Pedro Miguel Frazao Fernandes Ferreira <pfrazao@ualg.pt> writes:
> When I store a float8
> number in a database, I am supposed to fetch the number as inserted.
I fear you have a fundamental misconception about the nature of
floating-point representation. There is no such thing as "fetching the
number as inserted". You get only as much accuracy as float8 format
will store, which is about half a digit less than you are asking for in
this case.
The fact that you might be able to store this particular 16-digit value
exactly is no guarantee that you'll be able to store other 16-digit
values exactly, so I recommend choosing another representation if
that's your requirement. Blaming the output formatting is the wrong
way to look at it.
regards, tom lane
