Re: fn timestampdiff gives error...

[Lance J. Andersen]

Daniel John Debrunner wrote: Lance J. Andersen wrote: JDBC/ODBC define that the return type range for the TIMESTAMPDIFF function is an INTEGER regardless of what the backend engine supports. Just to be picky, do you actually see 'INTEGER' anywhere, or just 'integer'? Interesting question which sadly is not clearly articulated anywhere so i am afraid milage may vary.  Guess I will clarify this for JDBC 4 I would see SQL BIGINT, Types.BIGINT and long as all of type integer. Because if 'integer' means only SQL INTEGER or Types.INTEGER, then the numeric functions floor and mod should not work against BIGINT values. I'd always assummed that these types in appendix C of JDBC 3.0 meant: integer - any integral type float - any imprecise type number - any number type Dan.