Question

What different?

RIGHT:

select distinct maker, price from product
inner join printer
on product.model = printer.model
where color = 'y' and price <= (select min(price) from printer where color = 'y')

WRONG:

select distinct maker, price from product
inner join printer
on product.model = printer.model
where color = 'y' and price <= all (select distinct price from printer where color = 'y')

I know using “min” is better in performance.
But could anyone explain what wrong and different in the result?

Tables structures:

Product(maker, model, type)
Printer(code, model, color, type, price)

neikel

Answer 1

ALL is a short form for expanding the comparison operator against ALL rows, and combining the conditions with AND.
ANY is a short form for expanding the comparison operator against ALL rows, and combining the conditions with OR.

Specifically,

price <= all (select distinct price from printer where color = 'y')

expands to, assuming the subquery returns 4 rows

price <= <price1> AND
price <= <price2> AND
price <= <price3> AND
price <= <price4>

When any of those resove to NULL, the result is false because NULL is not comparable using <=. MIN doesn’t have this problem (because MIN skips NULLs) except for the edge case where there is NO result row, in which case <= (select MIN..) can also give you an unexpected result.

Using ALL in to perform a test against a nullable column should almost always be qualified with a filter, e.g.

select distinct maker, price from product
inner join printer
on product.model = printer.model
where color = 'y' and price <=
      all (select distinct price from printer where color = 'y'
           where price is not null)