On Thu, Oct 26, 2023 at 11:04:47PM -0400, Bruce Momjian wrote:
> On Thu, Oct 26, 2023 at 04:17:19PM -0700, David G. Johnston wrote:
> > On Thu, Oct 26, 2023 at 4:13 PM David G. Johnston <david.g.johnston@gmail.com>
> > wrote:
> >
> > On Thu, Oct 26, 2023 at 4:08 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> >
> > Bruce Momjian <bruce@momjian.us> writes:
> > > Ah, I was confused. I documented both in the attached patch.
> >
> > The function one should have the same annotation as some others:
> >
> > <entry>can be increased by recompiling <productname>PostgreSQL</
> > productname></entry>
> >
> >
> >
> > I'd like to see a comment on the parameter count one too.
> >
> > "Alternatives include using a temporary table or passing them in as a
> > single array parameter."
> >
> > About the only time this is likely to come up is with many parameters of
> > the same type and meaning, pointing that out with the array option seems
> > excessively wordy for the comment area.
> >
> > Needs a comma: 65,535
> >
> > Kinda think both should be tacked on to the end of the table. I'd also put
> > function arguments first so it appears under the compile time partition
> > keys limit.
> >
> >
> >
> > Cleanups for consistency:
> >
> > Move "identifier length" after "partition keys" (before the new "function
> > arguments")
> >
> > Add commas to: 1,600 and 1,664 and 8,192
>
> Okay, I made all the suggested changes in ordering and adding commas,
> plus the text about the ability to change function arguments via
> recompiling.
>
> I didn't put commas in 8192 since that is a power-of-two and kind of a
> magic number used in many places.
>
> I am not sure where to put text about using arrays to handle many
> function arguments. I just don't see it fitting in the table, or the
> paragraph below the table.
Patch applied back to Postgres 12.
--
Bruce Momjian <bruce@momjian.us> https://momjian.us
EDB https://enterprisedb.com
Only you can decide what is important to you.