gmane.mail.mh-e.user

mh-identities: what purpose ?

Hi,

I know what they are and what they do _but_ I do not see in what they are useful
compared to MH templates (comp, repl, etc...).

All of what mh-identity provides can be done at MH level and thus be used either
into the MUA or directly from a shell command.

Why not provide *official* support for templates and rely on them instead of
reinventing the wheel ?

I see there are solutions to compose a message according to a template, but they
look like *hacks* and not reliable solution.

WDYT ?

Regards,

Xavier

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone

headers

Peter S Galbraith | 21 Apr 16:07

Re: mh-identities: what purpose ?

Xavier Maillard <xma <at> gnu.org> wrote:

> Hi,
> 
> I know what they are and what they do _but_ I do not see in what they
> are useful compared to MH templates (comp, repl, etc...).
> 
> All of what mh-identity provides can be done at MH level and thus be
> used either into the MUA or directly from a shell command.
> 
> Why not provide *official* support for templates and rely on them
> instead of reinventing the wheel ?
> 
> I see there are solutions to compose a message according to a
> template, but they look like *hacks* and not reliable solution.
> 
> WDYT ?

Sorry that you don't like my _hack_.  My own view of MH-E is that
someone should be able to use it _without_ having the learn elisp
hacking (even to the point of `setq', everything should be configurable
using the custom interface), or having the learn all the internals of
MH/nmh.

mh-identity is simple to configure and to use, in spite of being a hack.

You are welcome to write and contribute alternate code that uses
templates if you wish.

--

--

(Continue reading)

headers

Xavier Maillard | 21 Apr 23:50

Re: mh-identities: what purpose ?

Hi Peter,

Peter S Galbraith <p.galbraith <at> globetrotter.net> wrote:

> > I see there are solutions to compose a message according to a
> > template, but they look like *hacks* and not reliable solution.
> > 
> > WDYT ?
> 
> Sorry that you don't like my _hack_.  My own view of MH-E is that
> someone should be able to use it _without_ having the learn elisp
> hacking (even to the point of `setq', everything should be configurable
> using the custom interface), or having the learn all the internals of
> MH/nmh.

Well, I did not mean to be rude with you or any hackers here.
Sorry if my wordings gave this feeling. I agree that the current
mh-identity is rather simple to configure and to use _but_ my
guess is that we should try to offer a more MH way so that any MH
applicable solution may also apply for MH-E. Bill has posted
something that permits to call comp with different templates,
this is, IMO, the right way to do things but till it becomes part
of MH-E officially, I won't use it but stick with mh-identity.

> mh-identity is simple to configure and to use, in spite of being a hack.

Well, you misunterpreted what I wanted to say :) mh-identity is
not what I am calling a hack. What I was calling a hack was
Bill's solution to have templates at hand. There were no rant
against mh-identity.

(Continue reading)

headers

Bill Wohler | 22 Apr 07:02

Re: mh-identities: what purpose ?

Xavier Maillard <xma <at> gnu.org> wrote:

> > mh-identity is simple to configure and to use, in spite of being a hack.
> 
> Well, you misunterpreted what I wanted to say :) mh-identity is
> not what I am calling a hack. What I was calling a hack was
> Bill's solution to have templates at hand.

Definitely a hack .

> I know what [identities] are and what they do _but_ I do not see in
> what they are useful compared to MH templates (comp, repl, etc...).

I guess you've read the Developer's Guide. In particular, 

http://mh-e.sourceforge.net/doc/devguide.html#Philosophy

    4 Philosophy

    This chapter discusses the philosophy and principles of the MH-E
    project. The first section covers our coding philosophy, while the
    second section summarizes the principles of the team that have
    evolved over time. 

    Code

    The core philosophies of the MH-E project regarding the code are as follows:

       1. Keep the code small and fast
       2. Refrain from adding lots of code to the codebase that would be

(Continue reading)

headers

Bill Wohler | 23 Apr 05:46

Re: mh-identities: what purpose ?

Bill Wohler <wohler <at> newt.com> wrote:

> Xavier Maillard <xma <at> gnu.org> wrote:
> 
> > > mh-identity is simple to configure and to use, in spite of being a hack.
> > 
> > Well, you misunterpreted what I wanted to say :) mh-identity is
> > not what I am calling a hack. What I was calling a hack was
> > Bill's solution to have templates at hand.
> 
> Definitely a hack .

OK, I looked over the code and perhaps I misspoke!

  (defun bw-mh-folder-mode-hook ()
    "Hook to set key bindings in MH-Folder mode."
    (make-local-variable 'transient-mark-mode)
    (transient-mark-mode t)
    (local-set-key "\C-cm" 'bw-compose-mail-with-template)
    (local-set-key "\C-cr" 'bw-reply-with-template))

Setting up a key binding certainly isn't a hack.

  (defun bw-compose-mail-with-template (form)
    "Compose and send a letter with alternate template FORM."
    (interactive (list (read-file-name
			"Form: "
			(concat mh-user-path "comps/") nil t)))
    (let ((mh-comp-formfile (expand-file-name form)))
      (call-interactively 'mh-send)))

(Continue reading)

headers

Peter Galbraith | 27 Apr 03:49

Re: mh-identities: what purpose ?

Xavier Maillard <xma <at> gnu.org> wrote:

> Hi Peter,
> 
> Peter S Galbraith <p.galbraith <at> globetrotter.net> wrote:
> 
> > > I see there are solutions to compose a message according to a
> > > template, but they look like *hacks* and not reliable solution.
> > > 
> > > WDYT ?
> > 
> > Sorry that you don't like my _hack_.  My own view of MH-E is that
> > someone should be able to use it _without_ having the learn elisp
> > hacking (even to the point of `setq', everything should be configurable
> > using the custom interface), or having the learn all the internals of
> > MH/nmh.
> 
> Well, I did not mean to be rude with you or any hackers here.

No, I'm the one who is sorry.  I had completely misread your post.
Sorry about that.  I was way too defensive anyway.  Hope there are no
hard feelings.

I should have replied earlier, but it's been a busy week!

Peter

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100.

(Continue reading)