On Wed, 2022-09-28 at 16:31 +0000, Tom Lane wrote:
> Doc: document bpchar, clarify relationship of text and varchar.
BPCHAR is documented as an alias to CHARACTER, but the former can end
up with no typmod at all whereas the latter defaults to CHARACTER(1).
Should we document the case of using BPCHAR with no length?
I noticed this because there exists a hash function bug[1] that seems
to be visible with BPCHAR but not CHARACTER.
[1]
https://www.postgresql.org/message-id/7692740d4736e79032a5dac689cf2e304c03fa78.camel@j-davis.com
--
Jeff Davis
PostgreSQL Contributor Team - AWS