push-input, accept-and-hold, etc., and vared

push-input, accept-and-hold, etc., and vared

[Bart Schaefer]

Using push-input or accept-and-hold during vared has unexpected (to me,
at least) behavior -- to wit, the value of the variable is pushed onto
the command input stack, and vared exits, whereupon the value of the
variable is immediately popped into the PS1 editor buffer.

There may be obscure circumstances in which this is what is meant, but
(particularly in the case of "vared mapfile[...]") that's seldom so.

I'm not precisely sure what to suggest we do about it, though.  It might
be nice to have accept-and-hold set the variable but not exit vared (this
would be particularly useful with "vared mapfile[...]").  I'm at a loss
with respect to push-input and friends.  Perhaps they just shouldn't work?

On a semi-related question, is it time that vared had its own keymap?
The problem is initializing it properly, I suppose.

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

RE: push-input, accept-and-hold, etc., and vared

[Andrej Borsenkow]

> 
> I was thinking about an option to vared allowing users to say which
> keymap they want to use there. And allowing users to define keymaps.
> But this would require some more discussion, I think, so I held myself 
> back (also thinking about the things Anthony once spoke about -- quite 
> a while ago in a message which number I don't seem to have any more).
> 

6938 and it was back in July 1999. 

-andrej

Re: push-input, accept-and-hold, etc., and vared

[Sven Wischnowsky]

Bart Schaefer wrote:

> Using push-input or accept-and-hold during vared has unexpected (to me,
> at least) behavior -- to wit, the value of the variable is pushed onto
> the command input stack, and vared exits, whereupon the value of the
> variable is immediately popped into the PS1 editor buffer.
> 
> There may be obscure circumstances in which this is what is meant, but
> (particularly in the case of "vared mapfile[...]") that's seldom so.
> 
> I'm not precisely sure what to suggest we do about it, though.  It might
> be nice to have accept-and-hold set the variable but not exit vared (this
> would be particularly useful with "vared mapfile[...]").  I'm at a loss
> with respect to push-input and friends.  Perhaps they just shouldn't work?
> 
> On a semi-related question, is it time that vared had its own keymap?
> The problem is initializing it properly, I suppose.

I was thinking about an option to vared allowing users to say which
keymap they want to use there. And allowing users to define keymaps.
But this would require some more discussion, I think, so I held myself 
back (also thinking about the things Anthony once spoke about -- quite 
a while ago in a message which number I don't seem to have any more).


Bye
 Sven


--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de