Horton

Re: Unexpected interval comparison

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
Cc: adrian.klaver@aklaver.com, frazer@frazermclean.co.uk, pgsql-general@postgresql.org
Date: 2017-04-03T15:35:25Z
Lists: pgsql-general 
Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp> writes:
> Ok, the attached patch changes the result type of
> interval_cmp_value from TimeOffset(=int64) to new 128 bit
> LinearInterval. The value is hidden under the functions
> interval_eq/ge.../cmp and all other stuff seems to use the
> functions.

Looking at this now ... why isn't the INT64_AU32 macro just

#define INT64_AU32(i64) ((i64) >> 32)

?  The business with subtracting and re-adding 1 seems unnecessary, and it
also creates a risk of overflow with the minimum possible int64 value.

			regards, tom lane

Commits

  1. Fix integer-overflow problems in interval comparison.