Value Distributions (Histograms)

Real data is rarely uniform. In the orders table most orders are paid, and only a few are returned. A query planner needs to know this distribution (a histogram of values) to judge how selective a filter is: status = 'returned' matches few rows, status = 'paid' matches most.

Write a query returning, for every status, the number of orders as row_count and the share of all orders as pct (a percentage rounded to 1 decimal place). Order by row_count descending, then by status.

Constraints

• Columns, in order: status, row_count, pct
• pct = 100 × row_count ÷ total orders, rounded to 1 decimal
• Order by row_count descending, then status

Hints