Re: Changed SRF in targetlist handling

David Fetter <david(at)fetter(dot)org> writes:
> On Mon, May 23, 2016 at 01:36:57PM -0400, Tom Lane wrote:
>> David Fetter <david(at)fetter(dot)org> writes:
>>> How about making "TABLE generate_series(1,n)" work? It's even
>>> shorter in exchange for some cognitive load.

>> No thanks --- the word after TABLE ought to be a table name, not some
>> arbitrary expression. That's way too much mess to save one keystroke.

> It's not just about saving a keystroke. This change would go with
> removing the ability to do SRFs in the target list of a SELECT
> query.

I guess you did not understand that I was rejecting doing that.
Telling people they have to modify existing code that does this and
works fine is exactly what I felt we can't do. We might be able to
blow off complicated cases, but I think simpler cases are too common
in the field.

I'm on board with fixing things so that the *implementation* doesn't
support SRF-in-tlist. But we can't just remove it from the language.

regards, tom lane