[Axiom-developer] Re: axiom-mode

[C Y]

--- Martin Rubey <address@hidden> wrote:

> C Y <address@hidden> writes:
> 
> > Here's the two cent tour:
> 
> Oh, that was great!  It's not really what I'm used to, but maybe it
> could accomodate the behaviour I like, to.Currently, I use emacs
> shell mode.  Here goes:

It was sort of a "notebook in textmode" as far as concept went.
 
> * at times, I like to edit output.  For example, I like to mark some
> parts of
>   the output with a different font, so that I can more easily focus
> my
>   attention.  Sometimes I also rearange things a little.  Is this
> possible with
>   your mode?  You write:

No, editing output in-notebook isn't possible currently. If one wants
to do that the output can be copied to another buffer, but in the Axiom
session itself I wanted to ensure the integrity of the output as much
as I could (and prevent stupid accidental deleting if I stuck the
cursor in the wrong place with the mouse...)

> so it sounds as though the answer is rather "no".  But I must admit
> I do not understand what C-a c does, neither M-up and M-down.  Does
> C-a c copy the input-output combination into the kill ring?

I think that's right - I'm not sure where it stuck it but that rings a
bell.

> Maybe it would be possible to have a simple "magic-key-combination"
> that allows editing of output?  Or, maybe even better: that turns
> the cursor into a brush and marks the character under point blue if
> I hit space?  (That would be really useful for me!)

It's a good idea, but I'm not quite sure what Emacs lets you get away
with for things like that.  Certain it can be looked into.

>  To actually edit output, copying it into a separate buffer may be
> the wiser choice anyway.

That was my feeling, but selective color changing/highlighting doesn't
change the output and still lets you direct the eye - I like it.

> * I'd really like to have an easy way to send definitions etc. from
> one buffer (containing a .input file) to an axiom process.  There
> are two (small) problems with that:
> 
>   1) I think it's difficult to detect where a function definition
> starts and where it ends.  The MuPAD people told me that the would
> simple mark the definition boundaries themselves with some special
> characters.  I think, that's quite wise.  Maybe -- << and -- @ would
> make sense? (They should be acceptable as comments, since I will
> often want to )read the input file, too.  Another shortcut that
> sends only the current line, and possible a third shortcut that
> sends the current region would also be good.

The line and region I like, but as you note trying to automatically
identify the function is tough.  Lisp can get away with it thanks to
the parenthesis, IIRC, but for other languages it's a problem.  I'm not
real keen on the special characters but if there's great demand for it
I suppose it could be done.
 
>   2) When you try to yank a definition into the axiom buffer, you'll
> be disappointed, because the axiom interpreter doesn't seem to like
> multiline input, does he?  So, maybe you would need to save the input
> which should be sent to axiom into a temporary file and )read it.

I think that came up last time, and the only workable solution was
something like that.  (Yuck.)

I'm not really that crazy about Emacs based on my experiences to date
with it, but I will admit I don't know of anything better for this kind
of work.

The pamphlet file is in pretty sad shape, so I guess that needs to be
fixed, but I was under the impression there was very little interest in
going further with this work and it did what I needed ;-).  That can be
revisited if we want to do more with it.

Cheers,
CY


       
____________________________________________________________________________________Need
 a vacation? Get great deals
to amazing places on Yahoo! Travel.
http://travel.yahoo.com/