On Nov 16, 2005, at 16:15, Hans van der Meer wrote:

> I am trying to define bold math myself, but encounter a problem  
> with the typescripts.

As always, when desperately posting the problem, something suggests  
itself that brings a solution of some sort.

I looked into the definition of \bfmath, traced its execution and  
found out that afterall my own [bfmath] typescript was indeed used!  
But in order to get the same behaviour between normal math (variables  
set in MathItalic) and bold math I had to change the definition of  
\bfmath:

\def\bfmath{\bf\mf\synchronizesymb\mi}

That is, the last font switch made to \mi instead of \mf as is done  
font-ini.tex.
Then I see the same behaviour in math and bfmath, that is MathItalic  
with $etc$ and MathItalicBold with $\bfmath etc$.

My question to Hans Hagen: Should this change be followed in font- 
ini.tex?
Personally, I feel confused by having math and bold math behaving  
differently.

Hans van der Meer

>
>
> Further probing into this matter (previous mail, see below) shows  
> that the typescript delineated by ****'s is NOT USED AT ALL!
> (Calling \setupformulas[method=bold] makes no difference).
> It therefore does not seem a problem in the math family assignments  
> in font-bfm.tex, as I first suspected.
>
> Is there some special invocation needed for hooking up a user  
> defined font set into $\bfmath ...$?
> If have looked into font-bfm.tex  and font-ini.tex but could not  
> figure it out.
> Clearly I am misunderstanding the inner workings here, but I have  
> no clue in which direction to look.
>
>> I am using the follwing (extracted) typescript lines for math in  
>> the lucida fonts:
>>
>> \starttypescript [luc]
>>     \definetypeface [luc][mm][math][lucidayy][default] 
>> [encoding=texnansi]
>>     \definetypeface [luc][mm][bfmath][lucidayy][default] 
>> [encoding=texnansi]
>> \stoptypescript
>>
>> \starttypescript [math] [lucidayy] [default]
>>     \definebodyfont [default] [mm]
>>       [mr=MathRoman mo 1, mi=MathItalic mo 1, ....]
>> \stoptypescript
>>
> ********************************************************************** 
> ********
>> \starttypescript [bfmath] [lucidayy] [default]
>>     \definebodyfont [default] [mm]
>>       [mrbf=MathRomanBold mo 1, mibf=MathItalicBold mo 1, ....]
>> \stoptypescript
> ********************************************************************** 
> ********
>
>> \starttypescript [math] [lucidayy] [name]
>>   \definefontsynonym [MathRoman]         [LucidaBright]
>>   \definefontsynonym [MathItalic]        [LucidaNewMath-AltItalic]
>> \stoptypescript
>>
>> \starttypescript [math] [lucidayy] [texnansi,ec,8r]
>>   \definefontsynonym [LucidaBright]                [hlhr8y]    
>> [encoding=texnansi]
>>   \definefontsynonym [LucidaNewMath-AltItalic]     [hlcrima]
>> \stoptypescript
>>
>> \starttypescript [bfmath] [lucidayy] [name]
>>   \definefontsynonym [MathRomanBold]     [LucidaBright-Demi]
>>   \definefontsynonym [MathItalicBold]    [LucidaNewMath- 
>> AltDemiItalic]
>> \stoptypescript
>>
>> \starttypescript [bfmath] [lucidayy] [texnansi,ec,8r]
>>   \definefontsynonym [LucidaBright-Demi]           [hlhb8y]    
>> [encoding=texnansi]
>>   \definefontsynonym [LucidaNewMath-AltDemiItalic] [hlcdima]
>> \stoptypescript
>>
>> Typesetting:
>> \starttext
>> \usetypescript[luc][texnansi]\setupbodyfont[luc,10pt]\lucidatrue
>> {$\sqrt{a^2 + b^2} = c + \mf\sin{}x\quad {\ma NZRC}$}\crlf
>> \stoptext
>>  Gives me for the variables neatly LucidaNewMath-AltItalic.
> *** Changing the fonts in the above typescript does work ***
>>
>> But typesetting
>> \starttext
>> \usetypescript[luc][texnansi]\setupbodyfont[luc,10pt]\lucidatrue
>> {$\bfmath\sqrt{a^2 + b^2} = c + \mf\sin{}x\quad {\ma NZRC}$}\crlf
>> \stoptext
>> Gives me LucidaBright-Demi instead of LucidaNewMath-AltDemiItalic.
>> Everything else however (even the blackletters from ma) are in the  
>> correct bold math font.
>> The font must be present, because substituting MathItalicBold for  
>> MathItalic in the [math] script proves that.
> *** But not coming from my typescript I conclude from several  
> experiments ***
>
> Hans van der Meer
> _______________________________________________
> ntg-context mailing list
> ntg-context@ntg.nl
> http://www.ntg.nl/mailman/listinfo/ntg-context
>