fundamental-mode-hook

fundamental-mode-hook

[Lennart Borgman (gmail)]

The elisp manual

   (info "(elisp) Major Mode Conventions")

says that

   Each major mode should have a normal "mode hook" named
   `MODENAME-mode-hook'.

fundamental-mode does not have such a hook.

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Karl Fogel wrote:
> "Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:
>> The elisp manual
>>
>>   (info "(elisp) Major Mode Conventions")
>>
>> says that
>>
>>   Each major mode should have a normal "mode hook" named
>>   `MODENAME-mode-hook'.
>>
>> fundamental-mode does not have such a hook.
> 
> Do you have a use in mind for such a hook, or is the inconsistency
> just bugging you?  (Not a facetious question, I'm honestly trying to
> figure out if the situation is a problem or not.)
> 
> In simple.el, fundamental-mode's definition is pretty tiny:
> 
>    (defun fundamental-mode ()
>      "Major mode not specialized for anything in particular.
>    Other major modes are defined by comparison with this one."
>      (interactive)
>      (kill-all-local-variables)
>      (unless delay-mode-hooks
>        (run-hooks 'after-change-major-mode-hook)))
> 
> Hmm, for one thing, I'm not sure why it doesn't just say
> 
>    (delay-mode-hooks 'after-change-major-mode-hook)
> 
> instead of checking the variable delay-mode-hooks manually.  Aside
> from that, it would be pretty easy to stick a run-hooks call in there,
> but would it really help anyone do anything they couldn't do before?


I am not sure what to answer at the moment. I just found this 
inconsistency when I tried to check if a function was a major mode 
function. I tested to see if the hook variable existed (probably not the 
best way, but that is what I did).

I then remembered that there are no hook for new buffers. Are they not 
set up in fundamental-mode?

BTW, there are more major modes missing this hook: Info-mode, xml-mode 
is what I have noticed so far.

Re: fundamental-mode-hook

[Juanma Barranquero]

On 2/8/07, Karl Fogel <kfogel@red-bean.com> wrote:

> (Not a facetious question, I'm honestly trying to
> figure out if the situation is a problem or not.)

http://lists.gnu.org/archive/html/help-gnu-emacs/2007-02/msg00042.html

> Hmm, for one thing, I'm not sure why it doesn't just say
>
>    (delay-mode-hooks 'after-change-major-mode-hook)
>
> instead of checking the variable delay-mode-hooks manually.

http://lists.gnu.org/archive/html/emacs-devel/2005-06/msg00886.html

                    /L/e/k/t/u

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Juanma Barranquero wrote:
> On 2/8/07, Karl Fogel <kfogel@red-bean.com> wrote:
> 
>> (Not a facetious question, I'm honestly trying to
>> figure out if the situation is a problem or not.)
> 
> http://lists.gnu.org/archive/html/help-gnu-emacs/2007-02/msg00042.html
> 
>> Hmm, for one thing, I'm not sure why it doesn't just say
>>
>>    (delay-mode-hooks 'after-change-major-mode-hook)
>>
>> instead of checking the variable delay-mode-hooks manually.
> 
> http://lists.gnu.org/archive/html/emacs-devel/2005-06/msg00886.html


I thought that a hook like (the now non-existent) fundamental-mode-hook 
was meant for adding things for minor mode for example.

I can not really understand the rational for not having it. There can 
still be minor mode active in a buffer with in fundamental-mode, or?

(And currently Info is incorrect.)

Re: fundamental-mode-hook

[Karl Fogel]

"Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:
> The elisp manual
>
>   (info "(elisp) Major Mode Conventions")
>
> says that
>
>   Each major mode should have a normal "mode hook" named
>   `MODENAME-mode-hook'.
>
> fundamental-mode does not have such a hook.

Do you have a use in mind for such a hook, or is the inconsistency
just bugging you?  (Not a facetious question, I'm honestly trying to
figure out if the situation is a problem or not.)

In simple.el, fundamental-mode's definition is pretty tiny:

   (defun fundamental-mode ()
     "Major mode not specialized for anything in particular.
   Other major modes are defined by comparison with this one."
     (interactive)
     (kill-all-local-variables)
     (unless delay-mode-hooks
       (run-hooks 'after-change-major-mode-hook)))

Hmm, for one thing, I'm not sure why it doesn't just say

   (delay-mode-hooks 'after-change-major-mode-hook)

instead of checking the variable delay-mode-hooks manually.  Aside
from that, it would be pretty easy to stick a run-hooks call in there,
but would it really help anyone do anything they couldn't do before?

-Karl

Re: fundamental-mode-hook

[Karl Fogel]

"Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:
> Juanma Barranquero wrote:
> I thought that a hook like (the now non-existent)
> fundamental-mode-hook was meant for adding things for minor mode for
> example.
>
> I can not really understand the rational for not having it. There can
> still be minor mode active in a buffer with in fundamental-mode, or?
>
> (And currently Info is incorrect.)

I've added the hook now.  Thanks, Juanma Barranquero, for the timely
archive pointers!

   2007-02-08  Karl Fogel  <kfogel@red-bean.com>

	* emacs/lisp/simple.el (fundamental-mode-hook): Declare new hook.
	(fundamental-mode): Run the new dedicated hook, and don't run
	after-change-major-mode-hooks manually anymore.

   $ cvs ci -F msg simple.el ChangeLog
   /cvsroot/emacs/emacs/lisp/simple.el,v  <--  simple.el
   new revision: 1.844; previous revision: 1.843
   /cvsroot/emacs/emacs/lisp/ChangeLog,v  <--  ChangeLog
   new revision: 1.10663; previous revision: 1.10662
   Mailing notification to emacs-commit@gnu.org... sent.
   Mailing diffs to emacs-diffs@gnu.org... 2 sent.
   $ 

-Karl

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Karl Fogel wrote:
> "Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:
>> Juanma Barranquero wrote:
>> I thought that a hook like (the now non-existent)
>> fundamental-mode-hook was meant for adding things for minor mode for
>> example.
>>
>> I can not really understand the rational for not having it. There can
>> still be minor mode active in a buffer with in fundamental-mode, or?
>>
>> (And currently Info is incorrect.)
> 
> I've added the hook now.  Thanks, Juanma Barranquero, for the timely
> archive pointers!
> 
>    2007-02-08  Karl Fogel  <kfogel@red-bean.com>
> 
> 	* emacs/lisp/simple.el (fundamental-mode-hook): Declare new hook.
> 	(fundamental-mode): Run the new dedicated hook, and don't run
> 	after-change-major-mode-hooks manually anymore.
> 
>    $ cvs ci -F msg simple.el ChangeLog
>    /cvsroot/emacs/emacs/lisp/simple.el,v  <--  simple.el
>    new revision: 1.844; previous revision: 1.843
>    /cvsroot/emacs/emacs/lisp/ChangeLog,v  <--  ChangeLog
>    new revision: 1.10663; previous revision: 1.10662
>    Mailing notification to emacs-commit@gnu.org... sent.
>    Mailing diffs to emacs-diffs@gnu.org... 2 sent.
>    $ 


Thanks. I just recompiled from CVS and tested. I can find the hook, but 
it does not seem to be run. Or I am doing something bad. This is what I 
used for testing:

(defun temp-fund()
   (message "temp-fund here")
   (sit-for 2))
(temp-fund)

(add-hook 'fundamental-mode-hook 'temp-fund)


(get-buffer-create "Test fund")

(with-temp-buffer
   (message "%s, major-mode=%s" (current-buffer) major-mode))

Re: fundamental-mode-hook

[Richard Stallman]

Fundamental mode is not supposed to run a hook, it is supposed
to make the global definitions of Emacs show through.
So please take out that change.

Re: fundamental-mode-hook

[Richard Stallman]

    I am not sure what to answer at the moment. I just found this 
    inconsistency when I tried to check if a function was a major mode 
    function. I tested to see if the hook variable existed (probably not the 
    best way, but that is what I did).

That approach won't work in general.  A major mode hook variable does
not have to "exist" in any sense unless the user chooses to put on
a hook.  Some major modes defvar the hook variable, but some don't.

Re: fundamental-mode-hook

[Richard Stallman]

       Each major mode should have a normal "mode hook" named
       `MODENAME-mode-hook'.

    fundamental-mode does not have such a hook.

Fundamental mode is an exception to many of these conventions.
I will update the manual.

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Richard Stallman wrote:
>     I am not sure what to answer at the moment. I just found this 
>     inconsistency when I tried to check if a function was a major mode 
>     function. I tested to see if the hook variable existed (probably not the 
>     best way, but that is what I did).
> 
> That approach won't work in general.  A major mode hook variable does
> not have to "exist" in any sense unless the user chooses to put on
> a hook.  Some major modes defvar the hook variable, but some don't.


Yes, thanks, that is what I thought. What is the right way to find out 
if a function is a major mode?

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Richard Stallman wrote:
>        Each major mode should have a normal "mode hook" named
>        `MODENAME-mode-hook'.
> 
>     fundamental-mode does not have such a hook.
> 
> Fundamental mode is an exception to many of these conventions.
> I will update the manual.


I thought a fundamental mode hook was a good solution to the problem 
that there is no "new buffer" hook. But a special hook would perhaps be 
better.

Re: fundamental-mode-hook

[Stefan Monnier]

> Yes, thanks, that is what I thought.  What is the right way to find out if
> a function is a major mode?

- look at its doc
- look at its code
- try it
- ...

I tried to get all major modes to use a standard macro
(define-derived-mode), but Richard didn't like some parts of it and it
all dies.  Still, all major modes defined with define-derived-mode have the
`derived-mode-parent' property.


        Stefan

Re: fundamental-mode-hook

[Edward O'Connor]

Lennart asked:

> What is the right way to find out if a function is a major mode?

Kevin Rodgers posted the following function to g.e.h or somewhere
similar some number of years ago. It's just a heuristic, but it works
well enough for my purposes.

,----
| (defun kr-major-mode-p (symbol)
|   "Return non-nil if SYMBOL is a major mode."
|   (and (fboundp symbol)
|        (let ((function-name (symbol-name symbol)))
|          (and (string-match "-mode\\'" function-name)
|               (not (string-match "\\`turn-\\(on\\|off\\)-"
|                                  function-name))))
|        (not (assq symbol minor-mode-alist))))
`----


Ted

-- 
Edward O'Connor
hober0@gmail.com

Ense petit placidam sub libertate quietem.

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Stefan Monnier wrote:
>> Yes, thanks, that is what I thought.  What is the right way to find out if
>> a function is a major mode?
> 
> - look at its doc
> - look at its code
> - try it
> - ...



Some of it could be done from elisp of course, but it seems complicated 
... ;-)


> I tried to get all major modes to use a standard macro
> (define-derived-mode), but Richard didn't like some parts of it and it
> all dies.  Still, all major modes defined with define-derived-mode have the
> `derived-mode-parent' property.


Seems easier. Combining that with the code that checks if it is a minor 
mode ...

But something even easier would be nice.

Re: fundamental-mode-hook

[Lennart Borgman (gmail)]

Edward O'Connor wrote:
> Lennart asked:
> 
>> What is the right way to find out if a function is a major mode?
> 
> Kevin Rodgers posted the following function to g.e.h or somewhere
> similar some number of years ago. It's just a heuristic, but it works
> well enough for my purposes.
> 
> ,----
> | (defun kr-major-mode-p (symbol)
> |   "Return non-nil if SYMBOL is a major mode."
> |   (and (fboundp symbol)
> |        (let ((function-name (symbol-name symbol)))
> |          (and (string-match "-mode\\'" function-name)
> |               (not (string-match "\\`turn-\\(on\\|off\\)-"
> |                                  function-name))))
> |        (not (assq symbol minor-mode-alist))))
> `----



Thanks, looks useful.

Re: fundamental-mode-hook

[Karl Fogel]

Richard Stallman <rms@gnu.org> writes:
> Fundamental mode is not supposed to run a hook, it is supposed
> to make the global definitions of Emacs show through.
> So please take out that change.

Done; thanks for updating the manual.

-Karl

Re: fundamental-mode-hook

[Richard Stallman]

    Yes, thanks, that is what I thought. What is the right way to find out 
    if a function is a major mode?

I think there is no reliable way.

In a future version we could add a way.

Re: fundamental-mode-hook

[Richard Stallman]

    I thought a fundamental mode hook was a good solution to the problem 
    that there is no "new buffer" hook. But a special hook would perhaps be 
    better.

Creating a buffer does not call `fundamental-mode', so adding a
fundamental-mode-hook wouldn't do this anyway.  Buffer creation
happens at a rather low level, but it can call Lisp code.
Adding a hook for that makes me worry, because I know that I hesitated
to do so, but I don't remember any specific reason why it was bad,
and it does call a function to help initialize UCS.

I don't want to consider such a change now.