Horton

Re: New "single-call SRF" APIs are very confusingly named

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Melanie Plageman <melanieplageman@gmail.com>
Cc: Andres Freund <andres@anarazel.de>, pgsql-hackers@postgresql.org, Michael Paquier <michael.paquier@gmail.com>
Date: 2022-10-14T21:32:50Z
Lists: pgsql-hackers 
Melanie Plageman <melanieplageman@gmail.com> writes:
> So, while I agree that the "Single" in SetSingleFuncCall() could be
> confusing given the name of ExprSingleResult, I feel like actually all
> of the names are somewhat wrong.

Maybe, but ExprSingleResult et al. have been there for decades and
are certainly embedded in a ton of third-party code.  It's a bit
late to rename them, whether you think they're confusing or not.
Maybe we can get away with changing names introduced in v15, but
even that I'm afraid will get some pushback.

Having said that, I'd probably have used names based on "materialize"
not "single" for what this code is doing.

			regards, tom lane

Commits

  1. Remove compatibility declarations for InitMaterializedSRF()

  2. Rename SetSingleFuncCall() to InitMaterializedSRF()

  3. Create routine able to set single-call SRFs for Materialize mode