Hi Christian, > I'm not sure of other uses (incremental search) of rl_get_previous_history() > but in this case it feels like _rl_maybe_replace_line() should be called > after > _rl_previous_history_internal() has loaded the history entry?
Just want to report that this change also fixes the issue for me. Thanks! I use history-search-backward with PgUp in pretty much the same way (gentoo default for 20+ years), and I think the new behavior in readline-8.3 should be considered as a regression. Can the fix be included in the next official patch? Thanks Regards, Yap
