On 2020-12-17 02:30, Michael Paquier wrote:
> On Wed, Dec 16, 2020 at 06:07:08PM +0100, Peter Eisentraut wrote:
>> Right, we would very likely not add it now. But it doesn't seem to cause a
>> lot of ongoing maintenance burden, so if there is a use case, it's not
>> unreasonable to keep it around. I have no other motive here, I was just
>> curious.
>
> From the point of view of the code, this would simplify the grammar
> rules of SELECT, which is a good thing. And this would also simplify
> some code in the rewriter where we create some CreateTableAsStmt
> on-the-fly where an IntoClause is specified, but my take is that this
> is not really a burden when it comes to long-term maintenance. And
> the git history tells, at least it seems to me so, that this is not
> manipulated much.
I have committed my small documentation patch that points out
compatibility with other implementations.
