Disable ncurses extended key values so that esc-codes can be used in …#1049
Disable ncurses extended key values so that esc-codes can be used in …#1049mckellygit wants to merge 1 commit intojonas:masterfrom
Conversation
…mappings for Shift/Ctrl/Alt Up/Dn/Lt/Rt/Hm/En/In/De/PgUp/PgDn
This makes it possible to use esc-codes in mappings for Shift/Ctrl/Alt Up/Down/Left/Right/Home/End/Insert/Delete/PgUp/PgDn. [tk: minor tweaks and NCURSES_EXT_FUNCS was only made visible in ncurses v5_6_20071117, so use NCURSES_VERSION instead for compatibility.] References #1046
|
Merged manually in aee8eb1. |
|
@koutcher thx. Something is weird tho, in my original PR a binding such as: worked fine - but now it does not, and I need to use: I will dig into it and post what I find. |
|
@koutcher its a problem with my other commit that was merged and the values for Esc and C-_ |
|
ok, PR added: #1058 |
|
Good catch. I should have spotted that during the review. I was so focused on the possible side-effect of changing the < in <= that I was happy enough when I checked that CTRL('z') was handled separately and I completely overlooked that when going from 25 to 30, 27 was also invited to the party. Hopefully 28 is only ^\. |
|
@koutcher @jonas just an fyi - I was working with xaizek on vifm for this same issue and he suggested a cleaner way to solve this. just BEFORE initscr() is called. I just thought I'd mention it. If you want I can create a PR to remove the keyok() loop and use this instead ? thx, |
|
It is much cleaner indeed. Why 20081102 ? |
|
Ok sure, NCURSES_VERSION is good - I just used an old version thinking it was old enough. |
|
Created #1074 |
…mappings for Shift/Ctrl/Alt Up/Dn/Lt/Rt/Hm/En/In/De/PgUp/PgDn
This is the other way to go, as compared to PR #1048
thx,
-m