Fix interval test, which was broken for floating-point timestamps.
authorJeff Davis <jdavis@postgresql.org>
Wed, 7 May 2014 02:35:24 +0000 (19:35 -0700)
committerJeff Davis <jdavis@postgresql.org>
Wed, 7 May 2014 02:53:59 +0000 (19:53 -0700)
commit348aa75a678069569923c501206d5a1ca03654e3
tree61589c4a2a13fce9adfe44ed77eea711995976c4
parent2c22afaa4e29cbd773bae3f043a941f82ff30e2d
Fix interval test, which was broken for floating-point timestamps.

Commit 4318daecc959886d001a6e79c6ea853e8b1dfb4b introduced a test that
couldn't be made consistent between integer and floating-point
timestamps.

It was designed to test the longest possible interval output length,
so removing four zeros from the number of hours, as this patch does,
is not ideal. But the test still has some utility for its original
purpose, and there aren't a lot of other good options.

Noah Misch suggested a different approach where we test that the
output either matches what we expect from integer timestamps or what
we expect from floating-point timestamps. That seemed to obscure an
otherwise simple test, however.

Reviewed by Tom Lane and Noah Misch.
src/test/regress/expected/interval.out
src/test/regress/sql/interval.sql