Re: [PATCH v4] Add \warn to psql - Mailing list pgsql-hackers

From David G. Johnston
Subject Re: [PATCH v4] Add \warn to psql
Date
Msg-id CAKFQuwaZv9UtR2tF8bumLvVR08SxKaw-aNqCEYs2C-Sgv-mY1g@mail.gmail.com
Whole thread Raw
In response to Re: [PATCH v4] Add \warn to psql  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
On Mon, Jul 8, 2019 at 8:35 PM Bruce Momjian <bruce@momjian.us> wrote:
On Mon, Jul  8, 2019 at 11:29:00PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > On Fri, Jul  5, 2019 at 11:29:03PM +0200, David Fetter wrote:
> >>> I fixed that, but I'm wondering if we should back-patch that fix
> >>> or leave the back branches alone.
>
> >> +0.5 for back-patching.
>
> > Uh, if this was done in a major release I am thinking we have to mention
> > this as an incompatibility, which means we should probably not backpatch
> > it.
>
> How is "clearly doesn't match the documentation" not a bug?

Uh, it is a bug, but people might be expecting the existing behavior
without consulting the documentation, and we don't expect people to be
testing minor releases.

Anyway, it seems to be have been applied only to head so far.

I would leave it at that.  Won't Fix for released versions (neither code nor documentation) as we describe the intended usage so people do the right thing (which is highly likely anyway - though something like "\echo :content_to_echo -n" wouldn't surprise me) but those that learned through trial and error only experience a behavior change on a major release as they would expect.  This doesn't seem important enough to warrant breaking the general rule.  Though I'd give a +1 to v12; at least for me Beta is generally fair game.

David J.

pgsql-hackers by date:

Previous
From: Ryan Lambert
Date:
Subject: Re: [Proposal] Table-level Transparent Data Encryption (TDE) and KeyManagement Service (KMS)
Next
From: Prabhat Sahu
Date:
Subject: Re: tableam vs. TOAST