Re: fix for readline terminal size problems when window is resized with open pager

On 2015-12-16 12:02:26 -0500, Tom Lane wrote:
> [ squint... ] What readline version are you using, and do you have
> LINES/COLUMNS set in your terminal environment?

libreadline-dev:
Installed: 6.3-8+b4

Both are set - I think bash does that.

> It may be that the echo stuff is not good, but I'm pretty dubious
> about ignore_env being the issue.

Indeed it's not, I'm not entirely sure whether I managed to repeatedly
run with the wrong version of psql (doubtful, same terminal), or what
strange state I observed.

But
_rl_get_screen_size (fileno (rl_instream), 0);
if (RL_ISSTATE(RL_STATE_REDISPLAYING) == 0)
_rl_redisplay_after_sigwinch ();
in ClosePager() shows the problem independent of ignore_env or not.

I think the difference here is actually that redrawing shows the query
after the pager, without it we don't show it again.

If you configure less to automatically quit if the query is below one
screen (-F) and use \pset pager always the difference is clearly
visible:

resize (just _rl_get_screen_size()):

postgres[12561][1]=# SELECT 1;
┌──────────┐
│ ?column? │
├──────────┤
│ 1 │
└──────────┘
(1 row)

Time: 0.797 ms
postgres[12561][1]=#

resize & redraw (_rl_get_screen_size() & _rl_redisplay_after_sigwinch):

postgres[12393][1]=# SELECT 1;
┌──────────┐
│ ?column? │
├──────────┤
│ 1 │
└──────────┘
(1 row)

postgres[12393][1]=# SELECT 1;Time: 0.555 ms

based on that I'm inclined to just go with resize - redisplaying the
query after the pager might be desirable, but I think it's an actual
behavioural change.