Boxplot - is there a way to provide the metrics via a predefined query

[matt-darwin]

I'm looking at producing a Boxplot chart to demonstrate Firebolt's aggregating indexing capabilities for fast performance.

The query for this would look something like this:-

SELECT "ts_day" AS ts_day,
       max("price") AS max_price,
       min("price") AS min_price,
       MEDIAN("price") AS median_price,
       APPROX_PERCENTILE(price, 0.75) AS seventy_fith_percentile,
       APPROX_PERCENTILE(price, 0.25) AS twenth_fifth_percentil,
       COUNT(*) AS number_samples
FROM "public"."trades"
WHERE "ts_day" >= CAST('2023-04-17' AS PGDATE)
  AND "ts_day" < CAST('2023-04-24' AS PGDATE)
  AND "currencyPair" = 'BTC_USDT'
GROUP BY "ts_day"
ORDER BY "ts_day"

Which would then leverage an automatically maintained aggregating index built to calculate the metrics required for the chart automatically.

However, looking at the box plot chart, it seems to want raw data to be provided, and then perform the calculations itself. This will be problematic in my example as there are millions of rows being returned, thus seriously impacting the performance of the chart.

Is there a boxplot visualisation that allows you to just provide the required values?