Re: Reduce WAL logging of INSERT SELECT

On 08/04/2011 04:55 PM, Tom Lane wrote:
> Bruce Momjian<bruce@momjian.us>  writes:
>> One thing we don't optimize is INSERT ... SELECT when the table is
>> created or truncated in the same transaction.  Seems we could.
>> We optimize CREATE TABLE AS which is effectively SELECT ... INTO using a
>> different syntax.  Is this a TODO?
> Considering that SELECT INTO is deprecated, I don't think we should be
> expending effort to encourage people to use it.
>
>             


Right, but the original point about INSERT ... SELECT seems reasonable, no?

cheers

andrew