On Mon, Jan 3, 2022 at 1:14 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Hmm, I think you went a bit too far here. The existing code intends
> to draw a distinction between "not recognized" (i.e., "we don't know
> what that word was you used") and "not supported" (i.e., "we know
> that word, but it doesn't seem to make sense in context, or we
> haven't got round to the case yet"). You've mashed those into the
> same error text, which I don't think we should do, especially
> since we're using distinct ERRCODE values for them.
Oops. I noticed that "inconsistency" between
ERRCODE_FEATURE_NOT_SUPPORTED and ERRCODE_INVALID_PARAMETER_VALUE and
then promptly blazed past it. Thanks for catching that.
> Attached v3 restores that distinction, and makes some other small
> tweaks. (I found that there were actually a couple of spots in
> date.c that got it backwards, so admittedly this is a fine point
> that not everybody is on board with. But let's make it consistent
> now.)
LGTM too, for whatever that's worth.