\boldsymbol (m-math)

\boldsymbol (m-math)

[AlterEgo Qasars]

Good evening.

I need in my documents \boldsymbol macro present in amslatex. It typesets
mathematical symbols in bold face (e. g. \beta). Some time ago I used
m-math module (version version=1999.9.10). But now it is not possible -- 
there is some deadlock in the code probably. I can't find a newer version. 
Is it somewhere? Or is it handled in another way?

Many thanks for your help.
Michal Kvasnicka

Re: \boldsymbol (m-math)

[Taco Hoekwater]

On Mon, 14 Jan 2002 12:52:57 +0100
"AlterEgo Qasars" <quasar@econ.muni.cz> wrote:

> Good evening.
> 
> I need in my documents \boldsymbol macro present in amslatex. It typesets
> mathematical symbols in bold face (e. g. \beta). Some time ago I used
> m-math module (version version=1999.9.10). But now it is not possible -- 
> there is some deadlock in the code probably. I can't find a newer version. 
> Is it somewhere? Or is it handled in another way?

Since the new font selection scheme in context was introduced the math 
module has been broken. I've been too busy to look into the changes
that are required. With some luck, I may be able to do something about
it next week. Until then, you are on your own. Sorry.

-- 
groeten,

Taco

Re[2]: \boldsymbol (m-math)

[Giuseppe Bilotta]

TH> Since the new font selection scheme in context was introduced the math 
TH> module has been broken. I've been too busy to look into the changes
TH> that are required. With some luck, I may be able to do something about
TH> it next week. Until then, you are on your own. Sorry.

My limited testings in trying to make the m-math module work in
modern ConTeXt led to this patch. Find

\def\docompletefontswitch[#1]%
  {\bodyfontsize=#1\relax
   \dimensiontocount\bodyfontsize\bodyfontpoints
   \edef\bodyfontfactor{\withoutpt\the\bodyfontsize}%
   \normalizebodyfontsize\bodyfontsize\to\normalizedbodyfontsize
   \edef\textface{\getvalue{\??ft\normalizedbodyfontsize\s!text}}%
   \edef\scriptface{\getvalue{\??ft\normalizedbodyfontsize\s!script}}%
   \edef\scriptscriptface{\getvalue{\??ft\normalizedbodyfontsize\s!scriptscript}}%
   \settextfonts
   \setmathfamilies\fontstyle
   \rmtf
   \the\everybodyfont}  

and change it to:

\def\docompletefontswitch[#1]%
  {\bodyfontsize=#1\relax
   \dimensiontocount\bodyfontsize\bodyfontpoints
   \edef\bodyfontfactor{\withoutpt\the\bodyfontsize}%
   \normalizebodyfontsize\bodyfontsize\to\normalizedbodyfontsize
   \dosetbodyfontface \textface         \s!text
   \dosetbodyfontface \scriptface       \s!script
   \dosetbodyfontface \scriptscriptface \s!scriptscript
   \settextfonts
   \setmathfamilies\fontstyle
   \rmtf
   \the\everybodyfont}

which should allow it to load.

Re[2]: \boldsymbol (m-math)

[Hans Hagen]

At 09:45 PM 1/17/2002 +0100, Giuseppe Bilotta wrote:

>TH> Since the new font selection scheme in context was introduced the math
>TH> module has been broken. I've been too busy to look into the changes
>TH> that are required. With some luck, I may be able to do something about
>TH> it next week. Until then, you are on your own. Sorry.

some code from the math module is now present in the core; i would prefer 
to see only the fancy math breaking to be in the module, and the rest can 
be core functionality (i can do that if i have code snippets and examples 
i.e. understand what the commands are supposed to do; i can then also take 
care of configurability]

! ! ! see m-newmat for the things to do

>My limited testings in trying to make the m-math module work in
>modern ConTeXt led to this patch. Find

.....

those are dangerous patches, esp since they may slow down font switching 
significantly;

if you look into the font-ini file, you will see some math sync code; 
actually, math kind of auto syncs with everymath

Hans

-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------

Re[3]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Sunday, January 20, 2002 Hans Hagen wrote:

HH> some code from the math module is now present in the core; i would prefer
HH> to see only the fancy math breaking to be in the module, and the rest can 
HH> be core functionality (i can do that if i have code snippets and examples 
HH> i.e. understand what the commands are supposed to do; i can then also take 
HH> care of configurability]

HH> ! ! ! see m-newmat for the things to do

Nowhere to be found here. Is a new ConTeXt out? Date?

For the code snippets and their use you can have a look at the
amsmath LaTeX package: the documentation comes with examples, and
the source code (.dtx file) is latex-able to show you the code and
some comments from the coder. (Nothing on the level of a ConTeXt
compiled module, of course ...)

HH> those are dangerous patches, esp since they may slow down font switching 
HH> significantly;

I just put them in to be able to load the module when doing some
testing. I don't use it by now.

--
Giuseppe "Oblomov" Bilotta

Re: Re[3]: \boldsymbol (m-math)

[Taco Hoekwater]

On Mon, 21 Jan 2002 16:39:25 +0100
"Giuseppe Bilotta" <bourbaki@bigfoot.com> wrote:

> For the code snippets and their use you can have a look at the
> amsmath LaTeX package: the documentation comes with examples, and
> the source code (.dtx file) is latex-able to show you the code and
> some comments from the coder. (Nothing on the level of a ConTeXt
> compiled module, of course ...)

That LaTeX code is (unfortunately) totally unusable, unless you 
write a t-latexcore.tex first. 

-- 
groeten,

Taco

Re[5]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Tuesday, January 22, 2002 Taco Hoekwater wrote:

TH> On Mon, 21 Jan 2002 16:39:25 +0100
TH> "Giuseppe Bilotta" <bourbaki@bigfoot.com> wrote:

>> For the code snippets and their use you can have a look at the
>> amsmath LaTeX package: the documentation comes with examples, and
>> the source code (.dtx file) is latex-able to show you the code and
>> some comments from the coder. (Nothing on the level of a ConTeXt
>> compiled module, of course ...)

TH> That LaTeX code is (unfortunately) totally unusable, unless you 
TH> write a t-latexcore.tex first. 

Who, me? Naaah

Yes, I expected the code to be non-recyclable. We'd need someone
to know the LaTeX kernel good enough to be able to translate LaTeX
code into ConTeXt code.

--
Giuseppe "Oblomov" Bilotta

Re[5]: \boldsymbol (m-math)

[Hans Hagen]

At 10:40 AM 1/22/2002 +0100, Giuseppe Bilotta wrote:

>Tuesday, January 22, 2002 Taco Hoekwater wrote:
>
>TH> On Mon, 21 Jan 2002 16:39:25 +0100
>TH> "Giuseppe Bilotta" <bourbaki@bigfoot.com> wrote:
>
> >> For the code snippets and their use you can have a look at the
> >> amsmath LaTeX package: the documentation comes with examples, and
> >> the source code (.dtx file) is latex-able to show you the code and
> >> some comments from the coder. (Nothing on the level of a ConTeXt
> >> compiled module, of course ...)
>
>TH> That LaTeX code is (unfortunately) totally unusable, unless you
>TH> write a t-latexcore.tex first.
>
>Who, me? Naaah
>
>Yes, I expected the code to be non-recyclable. We'd need someone
>to know the LaTeX kernel good enough to be able to translate LaTeX
>code into ConTeXt code.

no need for that, i will just reimplement the things needed using low level 
context things as well as add configrability (already done with a couple of 
things).

concerning the formula breaking stuff, taco has done that without the need 
for low level latex code, since it is kind of self contained;

so, the main things for me is to have isolated ams-code snippets for the 
things that should go into the core [take a look at newmat to see what 
already is moved/redone]

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------

Re: Re[5]: \boldsymbol (m-math)

[Taco Hoekwater]

Giuseppe said:
> TH> That LaTeX code is (unfortunately) totally unusable, unless you 
> TH> write a t-latexcore.tex first. 
> 
> Who, me? Naaah
> 
> Yes, I expected the code to be non-recyclable. We'd need someone
> to know the LaTeX kernel good enough to be able to translate LaTeX
> code into ConTeXt code.

There is also another thing. Amsmath defines a large amount of
environments, and I don't see why we should mimic that in
context. A command like:

\startdisplaymath[<label>][type=eqalign,location=left,number=no]
.....
\stopdisplaymath

should be enough, yes?

-- 
groeten,

Taco

Re[7]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Tuesday, January 22, 2002 Taco Hoekwater wrote:

TH> There is also another thing. Amsmath defines a large amount of
TH> environments, and I don't see why we should mimic that in
TH> context. A command like:

TH> \startdisplaymath[<label>][type=eqalign,location=left,number=no]
TH> .....
TH> \stopdisplaymath

TH> should be enough, yes?

Yes and no. In principle, yes. From a usability standpoint, no: if
you repeatedly use two or three of such types, you'd like to have
\starteqalign, \startsplit \startwhateverelse without having to
specify the type each time. So the answer is yes, provided the
user can define his own new displaymath environments with
different predefined options. Something on the line of itemgroups
vs itemize.

And one more thing, while we're at it: I think ConTeXt should
provide a quick way to choose between two alternatives of the same
environment, something on the line of the starred
commands/environments in LaTeX. For example, if I use both
numbered and unnumbered equations, in LaTeX I can differentiate by
using a * in the name (or not using it).

It would be nice if ConTeXt provided for a similar thing, letting
the user choose which alternative matches what

If I had for example:

\definestartstop[whatever][options]

and then I had:

\setupdefaultwhatever[option=value]
\setupalternativewhatever[option=other value]

it would allow me to use

\startwhatever
This has the default options
\stopwhatever

\startwhatever*
This has the alternative options
\stopwhatever

What do you think?

Re[7]: \boldsymbol (m-math)

[Hans Hagen]

At 06:17 PM 1/22/2002 +0100, Giuseppe Bilotta wrote:

>using a * in the name (or not using it).
>
>It would be nice if ConTeXt provided for a similar thing, letting
>the user choose which alternative matches what
>
>If I had for example:
>
>\definestartstop[whatever][options]
>
>and then I had:
>
>\setupdefaultwhatever[option=value]
>\setupalternativewhatever[option=other value]
>
>it would allow me to use
>
>\startwhatever
>This has the default options
>\stopwhatever
>
>\startwhatever*
>This has the alternative options
>\stopwhatever
>
>What do you think?

- for sure one (esp you i guess) would like multiple alternatives then 
\startgb***
- and combinations of options, so \startgb*@#%%!()
- i hate the *'d things

why not use the existing:

\definestartstop[whatever,whateverx][style=bold,before=\blank]
\setupstartstop[whateverx][style=slanted]

[inheritance of start-stops is probably an alternative]

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------

Re[8]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Wednesday, January 23, 2002 Hans Hagen wrote:

>>What do you think?

HH> - for sure one (esp you i guess) would like multiple alternatives then 
HH> \startgb***
HH> - and combinations of options, so \startgb*@#%%!()
HH> - i hate the *'d things

HH> why not use the existing:

HH> \definestartstop[whatever,whateverx][style=bold,before=\blank]
HH> \setupstartstop[whateverx][style=slanted]

HH> [inheritance of start-stops is probably an alternative]

Yes, start-stop inheritance would be a nice way.

A question: if I define and set up whateverx to have
option=oldvalue, can I do

\startwhateverx[option=changedonce]

?

--
Giuseppe "Oblomov" Bilotta

Re[8]: \boldsymbol (m-math)

[Hans Hagen]

At 01:09 PM 1/23/2002 +0100, Giuseppe Bilotta wrote:

>Wednesday, January 23, 2002 Hans Hagen wrote:
>
> >>What do you think?
>
>HH> - for sure one (esp you i guess) would like multiple alternatives then
>HH> \startgb***
>HH> - and combinations of options, so \startgb*@#%%!()
>HH> - i hate the *'d things
>
>HH> why not use the existing:
>
>HH> \definestartstop[whatever,whateverx][style=bold,before=\blank]
>HH> \setupstartstop[whateverx][style=slanted]
>
>HH> [inheritance of start-stops is probably an alternative]
>
>Yes, start-stop inheritance would be a nice way.
>
>A question: if I define and set up whateverx to have
>option=oldvalue, can I do
>
>\startwhateverx[option=changedonce]

i can provide that feature, but i think it's not what you want: it would 
add typesetting directives to the text stream while actually one wants to 
separete setup and content

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------

Re[9]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Thursday, January 24, 2002 Hans Hagen wrote:
>>
>>A question: if I define and set up whateverx to have
>>option=oldvalue, can I do
>>
>>\startwhateverx[option=changedonce]

HH> i can provide that feature, but i think it's not what you want: it would 
HH> add typesetting directives to the text stream while actually one wants to 
HH> separete setup and content

Only partly true. If whatevers have, say, six options, (option1,
option2, ... option6) and I use mostly two alternatives (whatever,
whateverx) where each has option1 set to a different value, I may
still want to change any of the other (5) options for one
equation, even assuming that it's relative to markup only.

--
Giuseppe "Oblomov" Bilotta

Re: Re[7]: \boldsymbol (m-math)

[Taco Hoekwater]

On Wed, 23 Jan 2002 12:32:54 +0100
"Hans Hagen" <pragma@wxs.nl> wrote:

> \definestartstop[whatever,whateverx][style=bold,before=\blank]
> \setupstartstop[whateverx][style=slanted]
> 
> [inheritance of start-stops is probably an alternative]

I agree with Hans that the starred forms are weird and ugly, and 
appending eg. an x sounds fine to me. It is even possible
to create something similar to \startitemize (with [X] options),
but that is a bit overkill IMO.

For the new math, I think this scheme is the most sensible:

  \startmathdisplay[<ref>][<options>]
  ...
  \stopmathdisplay

  \setupmathdisplay[<options>]

or another name, but this sounds reasonable. This is the core
command that does all of the 'real work'.

Then we should also have:

  \definemathdisplay[whatever][<options>]

which creates

  \setupwhatever[<options>]

and

  \startwhatever[<ref>][<overruled options>]
  ...
  \stopwhatever

Agreed? (will think about possible options later)

-------

Besides \formulanumber, we also need \subformulanumber.

-------

And then there is (still) the problem of pronunciations.

-- 
groeten,

Taco

Re[9]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Wednesday, January 23, 2002 Taco Hoekwater wrote:

TH> For the new math, I think this scheme is the most sensible:

TH>   \startmathdisplay[<ref>][<options>]
TH>   ...
TH>   \stopmathdisplay

TH>   \setupmathdisplay[<options>]

TH> or another name, but this sounds reasonable. This is the core
TH> command that does all of the 'real work'.

TH> Then we should also have:

TH>   \definemathdisplay[whatever][<options>]

TH> which creates

TH>   \setupwhatever[<options>]

TH> and

TH>   \startwhatever[<ref>][<overruled options>]
TH>   ...
TH>   \stopwhatever

TH> Agreed? (will think about possible options later)

Yes. Exactly what I was looking for.

TH> -------

TH> And then there is (still) the problem of pronunciations.

Uh?

--
Giuseppe "Oblomov" Bilotta

Re: Re[9]: \boldsymbol (m-math)

[Taco Hoekwater]

On Wed, 23 Jan 2002 17:30:29 +0100
"Giuseppe Bilotta" <bourbaki@bigfoot.com> wrote:

> 
> TH> And then there is (still) the problem of pronunciations.
> 
> Uh?

\starttheorem & co.

-- 
groeten,

Taco

Re[11]: \boldsymbol (m-math)

[Giuseppe Bilotta]

Thursday, January 24, 2002 Taco Hoekwater wrote:

TH> On Wed, 23 Jan 2002 17:30:29 +0100
TH> "Giuseppe Bilotta" <bourbaki@bigfoot.com> wrote:

>> 
>> TH> And then there is (still) the problem of pronunciations.
>> 
>> Uh?

TH> \starttheorem & co.

Oh. Well, one can use enumerations for them, the only problem
being that one cannot add optional text in the header (say, the
author name or theorem title). I was thinking aboud adapting my
xdesc module for this.

\starttheorem
This is a standard, unnamed theorem
\stoptheorem

\starttheorem[author=Lebesgue]
This is Lebesgue's theorem
\stoptheorem

\starttheorem[topic={total differential}]
This is the Total differential theorem.
\stoptheorem

\starttheorem[author=Cantor,topic={Uniform continuity}]
This theorem has both and author and a topic
\stoptheorem

The xdesc module allows to choose which (optional) entries to
allow in the header and how to typeset the header depending on
which are present, but xdescs miss a few things like referencing
and such.

--
Giuseppe "Oblomov" Bilotta

Re: Re[7]: \boldsymbol (m-math)

[Johannes H?sing]

On Wed, Jan 23, 2002 at 02:43:07PM +0100, Taco Hoekwater wrote:

[...]
> For the new math, I think this scheme is the most sensible:
> 
>   \startmathdisplay[<ref>][<options>]
>   ...
>   \stopmathdisplay
> 

This would not allow for multi-equation displays where each equation has
a separate and ref'able number.

> 
> Besides \formulanumber, we also need \subformulanumber.
> 

Or are you meaning the same here?

Greetings

Johannes
-- 
Johannes Hüsing   There is something fascinating about science. One gets
hannes@ruhrau.de  such wholesale returns of conjecture from such a 
                  trifling investment of fact.                Mark Twain

Re: Re[7]: \boldsymbol (m-math)

[Taco Hoekwater]

On Wed, 23 Jan 2002 21:29:38 +0100
"Johannes H?sing" <hannes@ruhrau.de> wrote:

> Or are you meaning the same here?

I believe that the m-math module did something like this
(pseudo-code, doesn't actually work!!):

\def\formulanumber[#1]{%
 \makethenumber[formula]
 \ifargument#1
   % set extra reference.
   \reference[#1]{\thenumber}%
 \else
   % 
   \ifformularefdonealready % by \startdisplay perhaps
   \else
     \reference[\inheritedvalue]{\thenumber}%
     \formularefdonealreadytrue
    \fi
 \fi
 \placetheformulanumber
 \incrementtheformulanumber
}

\startdisplaymath[AA]
a = b \formulanumber     \\        % (1), with ref AA
c = d \formulanumber     \\        % (2), no ref
d = e                    \\        % nothing
i = j \formulanumber[BB] \\        % (3), with ref BB
\stopdisplaymath

That should be enough, yes?

-- 
groeten,

Taco

Re: Re[5]: \boldsymbol (m-math)

[Hans Hagen]

At 01:46 PM 1/22/2002 +0100, Taco Hoekwater wrote:

>There is also another thing. Amsmath defines a large amount of
>environments, and I don't see why we should mimic that in
>context. A command like:
>
>\startdisplaymath[<label>][type=eqalign,location=left,number=no]
>.....
>\stopdisplaymath
>
>should be enough, yes?

right, actually, there is already

   \defineformulaalternative

(see core-mat.tex)

some 2+4 types are predefined, like:

% sp = single line paragraph  sd = single line display
% mp = multi  line paragraph  md = multy  line display

is is no problem to extend this mechanism

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------

Re[3]: \boldsymbol (m-math)

[Hans Hagen]

At 04:39 PM 1/21/2002 +0100, Giuseppe Bilotta wrote:

>Sunday, January 20, 2002 Hans Hagen wrote:
>
>HH> some code from the math module is now present in the core; i would prefer
>HH> to see only the fancy math breaking to be in the module, and the rest can
>HH> be core functionality (i can do that if i have code snippets and examples
>HH> i.e. understand what the commands are supposed to do; i can then also 
>take
>HH> care of configurability]
>
>HH> ! ! ! see m-newmat for the things to do
>
>Nowhere to be found here. Is a new ConTeXt out? Date?

i'll send you that file

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
  tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------
                                   fall-back web server: 
www.pragma-pod.nl
-------------------------------------------------------------------------