input method for unicode math characters

input method for unicode math characters

[Roland Winkler]

Recently I discovered for myself that sometimes it can be convenient
to use directly the many unicode math characters in text documents
(i.e., non-LaTeX documents, which is otherwise my first choice).
Yet up to now this is a bit clumsy in emacs because (to the best of
my knowledge) there is no particular support for these characters,
say via some handy input method. Is this something anybody has
already thought about? Or am I missing something here?

Roland

Re: input method for unicode math characters

[Antoine Levitt]

30/09/11 11:32, Roland Winkler
> Recently I discovered for myself that sometimes it can be convenient
> to use directly the many unicode math characters in text documents
> (i.e., non-LaTeX documents, which is otherwise my first choice).
> Yet up to now this is a bit clumsy in emacs because (to the best of
> my knowledge) there is no particular support for these characters,
> say via some handy input method. Is this something anybody has
> already thought about? Or am I missing something here?
>
> Roland

There's C-x 8 RET. For instance C-x 8 RET double integral produces ∬.
Or do you mean some other mode like latex-math-abbrev ?

Re: input method for unicode math characters

[Eli Zaretskii]

> Date: Fri, 30 Sep 2011 04:32:18 -0500
> From: "Roland Winkler" <winkler@gnu.org>
> 
> Recently I discovered for myself that sometimes it can be convenient
> to use directly the many unicode math characters in text documents
> (i.e., non-LaTeX documents, which is otherwise my first choice).
> Yet up to now this is a bit clumsy in emacs because (to the best of
> my knowledge) there is no particular support for these characters,
> say via some handy input method. Is this something anybody has
> already thought about? Or am I missing something here?

Did you try "C-u C-\ TeX RET"?

Re: input method for unicode math characters

[Roland Winkler]

On Fri Sep 30 2011 Antoine Levitt wrote:
> There's C-x 8 RET. For instance C-x 8 RET double integral produces
> â ¬.

Thanks, well, the problem with this is that one needs to know these
names

> Or do you mean some other mode like latex-math-abbrev ?

...and here it's kind of similar with zillions of commands

On Fri Sep 30 2011 Eli Zaretskii wrote:
> Did you try "C-u C-\ TeX RET"?

...while here somehow I get completely irritated by the minibuffer
that pops up (or is it the message area? oh well!)
Probably this can be customized. But then how to keep track of the
available possibilities?

-- In any case: Thanks! I just started myself thinking about the
question what I might find useful here. I'd like to have something
that interrupts my flow of typing as little as possible, while at
the same time it's not easy to keep track of the plethora of symbols
that are available so that some kind of guidance comes handy. An
"acceptable" solution for these contradicting requirements is
certainly a matter of taste.

To begin with, I now have some approaches to play with.

Roland

Re: input method for unicode math characters

[Eli Zaretskii]

> Date: Fri, 30 Sep 2011 16:35:17 -0500
> From: "Roland Winkler" <winkler@gnu.org>
> Cc: emacs-devel@gnu.org
> 
> On Fri Sep 30 2011 Eli Zaretskii wrote:
> > Did you try "C-u C-\ TeX RET"?
> 
> ...while here somehow I get completely irritated by the minibuffer
> that pops up (or is it the message area? oh well!)

You asked for an input method.  _All_ Emacs input methods show
possible combinations in the echo area.  For that matter, any
non-Emacs input method with complex combinations will pop up a small
window with these possibilities.

IOW, if you want an input method, you should be prepared to deal with
these pop-ups.

OTOH, if you use TeX a lot, you should be able to type the likes of
\gamma without ever looking at the echo area.

Re: input method for unicode math characters

[Thierry Volpiatto]

"Roland Winkler" <winkler@gnu.org> writes:

> On Fri Sep 30 2011 Antoine Levitt wrote:
>> There's C-x 8 RET. For instance C-x 8 RET double integral produces
>> â ¬.
>
> Thanks, well, the problem with this is that one needs to know these
> names
I wrote today an anything command for that. (i.e 𝕦𝕔𝕤 𝕔𝕠𝕞𝕡𝕝𝕖𝕥𝕚𝕠𝕟 ⚡)
See http://repo.or.cz/w/anything-config.git

You can write with all these characters without quitting.

>> Or do you mean some other mode like latex-math-abbrev ?
>
> ...and here it's kind of similar with zillions of commands
>
> On Fri Sep 30 2011 Eli Zaretskii wrote:
>> Did you try "C-u C-\ TeX RET"?
>
> ...while here somehow I get completely irritated by the minibuffer
> that pops up (or is it the message area? oh well!)
> Probably this can be customized. But then how to keep track of the
> available possibilities?
>
> -- In any case: Thanks! I just started myself thinking about the
> question what I might find useful here. I'd like to have something
> that interrupts my flow of typing as little as possible, while at
> the same time it's not easy to keep track of the plethora of symbols
> that are available so that some kind of guidance comes handy. An
> "acceptable" solution for these contradicting requirements is
> certainly a matter of taste.
>
> To begin with, I now have some approaches to play with.
>
> Roland
>
>

-- 
A+ Thierry
Get my Gnupg key:
gpg --keyserver pgp.mit.edu --recv-keys 59F29997 

Re: input method for unicode math characters

[Stefan Monnier]

>> There's C-x 8 RET. For instance C-x 8 RET double integral produces ∬.
> Thanks, well, the problem with this is that one needs to know these names.

You can use completion to find it.  I personally find that substring
completion is very useful for it, so I have

(when (boundp 'completion-category-overrides)
  (add-to-list 'completion-category-overrides
               '(unicode-name (styles basic substring))))

> ...while here somehow I get completely irritated by the minibuffer
> that pops up (or is it the message area? oh well!)
> Probably this can be customized. But then how to keep track of the
> available possibilities?

I don't find the echo area guidance very useful for TeX, indeed, but
it's not like the guidance bothers me: I just ignore it.  This said, it
would be good to improve it for such circumstances (and the TAB
"completion" should work more like normal completion: e.g. after "\Rig"
a TAB should complete to \Rightarrow since it's the only sequence that
starts with \Rig; also showing the tree in *Quail Completions* is not
nearly as useful as showing the leaves in *Completions*).


        Stefan

Re: input method for unicode math characters

[Roland Winkler]

On Sat Oct 1 2011 Stefan Monnier wrote:
> >> There's C-x 8 RET. For instance C-x 8 RET double integral produces  .
> > Thanks, well, the problem with this is that one needs to know these names.
> 
> You can use completion to find it.  I personally find that substring
> completion is very useful for it, so I have
> 
> (when (boundp 'completion-category-overrides)
>   (add-to-list 'completion-category-overrides
>                '(unicode-name (styles basic substring))))

Thanks, I'll give it a try.

> I don't find the echo area guidance very useful for TeX, indeed,
> but it's not like the guidance bothers me: I just ignore it. This
> said, it would be good to improve it for such circumstances

I'd appreciate that!

There are things popping up in the message area that I do not want
to ignore. And there are other things that one can ignore without a
problem. Yet I only know which category a message belongs to if I do
pay some attention to what is happening in the message area. So the
bottom line is that too often one can get distracted for nothing.

Roland