Re: Changed SRF in targetlist handling

On 2016-08-17 17:41:28 -0700, Andres Freund wrote:
> Tom, do you think this is roughly going in the right direction? My plan
> here is to develop two patches, to come before this:
>
> a) Allow to avoid using a tuplestore for SRF_PERCALL SRFs in ROWS FROM -
> otherwise our performance would regress noticeably in some cases.
> b) Allow ROWS FROM() to return SETOF RECORD type SRFs as one column,
> instead of expanded. That's important to be able move SETOF RECORD
> returning functions in the targetlist into ROWS FROM, which otherwise
> requires an explicit column list.

I'm working on these. Atm ExecMakeTableFunctionResult() resides in
execQual.c - I'm inlining it into nodeFunctionscan.c now, because
there's no other callers, and having it separate seems to bring no
benefit.

Please speak soon up if you disagree.

Andres