summaryrefslogtreecommitdiffstats
path: root/main.c
diff options
context:
space:
mode:
authorKevin McCarthy <kevin@8t8.us>2021-07-22 15:44:42 -0700
committerKevin McCarthy <kevin@8t8.us>2021-07-24 17:36:21 -0700
commitf4ff768c2764502a974042e1fe9d861a7a92fbcf (patch)
treed94b884f09266a5281337deeef611c37a8f30908 /main.c
parente0232be752902950e379b8bb4bd84699d3781bb5 (diff)
Enable nonl() mode for ncurses. closes #362.
This patch is based on the patch from Mike Frysinger, archived from the old trac system at: https://gitlab.com/muttmua/trac-tickets/-/blob/master/attachments/3494/mutt-enter-return-hg-6658.patch This version creates a new key symbol, <KeypadEnter> for the KEY_ENTER value returned by getch(). According to Vincent Lefèvre's research, that corresponds to the "key pad" Enter key. If KEY_ENTER is not defined, it binds the symbol to "\n". It does this after the KeyNames[] definitions of <Enter> and <Return> so they remain the default shown by <what-key>. Since Mutt already accepts KEY_ENTER at the y/n prompt, add it for all the built-in operations too. Bindings with "\r" and "\n" in the binding_t structures don't need "<enter>" and "<return>" to be re-specified in km_init(), so remove those. Instead add the "<KeypadEnter>" bindings there. Remove unnecessary Alias menu bindings for OP_GENERIC_SELECT_ENTRY in km_init(). As a note: I don't believe non-extended key symbol bindings need to be bound inside km_init(). I'm not clear why this was done, or what purpose it serves. I'll do a bit more research, but if I can't uncover any issues, will move them inside the binding_t structures in another commit.
Diffstat (limited to 'main.c')
-rw-r--r--main.c1
1 files changed, 1 insertions, 0 deletions
diff --git a/main.c b/main.c
index 80a1d593..3fd930b6 100644
--- a/main.c
+++ b/main.c
@@ -604,6 +604,7 @@ static void start_curses (void)
keypad (stdscr, TRUE);
cbreak ();
noecho ();
+ nonl ();
#if HAVE_TYPEAHEAD
typeahead (-1); /* simulate smooth scrolling */
#endif