On Wed, Dec 16, 2015 at 12:06 PM, Andres Freund <andres@anarazel.de> wrote:
> On 2015-12-16 13:02:25 -0500, Tom Lane wrote:
>> I think the most reasonable way to handle this is to put the
>> call into a new function exported from input.c, where it can be
>> made conditional on useReadline.
>
> Sounds good.
I'll right, I'll follow up with that shortly.
For posterity, I think you guys arrived at the right conclusion, but
the difference between the two public screen reset calls is that one
(the original patch) forces a re-display of the prompt and the newer
one, rl_reset_screen_size(), does not. With the older patch the best
I could come up with was to call rl_crlf() so that the prompt display
did not amend to the end of \timing's output. However, that still
modified the output relative to what it was pre-patch, which would be
impolite to our users in a backpatch scenario.
merlin