jian he <jian.universal...@gmail.com> writes: > On Thu, Feb 8, 2024 at 1:17 PM Tom Lane <t...@sss.pgh.pa.us> wrote: >> >> "Jonathan S. Katz" <jk...@postgresql.org> writes: >> > On 2/6/24 3:19 AM, jian he wrote: >> >> On Tue, Feb 6, 2024 at 12:43 PM Jonathan S. Katz <jk...@postgresql.org> >> >> wrote: >> >>> * In PL/pgSQL, support SQL commands that are `CREATE FUNCTION`/`CREATE >> >>> PROCEDURE` with SQL-standard bodies. >> >> >> https://git.postgresql.org/cgit/postgresql.git/commit/?id=57b440ec115f57ff6e6a3f0df26063822e3123d2 >> >> says only for plpgsql routine or DO block. >> >> > I had some trouble wordsmithing this, but the commit title is pretty >> > clear to me so I opted for that. >> >> Your text seems fine to me. I'm confused about the objection here: >> exactly what uses of plpgsql aren't in a routine or DO block? >> >> regards, tom lane > > I guess I was confused with cases like this > ` > create function test1() returns int language plpgsql > begin atomic > select unique1 from tenk1 limit 1; > end ; > ` > looking back, the original text seems fine.
The word "routine" is the SQL standard's common term for both functions and procedures. - ilmari
