Pattern

A histogram in SQL

Count values per bin of a fixed width, keeping the empty bins.

A value's bin is FLOOR(x / width) * width. Count per bin, then LEFT JOIN the counts onto every bin from generate_series(0, last_bin, width) and COALESCE to 0: without the series, an empty bin disappears and the shape lies. Choosing the width is the analysis: too wide hides the shape, too narrow shows noise.

WITH baskets AS (
  SELECT o.order_id, SUM(l.quantity * l.unit_price) AS total
  FROM orders o JOIN order_lines l ON l.order_id = o.order_id
  GROUP BY o.order_id
)
SELECT CAST(FLOOR(total / 100) * 100 AS INTEGER) AS bin_start,
       COUNT(*) AS orders,
       repeat('█', CAST(FLOOR(COUNT(*) / 50) AS INTEGER)) AS bar
FROM baskets
GROUP BY 1
ORDER BY 1;

bin_start	orders	bar
0	2011	████████████████████████████████████████
100	420	████████
200	45
300	17
400	6
500	1

Try it in Randori

Practise:A histogram in SQL