Re: mgp.el

Re: mgp.el

[Christian Egli]

[-- Attachment #1: Type: text/plain, Size: 345 bytes --]

Pavel@janik.cz (Pavel Janík) writes:

> Hello Christian,
> 
> I have just read about mgp.el and I'd like to thank for that work. Can you
> please fix the following issues and send it to emacs-devel@gnu.org for
> inclusion? This could be really useful.

Here's a version that hopefully fixes all the issues you mention and
some more.


[-- Attachment #2: Magic Point mode --]
[-- Type: application/emacs-lisp, Size: 2952 bytes --]

[-- Attachment #3: Type: text/plain, Size: 119 bytes --]


-- 
Christian Egli
Acterna Zürich AG, Foerrlibuckstrasse 62, P.O. Box 74, CH-8037 Zuerich
http://www.acterna.com

Re: mgp.el

[Pavel Janík]

   From: Christian Egli <christian.egli@stest.ch>
   Date: 26 Feb 2002 17:17:32 +0100

Hi,

   > > I have just read about mgp.el and I'd like to thank for that work. Can you
   > > please fix the following issues and send it to emacs-devel@gnu.org for
   > > inclusion? This could be really useful.
   > 
   > Here's a version that hopefully fixes all the issues you mention and
   > some more.

[ ... mgp.el omitted ... ]

what do you think about this?

MagickPoint is quite good presentation program (released under BSD-like
license) which uses textual description of the presentation. The syntax of
presentation's source files is very easy and together with Emacs' editing
capabilities create a very nice presentation toolbox. I vote for including
that stuff into generic-x.el. If you agree on that with me, I will take
care of it.

BTW - what do you use for presentations? Shouldn't we implement something
like MagickPoint in Lisp to use only Emacs?

P.S. Christian, please add "%charset" in the KEYWORD-LIST argument of
define-generic-mode.
-- 
Pavel Janík

The only way to handle bad stuff is to make people _responsible_ for their
own sh*t, and have them maintain it themselves.
                  -- torvalds@transmeta.com (Linus Torvalds) in LKML

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: mgp.el

[Kenichi Handa]

Pavel@Janik.cz (Pavel Janík) writes:
> MagickPoint is quite good presentation program (released under BSD-like
> license) which uses textual description of the presentation. The syntax of
> presentation's source files is very easy and together with Emacs' editing
> capabilities create a very nice presentation toolbox. I vote for including
> that stuff into generic-x.el. If you agree on that with me, I will take
> care of it.

I agree.  I was also using MagicPoing before.

> BTW - what do you use for presentations? Shouldn't we implement something
> like MagickPoint in Lisp to use only Emacs?

As far as I know, actually there are already two such
programs.  One is elpoint
(http://sourceforge.net/projects/elpoint/).  Another is emp
(EMacs Point) which I wrote a while ago.  I recently always
use emp for my presentation but don't have time to write
documentation and make it stable enough to be released
public.  :-(

---
Ken'ichi HANDA
handa@etl.go.jp


_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: mgp.el

[Christian Egli]

[-- Attachment #1: Type: text/plain, Size: 661 bytes --]

Pavel@janik.cz (Pavel Janík) writes:

>    > > I have just read about mgp.el and I'd like to thank for that work. Can you
>    > > please fix the following issues and send it to emacs-devel@gnu.org for
>    > > inclusion? This could be really useful.
>    > 
>    > Here's a version that hopefully fixes all the issues you mention and
>    > some more.
> 
> I vote for including that stuff into generic-x.el. If you agree on
> that with me, I will take care of it.

That would be fine with me.

> P.S. Christian, please add "%charset" in the KEYWORD-LIST argument of
> define-generic-mode.

Yup, done. See the attached new version of mgp.el


[-- Attachment #2: mgp.el --]
[-- Type: application/emacs-lisp, Size: 2975 bytes --]

[-- Attachment #3: Type: text/plain, Size: 121 bytes --]



-- 
Christian Egli
Acterna Zürich AG, Foerrlibuckstrasse 62, P.O. Box 74, CH-8037 Zuerich
http://www.acterna.com

Re: mgp.el

[Christian Egli]

Christian Egli <christian.egli@stest.ch> writes:

> Pavel@janik.cz (Pavel Janík) writes:
> 
> >    > > I have just read about mgp.el and I'd like to thank for that work. Can you
> >    > > please fix the following issues and send it to emacs-devel@gnu.org for
> >    > > inclusion? This could be really useful.
> >    > 
> >    > Here's a version that hopefully fixes all the issues you mention and
> >    > some more.
> > 
> > I vote for including that stuff into generic-x.el. If you agree on
> > that with me, I will take care of it.
> 
> That would be fine with me.
> 
> > P.S. Christian, please add "%charset" in the KEYWORD-LIST argument of
> > define-generic-mode.
> 
> Yup, done. See the attached new version of mgp.el

Actually to followup to myself. After having done the research I
should have done in the first place, I found out that Magicpoint comes
with a file named mgp-mode.el which is also a Magicpoint mode.

It has more features than my barebones mgp-mode, e.g. it has keyword
completion, tempo.el-style code insertions and (that one is cute) can
jump to arbitrary pages of your presentation.

However, it is fairly old, has been subject to some bit rot
(e.g. fontification is done with hilit19, etc.) and it states that it
is based on Emacs 19.28.

I'll try to contact the original author and ask him if we could merge
the two modes.

-- 
Christian Egli
Acterna Zürich AG, Foerrlibuckstrasse 62, P.O. Box 74, CH-8037 Zuerich
http://www.acterna.com

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: mgp.el

[Richard Stallman]

    MagickPoint is quite good presentation program (released under BSD-like
    license) which uses textual description of the presentation. The syntax of
    presentation's source files is very easy and together with Emacs' editing
    capabilities create a very nice presentation toolbox. I vote for including
    that stuff into generic-x.el. If you agree on that with me, I will take
    care of it.

That is a good place to put the code, but we need papers first.
I have asked for them.

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: mgp.el

[Richard Stallman]

    I'll try to contact the original author and ask him if we could merge
    the two modes.

That is a useful project, so thank you.  We would need papers from that
original author.


_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://mail.gnu.org/mailman/listinfo/emacs-devel

Re: mgp.el

[Christian Egli]

Richard Stallman <rms@gnu.org> writes:

> We got the papers for mgp.el.  Can you send me the code
> to install?

I have worked some more on the code and it is now IMHO more or less
feature complete. There are two more things that could be done which
are mentioned in the documentation under TODO.

You can fetch it from http://www.umverkehr.ch/egli/mgp.el

There is one slight problem in that the name of this file clashes with
an Emacs Lisp program named mgp.el to convert mgp file to
LaTeX2e. This program is shipped in the contrib directory of
magicpoint.

I don't know if we have to go through the paper signing process again
if we choose to change the name of the .el file.

To me it looks like we have the following options:

1. Ignore the name clash and let the magicpoint developers handle this
   problem. This might be a problem for e.g. Debian where all
   magicpoint .el files are put into the site-lisp file.

2. Rename our version before putting it into Emacs. This might require
   a rerun of the paper signing process. Possible names are
   magicpoint.el or magicpoint-mode.el (mgp.el and mgp-mode.el are
   taken as they are shipped in the contrib directory of the
   magicpoint sources).

> Have you tested it with the latest development version of Emacs,
> as found in the CVS repository on savannah.gnu.org?  If not,
> it would be very useful if you did that now.

I have tested it with GNU Emacs 21.2.1 (i386-debian-linux-gnu, X
toolkit, Xaw3d scroll bars) on Debian

I'll try to test it with CVS this week.

-- 
Christian Egli
wyona: research & development
http://www.wyona.com

Re: mgp.el

[Stefan Monnier]

> 2. Rename our version before putting it into Emacs. This might require
>    a rerun of the paper signing process. Possible names are
>    magicpoint.el or magicpoint-mode.el (mgp.el and mgp-mode.el are
>    taken as they are shipped in the contrib directory of the
>    magicpoint sources).

There's no need to re-sign any paper.  As for the name, I'd
lean towards mgp-mode.el but only if you are confident that the
missing features from the existing mgp-mode.el will be added
soon and you are confident that this old mgp-mode.el will not be
developed further.

Have you gotten in touch with the author/maintainer of the old
mgp-mode.el ?  What does (s)he think ?


	Stefan

Re: mgp.el

[Christian Egli]

"Stefan Monnier" <monnier+gnu/emacs@rum.cs.yale.edu> writes:

> > 2. Rename our version before putting it into Emacs. This might require
> >    a rerun of the paper signing process. Possible names are
> >    magicpoint.el or magicpoint-mode.el (mgp.el and mgp-mode.el are
> >    taken as they are shipped in the contrib directory of the
> >    magicpoint sources).
> 
> There's no need to re-sign any paper.  As for the name, I'd
> lean towards mgp-mode.el but only if you are confident that the
> missing features from the existing mgp-mode.el will be added
> soon 

As far as I can tell the only two missing features that are only in
the existing mgp-mode.el are:

1. C-u C-c C-c calculates the current page and passes it to magicpoint
   so that the presentation starts with the current page.

   This shouldn't be very hard to add. If I take code from the
   original we need papers from the original author Hirotsugu
   Kakugawa.

2. The existing mgp-mode.el has some key-bindings that let you insert
   directives directly e.g. C-c C-p inserts "%page". 

   IMHO this can be done easier with tempo if there is a demand for
   this feature.

> and you are confident that this old mgp-mode.el will not be
> developed further.

I was in contact with the original developer. It sounded like he was
using Emacs 20.7 and his original version quite happily without doing
any development on it.

> Have you gotten in touch with the author/maintainer of the old
> mgp-mode.el ?  What does (s)he think ?

I had asked him if I could reuse parts of his code and he encouraged me
to to do so. At least that's how I read his mail :-)

For the record here's the mail that I got from the maintainer:

Hirotsugu Kakugawa <kakugawa@aial.hiroshima-u.ac.jp> writes:

> Dear Christian,
> 
> Thank you for your message.
> I'm using mgp-mode.el with Emacs 20.7, 
> but I have not tested it with Emacs 21. It may be outdated.
> My mgp-model.el is distributed under the terms of GPL.
> So you can freely modify it without permission.
> I am looking forward to improved mgp-mode.el!
> 
> 
> Cheers,
> 
> Hirotsugu Kakugawa


-- 
Christian Egli
wyona: research & development
http://www.wyona.com

Re: mgp.el

[Stefan Monnier]

> 2. The existing mgp-mode.el has some key-bindings that let you insert
>    directives directly e.g. C-c C-p inserts "%page".

Oh, that's what is meant by "Keyword completion" ?

>    IMHO this can be done easier with tempo if there is a demand for
>    this feature.

I'd recommend `skeleton' instead of `tempo'.  I'm not sure why I was
originally biased in favor of `skeleton' (they are *very* similar),
but right now I can justify it with the fact that we should strive to
stick to one of the two and `tempo' is only used by `snmp-mode'
and `dcl-mode' whereas `skeleton' is used by at least 6 packages.
Also `define-skeleton' is autoloaded (and indented properly) :-)


	Stefan

Re: mgp.el

[Christian Egli]

"Stefan Monnier" <monnier+gnu/emacs@rum.cs.yale.edu> writes:

> > 2. The existing mgp-mode.el has some key-bindings that let you insert
> >    directives directly e.g. C-c C-p inserts "%page".
> 
> Oh, that's what is meant by "Keyword completion" ?

How would you call this instead?

> >    IMHO this can be done easier with tempo if there is a demand for
> >    this feature.
> 
> I'd recommend `skeleton' instead of `tempo'.  I'm not sure why I was
> originally biased in favor of `skeleton' (they are *very* similar),
> but right now I can justify it with the fact that we should strive to
> stick to one of the two and `tempo' is only used by `snmp-mode'
> and `dcl-mode' whereas `skeleton' is used by at least 6 packages.

I have no preferences either way. I'll have a look at skeleton.

-- 
Christian Egli
wyona: research & development
http://www.wyona.com

Re: mgp.el

[Kai Großjohann]

Christian Egli <christian.egli@wyona.com> writes:

> "Stefan Monnier" <monnier+gnu/emacs@rum.cs.yale.edu> writes:
>
>> > 2. The existing mgp-mode.el has some key-bindings that let you insert
>> >    directives directly e.g. C-c C-p inserts "%page".
>> 
>> Oh, that's what is meant by "Keyword completion" ?
>
> How would you call this instead?

I wouldn't call it completion unless completion is really involved.
How about "keyword insertion"?

But it seems that C-c C-p just inserts a fixed string, right?  Hm.
So it's not really abbrevs, either, because abbrevs are sequences of
self-inserting characters.

Yeah, keyword insertion sounds good.

kai
-- 
A large number of young women don't trust men with beards.  (BFBS Radio)

Re: mgp.el

[Richard Stallman]

After we install the file, we can rename it if we want to.

Re: mgp.el

[Richard Stallman]

    1. C-u C-c C-c calculates the current page and passes it to magicpoint
       so that the presentation starts with the current page.

       This shouldn't be very hard to add. If I take code from the
       original we need papers from the original author Hirotsugu
       Kakugawa.

How many limes of code is that?  We don't need papers if it is small.
Alternatively, can you write your own code to do this job?

Re: mgp.el

[Hirotsugu Kakugawa]

Hi.

rms>     1. C-u C-c C-c calculates the current page and passes it to magicpoint
rms>        so that the presentation starts with the current page.
rms> 
rms>        This shouldn't be very hard to add. If I take code from the
rms>        original we need papers from the original author Hirotsugu
rms>        Kakugawa.
rms> 
rms> How many limes of code is that?  We don't need papers if it is small.
rms> Alternatively, can you write your own code to do this job?

I wrote two elisp functions, and they are only 23 lines of code in total.
They simply count the number of "\n%page" in a buffer.


Cheers,

Hirotsugu Kakugawa

Re: mgp.el

[Stefan Monnier]

> I wrote two elisp functions, and they are only 23 lines of code in total.
> They simply count the number of "\n%page" in a buffer.

BTW, assuming you've set page-delimiter accordingly, you should
be able to use `what-page' except that it's obviously only meant
for interactive use.  Maybe you should fix `what-page' to be usable
from elisp and then use that.


	Stefan

Re: mgp.el

[Richard Stallman]

    I wrote two elisp functions, and they are only 23 lines of code in total.
    They simply count the number of "\n%page" in a buffer.

This is on the borderline of how much code requires papers.
Would you be willing to sign a copyright disclaimer just to play safe?

Re: mgp.el

[Hirotsugu Kakugawa]

Dear RMS,

rms> This is on the borderline of how much code requires papers.
rms> Would you be willing to sign a copyright disclaimer just to play safe?

Yes, I am willing to sign.

-- h.kakugawa