Re: pgsql: Fix memory leak in PLySequence_ToJsonbValue()

On Fri, Jun 15, 2018 at 11:01:26AM -0400, Tom Lane wrote:
> David Fetter <david(at)fetter(dot)org> writes:
>> Perhaps we could acquire the short.pg domain and run it off that.
>
> We're already using postgr.es, so it's not going to get much shorter
> just from mucking with the domain part. We'd need to set up some
> abbreviation for the message-ID part, which I'm not for. It would
> add more overhead to writing commit log entries (to create or find
> out the right abbreviation), and it would make it harder to go back
> and forth between the commit log and local mail archives, and it'd
> be yet another Thing We Have To Keep Running Forever.

I am also -1 for any kind of message-id shorteners in the commit
messages. I cannot say for others, but sometimes I work offline with
only the git history and a cold copy of Postgres archives, meaning that
depending on an external source to get the raw message IDs would break
that.
--
Michael