Re: more questions

[Pavel Roskin <proski gnu org>]

> > > for example:
> > > alt-f7: \e\e[18~
> > > ctrl-insert: \e[2~
> >
> > I don't think mc could have written this.  Those keys are not redefinable
> > now.
>
> It did, at least for alt-f7. The ctrl-insert code, however, is the same
> as insert alone :( Can this be fixed somehow?  I don't dare to dream
> about the ability to distinguish between, say, plain insert and
> keypad-insert (as Emacs does) but what about just plain old ctrl-insert?

Please check key.c - you can read modifiers from X and from Linux console, 
so it's possible, but the necessary code doesn't exist yet.  I mean the 
code that would make this generic and easy to use.

Ctrl-Insert is not the same as Insert in rxvt-2.7.8, so it's not as simple
as reading the modifiers - the combination of the modifier state and the
keys should be considered as a whole to deal with this.  In other words, 
somewhere mc should have a table like this:

\e[2~ = insert
\e[2^ = ctrl-insert
Ctrl + \e[2~ = ctrl-insert
Ctrl + \e[2^ = ctrl-insert

Implementing this logic would take a lot of coding, of course.

Right now, modifiers are checked when the key is being processed.  They
are not used to identify the key initially.  See
edit/edit_key_translator.c.

-- 
Regards,
Pavel Roskin