Typo in psql doc

On Tue, Apr 13, 2021 at 07:57:54AM +0000, PG Doc comments form wrote:

There is a spurious space inside the documentation:
\set HISTFILE ~/.psql_history- :DBNAME
instead of
\set HISTFILE ~/.psql_history-:DBNAME

Both commands sey in your .psqlrc results in the same path being
used, as I guess that psqlscanslash.l eats all the whitespaces
in-between. So the documentation is not wrong here (see also commit
e4c7619).
--
Michael

Michael Paquier <michael@paquier.xyz> writes:

On Tue, Apr 13, 2021 at 07:57:54AM +0000, PG Doc comments form wrote:

There is a spurious space inside the documentation:
\set HISTFILE ~/.psql_history- :DBNAME
instead of
\set HISTFILE ~/.psql_history-:DBNAME

Both commands sey in your .psqlrc results in the same path being
used, as I guess that psqlscanslash.l eats all the whitespaces
in-between. So the documentation is not wrong here (see also commit
e4c7619).

I think that e4c7619 may have been working around a since-fixed
limitation in variable expansion. [experiments ...] Ah, looks like
this behavior changed in 9.2, which is later than I would've guessed:

psql (9.1.24)
Type "help" for help.

regression=# \set FOO bar-:DBNAME
regression=# \echo :FOO
bar-:DBNAME

psql (9.2.24)
Type "help" for help.

regression=# \set FOO bar-:DBNAME
regression=# \echo :FOO
bar-regression

As you say, both ways now give the same result. Since it's not the
point of this example to illustrate \set's space-eating behavior,
it might be clearer to revert the addition of the space.

regards, tom lane

On Wed, Apr 14, 2021 at 10:12:02AM -0400, Tom Lane wrote:

As you say, both ways now give the same result. Since it's not the
point of this example to illustrate \set's space-eating behavior,
it might be clearer to revert the addition of the space.

Oh, interesting point. I did not notice that this was different
before.
--
Michael

Yes indeed, thanks. I re-tested the example this morning and it worked
correctly with the space. I guess I had messed up things when I first
tried it.

On Thu, Apr 15, 2021 at 09:55:53AM +0200, Ludovic Kuty wrote:

Yes indeed, thanks. I re-tested the example this morning and it worked
correctly with the space. I guess I had messed up things when I first tried
it.

After seeing Tom's argument that this behavior become more consistent
in 9.2, and that e4c7619 introduced it initially because of what ~9.1
was doing, I have considered both points and just applied a small
patch for HEAD with 1840d9f to remove the space.
--
Michael