On Mon, 2011-10-10 at 14:27 +0100, Thom Brown wrote:
> I don't know if this has already been discussed, but can you explain
> the following:
>
> postgres=# select '[1,8]'::int4range;
> int4range
> -----------
> [1,9)
> (1 row)
>
> It seems unintuitive to represent a discrete range using an exclusive
> upper bound. While I agree that the value itself is correct, it's
> representation looks odd. Is it necessary?
The "canonicalize" function (specified at type creation time) allows you
to specify the canonical output representation. So, I can change the
canonical form for discrete ranges to use '[]' notation if we think
that's more expected.
But then "int4range(1,8)" would still mean "int4range(1,8,'[)')" and
therefore '[1,7]'. I used to have a "default_flags" parameter that could
also be specified at type creation time that would control the default
third parameter (the parameter that controls inclusivity) of the
constructor. However, I removed the "default_flags" parameter because,
as Florian pointed out, it's better to have a consistent output from the
constructor.
I'm open to suggestions, including potentially bringing back
"default_flags".
Regards,Jeff Davis