Array(Float32) AS values
SELECT timeSeriesGroupArray(timestamp, value)
FROM
(
    SELECT
        arrayJoin(arrayZip(timestamps, values)) AS ts_and_val,
        ts_and_val.1 AS timestamp,
        ts_and_val.2 AS value
);

Response

┌─timeSeriesGroupArray(timestamp, value)───────────────┐
│ [(100, 5), (110, 1), (120, 6), (130, 8), (140, 19)]  │
└──────────────────────────────────────────────────────┘

Passing multiple samples of timestamps and values as arrays of equal size

Query

WITH
    [110, 120, 130, 140, 140, 100]::Array(UInt32) AS timestamps,
    [1, 6, 8, 17, 19, 5]::Array(Float32) AS values
SELECT timeSeriesGroupArray(timestamps, values);

Response

┌─timeSeriesGroupArray(timestamps, values)──────────────┐
│ [(100, 5), (110, 1), (120, 6), (130, 8), (140, 19)]   │
└───────────────────────────────────────────────────────┘