Horton

Re: ecpg command does not warn COPY ... FROM STDIN;

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Fujii Masao <masao.fujii@oss.nttdata.com>
Cc: Ryo Kanbayashi <kanbayashi.dev@gmail.com>, "Hayato Kuroda (Fujitsu)" <kuroda.hayato@fujitsu.com>, pgsql-hackers@lists.postgresql.org
Date: 2025-01-08T15:42:09Z
Lists: pgsql-hackers 
Fujii Masao <masao.fujii@oss.nttdata.com> writes:
> On 2025/01/08 23:04, Ryo Kanbayashi wrote:
> But it is not working.
> ecpg command fails to notice though code like above exits on pgc code.

> This issue seems to have been introduced in commit 3d009e45bd.

Indeed :-(

> As for COPY FROM STDOUT, while it's possible, mentioning it might be
> confusing since it's not official syntax. So I'm not sure if this is
> good idea.

There's another problem: the correct syntax is COPY TO STDOUT,
and that won't trigger this warning either.

I'm inclined to drop the test on @5 altogether, and just check for
"stdin" or "stdout" in @7.  There is no variant of those that will
work.  (But maybe we should allow it if opt_program isn't empty?)

The warning message could perhaps be written
"COPY FROM STDIN/TO STDOUT is not implemented".

			regards, tom lane

Commits

  1. ecpg: Restore detection of unsupported COPY FROM STDIN.

  2. plpgsql: pure parser and reentrant scanner

  3. Add support for piping COPY to/from an external program.