"Glen Parker" <glenebob@nwlink.com> writes:
> DB=# select round(1.25::float, 2);
> ERROR: function round(double precision, integer) does not exist
> What the heck? This can't be right... Can it?
regression=# \df round
List of functions
Result data type | Schema | Name | Argument data types
------------------+------------+-------+---------------------
double precision | pg_catalog | round | double precision
numeric | pg_catalog | round | numeric
numeric | pg_catalog | round | numeric, integer
(3 rows)
Looks right to me: the only 2-arg flavor of round() is on numeric,
not float8.
You could always cast the float to numeric, of course. I think in
7.2 such conversions were allowed silently, but we have reduced the
number of implicit type coercions.
regards, tom lane
