Apache DataFusion SQL Query Engine
CREATE TABLE grouped_distinct AS
SELECT value % 2000 AS g,
value % 8 AS few_g,
(value * 48271) % 999983 AS x,
(value / 2000) % 8 AS low_x,
arrow_cast((value * 48271) % 999983, 'UInt64') AS unsigned_x,
CAST((value * 48271) % 999983 AS VARCHAR) AS string_x,
CAST((value * 48271) % 999983 AS DOUBLE) AS float_x,
CASE WHEN value % 100 < 90 THEN 0 ELSE value % 8 END AS skew_g,
CAST(CASE WHEN value % 100 < 90 THEN 0 ELSE value % 8 END AS VARCHAR) AS string_g,
value % 1000000 AS many_g,
value % 2 AS dup_x
FROM range(${GROUPED_DISTINCT_ROWS:-4000000});