Horton

Re: Bug fix for psql's meta-command \ev

Ryoga Yoshida <bt23yoshidar@oss.nttdata.com>

View thread View on postgr.es
From: Ryoga Yoshida <bt23yoshidar@oss.nttdata.com>
To: Michael Paquier <michael@paquier.xyz>
Cc: Aleksander Alekseev <aleksander@timescale.com>, pgsql-hackers@lists.postgresql.org, Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Date: 2023-09-20T05:08:34Z
Lists: pgsql-hackers 
On 2023-09-20 09:32, Michael Paquier wrote:
> Actually there was a bit more to it in the presence of \e, that could
> also get some unpredictible behaviors if some errors happen while
> editing a query, which is something unlikely, still leads to strange
> behaviors on failure injections.  I was considering first to move the
> reset in do_edit(), but also we have the case of \e[v|f] where the
> buffer has no edits so it felt a bit more natural to do that in the
> upper layer like in this patch.

Indeed, similar behaviours can happen with the \e. The patch you 
committed looks good to me. Thank you.

Ryoga Yoshida

Commits

  1. psql: Reset query buffer of \e, \ef and \ev on error