message-x.el

message-x.el

[Robert Epprecht]

;; message-x.el -- customizable completion in message headers
;; Copyright (C) 1998 Kai Großjohann

I do use and love Kai's message-x.el
I wonder why this is not included in Gnus?  Kai?

Robert Epprecht

Re: message-x.el

[Matthieu Moy]

Robert Epprecht <epprecht@sunweb.ch> writes:

> ;; message-x.el -- customizable completion in message headers
> ;; Copyright (C) 1998 Kai Großjohann
>
> I do use and love Kai's message-x.el
> I wonder why this is not included in Gnus?  Kai?

I also do ! 

(same question for BBDB and Emacs)

-- 
Matthieu

Re: message-x.el

[ShengHuo ZHU]

Matthieu Moy <matthieumNOSPAM@yahoo.com.invalid> writes:

[...]

> (same question for BBDB and Emacs)

I think it is a copyright problem.

ShengHuo

Re: message-x.el

[Kai Großjohann]

Robert Epprecht <epprecht@sunweb.ch> writes:

> I do use and love Kai's message-x.el
> I wonder why this is not included in Gnus?  Kai?

I think the most important functionality is already part of Gnus, see
the function message-tab.

There is some functionality which message-x has but message-tab does
not: message-x-tab tries to find out whether it should jump to the
next header.  message-tab does nothing when it can't complete.

Do you think that this functionality is important?  (The
implementation is very hairy.)

kai
-- 
~/.signature is: umop 3p!sdn    (Frank Nobis)

Re: message-x.el

[Matthieu Moy]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> I think the most important functionality is already part of Gnus, see
> the function message-tab.

I've just seen that in your previous post. 

> There is some functionality which message-x has but message-tab does
> not: message-x-tab tries to find out whether it should jump to the
> next header.  message-tab does nothing when it can't complete.
>
> Do you think that this functionality is important?  (The
> implementation is very hairy.)

I  like it,  yes. When  I'm in  a  header, I  just hit  TAB, and  your
message-x guesses exactly what I want. To go to the body, TAB is a lot
shorter than C-c C-b, and  easier to remember (I had already forgotten
it !) 

-- 
Matthieu

Re: message-x.el

[Kai Großjohann]

Matthieu Moy <matthieumNOSPAM@yahoo.com.invalid> writes:

> I  like it,  yes. When  I'm in  a  header, I  just hit  TAB, and  your
> message-x guesses exactly what I want. To go to the body, TAB is a lot
> shorter than C-c C-b, and  easier to remember (I had already forgotten
> it !) 

You must be very lucky that message-x guesses what you want :-)

Anyhow, maybe it's a good idea to add this functionality to
message-tab, as well.  The problem is to find out, after invoking a
completion function, what the function did.

kai
-- 
~/.signature is: umop 3p!sdn    (Frank Nobis)

Re: message-x.el

[Fabien Penso]

Matthieu on Thu, 31 Jan 2002 18:00:00 +0100 wrote:

 >> There is some functionality which message-x has but message-tab does
 >> not: message-x-tab tries to find out whether it should jump to the
 >> next header.  message-tab does nothing when it can't complete.
 >> 
 >> Do you think that this functionality is important?  (The
 >> implementation is very hairy.)

 > I  like it,  yes. When  I'm in  a  header, I  just hit  TAB, and  your
 > message-x guesses exactly what I want. To go to the body, TAB is a lot
 > shorter than C-c C-b, and  easier to remember (I had already forgotten
 > it !) 

Same here. I just love it :-)

-- 
Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de :
http://perso.LinuxFR.org/penso/  | http://linuxfr.org/dons/

Re: message-x.el

[David S. Goldberg]

>>>>> On Mon, 04 Feb 2002 18:03:54 +0100, Fabien Penso
>>>>> <penso@linuxfr.org> said:

>>> Do you think that this functionality is important?  (The
>>> implementation is very hairy.)

>> I  like it,  yes. When  I'm in  a  header, I  just hit  TAB, and  your
>> message-x guesses exactly what I want. To go to the body, TAB is a lot
>> shorter than C-c C-b, and  easier to remember (I had already forgotten
>> it !) 

> Same here. I just love it :-)

To each his own.  I way too often find myself completing on multiple
addresses, particularly in Cc, but also in To.  Back when I used
message-x, I turned off the jumping functionality immediately upon
finding myself typing addresses into the body :-) I am not suggesting
that the "guess-where-to-go" functionality shouldn't be incorporated,
but it should be turned off by default.

-- 
Dave Goldberg
david.goldberg6@verizon.net

Re: message-x.el

[Daniel Pittman]

On Thu, 31 Jan 2002, Kai Großjohann wrote:

[...]

> There is some functionality which message-x has but message-tab does
> not: message-x-tab tries to find out whether it should jump to the
> next header.  message-tab does nothing when it can't complete.
> 
> Do you think that this functionality is important?  (The
> implementation is very hairy.)

No, because it's an AI-complete problem. ;)
        Daniel

-- 
Big whorls have little whorls that feed on their velocity
And little whorls have lesser whorls and so on to viscosity.
        -- Lewis F. Richardson, quoted in Gleick's 'Chaos'