Le 18/10/2021 à 17:24, Vik Fearing a écrit : > On 10/18/21 8:44 AM, Gilles Darold wrote: >> Le 17/10/2021 à 23:48, Isaac Morland a écrit : >>> On Sun, 17 Oct 2021 at 17:42, Gilles Darold <gil...@migops.com >>> <mailto:gil...@migops.com>> wrote: >>> >>> Note that psql doesn't display a separate line for each row in this >>> case, but the actual result coming back from the server does contain >>> the appropriate number of rows. >> I was not aware of that. In this case perhaps that we can remove the >> restriction on having at least on expandable column and we will have the >> same behavior but I can't think of an interest to allow that. > Allowing no-column tables removed the need to handle a bunch of corner > cases. Useful for users or not, the precedent is set.
I agree, now that I know that this is perfectly possible to return N rows without any data/column I also think that we should allow it in respect to PostgreSQL behavior with a table with no column. I will remove the check at SET UNEXPANDED. -- Gilles Darold
