Re: [Lynx-dev] backspace vs. delete can't have separate function

[Thomas Dickey]

On Mon, 28 May 2007, Rado S wrote:

> =- Thomas Dickey wrote on Mon 28.May'07 at  9:05:50 -0400 -=
>
> > It seems that curses is returning ^H as KEY_BACKSPACE (0x107 is
> > 0407 for which see curses.h rather than lynx.cfg's KEYMAP), which
> > lynx then translates into <delete> (see LYgetch_for()):
>
> Argl, this means the only way to use both separately is to undefine
> KEY_BACKSPACE or uncomment that code there?
> Ugh, this doesn't even help: although it lists both with separate
> functions, still the BS key doesn't produce the HISTORY function
> as desired but some DOWN_LINK.
>
> > GETCH: Got 0x107.
> > LYReadCmdKey(0) -><delete> (0x7f)
>
> Uh, here it produces this:
> GETCH: Got 0x107.
> LYReadCmdKey(0) ->End (0x107)
>
> Is this "normal" or even "ok"?

It doesn't look right - that looks as if it's not seeing the KEY_BACKSPACE
case and is returning Lynx's code.

It-would-be-nice if Lynx used curses codes more than it does,
e.g., so it didn't have its own meaning for 0x107.  It does
that for a couple of reasons: a lot of the keyboard code was
contrived to work with VMS curses, and then hacked up for slang...

> > > BTW, what kind of codes >0xFF are those?
> > > I know they are "function keys", but how are they translated?
> > > How do they relate to literal chars pressed or terminfo?
> >
> > it's written down in lynx.cfg (but as noted above, curses and slang
> > have their own codes which lynx first translates into its own):
>
> I've seen the list before, I wasn't looking for the literal
> translation, but rather why and how this happens the way it happens
> with ncurses.

For ncurses, I assume it happens because there's a "kbs=^H" in the 
terminal description...

--
Thomas E. Dickey
http://invisible-island.net
ftp://invisible-island.net