I just realized that Drill now supports a new construct that we hadn't previously tried out. We can use the VALUES operator in the FROM clause thereby generating our own data. Nice that Calcite already gave this to us when we fixed the large in clause problem:
Example: SELECT * FROM (VALUES (1, 1.0, DATE '2008-2-23', TIME '12:23:34', TIMESTAMP '2008-2-23 12:23:34.456', INTERVAL '1' YEAR, INTERVAL '2' DAY), (1, 1.0, DATE '2008-2-23', TIME '12:23:34', TIMESTAMP '2008-2-23 12:23:34.456', INTERVAL '1' YEAR, INTERVAL '2' DAY) ) AS Example(myInt, myFloat, myDate, myTime, myTimestamp, int1, int2); +--------+----------+-------------+-----------+--------------------------+-------+-------+ | myInt | myFloat | myDate | myTime | myTimestamp | int1 | int2 | +--------+----------+-------------+-----------+--------------------------+-------+-------+ | 1 | 1.0 | 2008-02-23 | 04:23:34 | 2008-02-23 04:23:34.456 | P12M | P2D | | 1 | 1.0 | 2008-02-23 | 04:23:34 | 2008-02-23 04:23:34.456 | P12M | P2D | +--------+----------+-------------+-----------+--------------------------+-------+-------+
