# [NTG-context] How to break laws in physics?

Hans Hagen pragma at wxs.nl
Mon Jan 9 19:39:52 CET 2006

Mojca Miklavec wrote:

>\definetypeface [iwona] [mm] [math] [iwona] [default]
>\setupbodyfont[iwona]
>
>${\partial\rho\over\partial t}+\vec{\nabla}\vec{\jmath}=0$
>
>(If I fail on the exam, I'll try to explain to professor that I found
>a few mistakes in the Maxwell equations and lost the whole time trying
>to fix them before I was able to learn further :)
>
>
ok, so  it's may fault that you failed that examn ... if so ... at
eurotex/bachotex i'll buy you a beer

>
>
>I've found three "kind-of-bugs" in font itself (\jmath and \imath were
>missing, \partial is mirrored, vectors placed strangely - the latter
>may not be connected with the font itself).
>
>
can you communicate this with J&J ...

>So I tried to redefine \jmath to use dotlessj from italic version of
>font and found another "bug" in ConTeXt. dotlessj is commented out in
>enco-ec.tex. Well, I understand that it may not be present in every
>font, but it's worse: it's already defined in the default encoding, so
>\dotlessj results in double quotes, which is worse than no char at
>all. enco-ec has some other faults present, I tried to fix a few of
>them ([t|s]cedilla should be [t|s]commaaccent, \Eth and \eth were
>commented out, while they are OK, one glyph is present twice, I
>redefined it to map to the first slot, since usually the second one is
>empty, upper/lower casing may still need some revision).
>
>Is there any chance that any numerical definitions of characters
>disappear from enco-def slowly? If anyone is using any other encoding
>than the standard one, the characters map to something completely
>weird then.
>
>
actually all chars in enco-def that use funny numbers should be present
in a (math) font vector; better a wrong something that nothing (we need
a fallback)

the alternative is to have a def which is mostly mapping to nothing and
then make the tex encoding on top of that (quite an effort and it may
temporary introduce bugs)

>Some fixes for enco-def:
>
>% DELETE THIS ONE: IT'S WORSE THAN NOTHING
>\definecharacter eth              {\dstroke} % wrong but better that nothing
>% and replace by:
>\definecharacter eth              {\unknownchar}
>
>
ok, gone are the dstrokes (only you use them -)

>\definecharacter scommaaccent     {\buildtextbottomcomma s}
>\definecharacter Scommaaccent     {\buildtextbottomcomma S}
>\definecharacter tcommaaccent     {\buildtextbottomcomma t}
>\definecharacter Tcommaaccent     {\buildtextbottomcomma T}
>
>\definecharacter Ohorndotbelow       {\buildtextbottomdot\Ohorn}
>\definecharacter ohorndotbelow       {\buildtextbottomdot\ohorn}
>\definecharacter Uhorndotbelow       {\buildtextbottomdot\Uhorn}
>\definecharacter uhorndotbelow       {\buildtextbottomdot\uhorn}
>
>

>enco-ini:
>
>Any chances to add something like:
>\definecharacter bottomcomma {,} % comma, lowered a bit
>\def\buildtextbottomcomma{\bottomaccent{.25ex}{0}{5}{\textbottomcomma}}
>
>I have no idea what \textbottomcomma should be, but probably just a
>comma, lowered a bit. The current definition with ' is a bit strange.
>
>
hm, so you suppose that i have an idea ...

\definetypeface[xxx][rm][serif][modern][default][encoding=qx]

{\xxx \scommaaccent}

\definecharacter textbottomcomma  {\hbox{\lower.35ex\hbox{\tx,}}}

\definecharacter scommaaccent     {\buildtextbottomcomma s}

\def\buildtextbottomcomma{\bottomaccent{.15ex}{0}{5}{\tx,}}

\scommaaccent \textbottomcomma

\stoptext

Hans