esc key

[bulia byak]

> > where 0 means "stick forever", 1 means "stick for some time" (controlled
> > by esc_delay) and 2 means "do not stick".
> 
> I have no idea what it means.  We had a problem with mc+xterm+ncurses
> recently when escape would not be recognized by ncurses under heavy load
> (i.e. you are entering a tar.bz2 archive).  The fix was to set ncurses'
> internal delay to a reasonable non-zero value.  I believe that we would
> have a lot of weird problems if the escape delay was exactly 0.

Of course to implement "do not stick", we don't need exactly 0. A few ms would be fine.

> I think we can live with one variable esc_delay, and 0 (or maybe -1) would
> mean "wait for the next escape forever".

This is possible too, but my proposal has the advantage of maintaining backward compatibility: we can alias "old_esc_mode" to be the same as the "esc_mode" and then the old ini files with either old_esc_mode=1 or old_esc_mode=0 would continue to work as before. Only if you use the new extended value of 2 you'll get a new behavior. Do you think that makes sense?

> I don't think so.  It's not uncommon for people to use different terminals
> with different speed.  I'm not saying that we necessarily need a GUI.
> Maybe we should finally implement terminal-specific settings?

Perhaps, although I personally do not use different terminals so I don't feel this as a priority. IMHO implementing an external keymap configuration is more important at this point.

> I think it's a huge annoyance when the command line is erased
> unexpectedly.  I'm used to running "Find File" by pressing Escape-?, so
> I'm losing the command line every time I attempt to do it in Far.

This is why I proposed to enable this binding _only_ for esc_mode=2, i.e. for those who want no delay whatsoever and never use esc as a prefix key. 


-- 
__________________________________________________________
Sign-up for your own FREE Personalized E-mail at Mail.com
http://www.mail.com/?sr=signup

Meet Singles
http://corp.mail.com/lavalife