Re: \bar missing in euler

[Aditya Mahajan <adityam@umich.edu>]

On Thu, 6 Jul 2006, Adam Lindsay wrote:

> Aditya Mahajan wrote:
>
>> gives just "a". I do not get any bar over "a".
>
> Yup. You seem to be the first to have noticed (or complained), as well.
>
>> This is how eulervm.sty defines these characters for T1 encoding.
>> ...
>
> That's the key, though, it apparently pulls in the text accents as well,
> and changes the math definitions depending on the text encoding. That
> seems rather opposite to how ConTeXt *should* work.
>
>> I do not know what is the equivalent of 'operators' family in context.
>> Any idea how to pull them out of the text fonts?
>
> Let's open this up to Hans and Taco: any idea? Much of the finer points
> of \definemathsymbol et al are voodoo to me.
>
> The fourth/sixth arguments of \definemathsymbol insist on a (hex) number
> to be in that slot. Anything else (like \textmacron) is a no-go.
>
> So right now, Euler is limited by the math mechanisms/the encoding of
> the font. I can imagine two "solutions" of differing hack-ish natures:
>
> 1) Allow further expansion in \definemathsymbol:
>  \definemathsymbol [breve] [accent] [tf] [\textbreve]
>
> 2) Do a hack following LaTeX's imperfect lead:
>  \definemathsymbol [breve] [accent] [tf] ["15] % for texnansi
>   % or
>  \definemathsymbol [breve] [accent] [tf] ["08] % for ec
>  etc.

I have no knowledge of how Context's font stuff works. The first 
definition looks better to me, as one would not need to worry about 
the encoding while defining symbols (one needs to worry about the 
encoding while defining \textbreve etc, anyway). Though the second 
does have the advantage that it works right away, so there is no need 
to change the definition of \definemathsymbol.

Right now I will use

\definemathsymbol [bar] [accent] [tf] ["09]

which seems to work with my present setup.

Thank you.

Aditya