headers
Peter S Galbraith | 1 Dec 2002 01:22
X-Face
Favicon

Re: Emacs21 mh-alias users no longer need multi-prompt.el

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

> Peter S Galbraith <p.galbraith <at> globetrotter.net> wrote:
> 
> > Jeffrey C Honig <jch <at> honig.net> wrote:
> > 
> > > Do you think it would make sense to prime mh-alias-timestamp from
> > > ``mhparam aliasfile''?
> > 
> > Is there a downside to changing `mh-alias-timestamp' to an internal
> > defvar instead of a defcustom?
> > 
> > I'd prime it from ``mhparam aliasfile'' and the user woulndn't need to
> > get involved at all.
> 
>   This is a great idea!

I found one small downside, but I think we can overlook it.  Tell me
whether you agree.

$ tail -1 /etc/nmh/MailAliases
testalias: p.galbraith <at> globetrotter.net
$ /usr/bin/mh/ali testalias
p.galbraith <at> globetrotter.net

The system alias file is translated, however I can't tell what that file
is from mhparam:

$ /usr/bin/mh/mhparam AliasFile
mailaliases
(Continue reading)

Permalink | Reply | 17 comments |
headers
Bill Wohler | 1 Dec 2002 01:26
Picon

CVS: src mh-utils.el,1.178,1.179 mh-speed.el,1.27,1.28 mh-seq.el,1.73,1.74 mh-pick.el,1.22,1.23 mh-mime.el,1.93,1.94 mh-index.el,1.52,1.53 mh-identity.el,1.12,1.13 mh-funcs.el,1.29,1.30 mh-e.el,1.201,1.202 mh-comp.el,1.149,1.150 Makefile,1.24,1.25 ChangeLog,1.551,1.552 .cvsignore,1.1,1.2

Update of /cvsroot/mh-e/src
In directory sc8-pr-cvs1:/tmp/cvs-serv2472

Modified Files:
	mh-utils.el mh-speed.el mh-seq.el mh-pick.el mh-mime.el 
	mh-index.el mh-identity.el mh-funcs.el mh-e.el mh-comp.el 
	Makefile ChangeLog .cvsignore 
Log Message:
* mh-funcs.el, mh-identity.el, mh-index.el, mh-pick.el, mh-seq.el:
Removed MH-E autoloads. These are now provided by mh-loaddefs.el.
Added ;;;###mh-autoload tag to selected functions so that autoloads
for those functions can be automatically created in mh-loaddefs.el
(closes half of SF #643722).

* mh-mime.el (mh-buffer-data): Moved to mh-utils where it is used.
Added ;;;###mh-autoload tag to selected functions so that autoloads
for those functions can be automatically created in mh-loaddefs.el
(closes half of SF #643722)..

* mh-utils.el (mh-loaddefs): Added require.
(mh-buffer-data): Moved here from mh-mime.el as it is used in
mh-display-msg.
(mh-set-cmd-note): Moved to mh-e.el. Removed MH-E autoloads. These are
now provided by mh-loaddefs.el. Added ;;;###mh-autoload tag to
selected functions so that autoloads for those functions can be
automatically created in mh-loaddefs.el (closes half of SF #643722)..

* mh-e.el (mh-set-cmd-note): Moved here from mh-utils.el since it uses
mh-scan-format-file which is defined here. Removed MH-E autoloads.
These are now provided by mh-loaddefs.el. Added ;;;###mh-autoload tag
(Continue reading)

Permalink | Reply | 0 comments |
headers
noreply | 1 Dec 2002 01:28
Picon
Favicon

[ mh-e-Feature Requests-643722 ] Autoload and defcustom reorg

Feature Requests item #643722, was opened at 2002-11-25 11:15
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=363357&aid=643722&group_id=13357

Category: General
Group: mh-e-7.0
Status: Open
Resolution: None
Priority: 7
Submitted By: Bill Wohler (wohler)
Assigned to: Bill Wohler (wohler)
Summary: Autoload and defcustom reorg

Initial Comment:
Unfortunately, in many cases we need to load libraries
to be able to read their documentation or customize
them. Thus, we need to add the full documentation to
the autoloads and perhaps add more of them (such as the
indexing functions).

But we don't want to have to manually keep the
documentation in the autoload in sync with the
function's documentation.

Note that there already exists a ";;;###autoload" tag
that Emacs uses to define autoloads for a module's
entry point. This has already been done for MH-E. We
should extend this functionality by defining:

(defconst generate-autoload-cookie ";;;###mh-autoload")
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bill Wohler | 1 Dec 2002 01:32
X-Face
Picon
Picon
Gravatar

Re: Emacs21 mh-alias users no longer need multi-prompt.el

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

> I found one small downside, but I think we can overlook it.  Tell me
> whether you agree.
> 
> $ tail -1 /etc/nmh/MailAliases
> testalias: p.galbraith <at> globetrotter.net
> $ /usr/bin/mh/ali testalias
> p.galbraith <at> globetrotter.net
> 
> The system alias file is translated, however I can't tell what that file
> is from mhparam:
> 
> $ /usr/bin/mh/mhparam AliasFile
> mailaliases
> 
> I figure that for most cases, the system alias file doesn't change very
> often, so checking the timestamp of only user alias file is probably
> sufficient.   Okay?

  I don't understand the question. The timestamps for all of the files
  in question can and should be checked. What exactly is the problem in
  doing that?

--
Bill Wohler <wohler <at> newt.com>  http://www.newt.com/wohler/  GnuPG ID:610BD9AD
Maintainer of comp.mail.mh FAQ and MH-E. Vote Libertarian!
If you're passed on the right, you're in the wrong lane.
(Continue reading)

Permalink | Reply | 10 comments |
headers
Peter S Galbraith | 1 Dec 2002 01:40
X-Face
Favicon

Re: mh-alias for everybody?

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

> Peter S Galbraith <p.galbraith <at> globetrotter.net> wrote:
> 
> > mh-comp.el has this:
> > 
> > (defun mh-read-address (prompt)
> >   "Read a To: or Cc: address, prompting in the minibuffer with PROMPT.
> > May someday do completion on aliases."
> >   (read-string prompt))
> > 
> > So... Should I replace it with mh-alias' version for everybody (if that
> > doesn't break bbdb, which I need to check but I'm pretty ure of)?
> 
>   It's a part of MH-E now. Might as well make it as transparent as
>   possible!

:-)

Cool.  I'll change it when it has reached some level of stability.  I'm
changing lots of stuff now, so it's not a good time to impose it on all
of you guys.
Permalink | Reply | 0 comments |
headers
Peter S Galbraith | 1 Dec 2002 01:45
X-Face
Favicon

Re: Emacs21 mh-alias users no longer need multi-prompt.el

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

>   I don't understand the question. The timestamps for all of the files
>   in question can and should be checked. What exactly is the problem in
>   doing that?

That I was going to kill the variable holding the filename in order to
obtain them from mhparam.  But mhparam doesn't output the system alias
file.
Permalink | Reply | 9 comments |
headers
Peter S Galbraith | 1 Dec 2002 01:47
X-Face
Favicon

Re: Automatic creation of autoloads

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

>   I'll check this in after a short break.

So do we all generate mh-loaddefs.el locally?  

Or do you generate it, check it into CVS, and we each update it in CVS
when we add an autoload tag?
Permalink | Reply | 0 comments |
headers
Jeffrey C Honig | 1 Dec 2002 01:48

Re: Emacs21 mh-alias users no longer need multi-prompt.el

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

> I figure that for most cases, the system alias file doesn't change very
> often, so checking the timestamp of only user alias file is probably
> sufficient.   Okay?

Can we assume it is MailAliases in mh-lib?

Thanks.

Jeff
Permalink | Reply | 5 comments |
headers
Peter S Galbraith | 1 Dec 2002 01:59
X-Face
Favicon

Re: Emacs21 mh-alias users no longer need multi-prompt.el

Jeffrey C Honig <jch <at> honig.net> wrote:

> Peter S Galbraith <p.galbraith <at> globetrotter.net> wrote:
> 
> > I figure that for most cases, the system alias file doesn't change very
> > often, so checking the timestamp of only user alias file is probably
> > sufficient.   Okay?
> 
> Can we assume it is MailAliases in mh-lib?

Nah.  On Debian, it's /etc/nmh/MailAliases
Permalink | Reply | 4 comments |
headers
Peter S Galbraith | 1 Dec 2002 02:03
X-Face
Favicon

Debian 7.0 package web page

http://packages.debian.org/unstable/mail/mh-e.html

That's nice to see...
Permalink | Reply | 1 comment |

Gmane