Re: Bug in enco-ec.tex and enco-fde.tex

[Tobias Burnus <burnus@gmx.de>]

Hallo Uwe, Hraban en/und Hans,

Uwe Koloska wrote:
> there are some errors in the files
>   tex/context/base/enco-ec.tex
>   tex/context/base/enco-fde.tex
> 
> I am sorry because this errors are caused by me thinking only half the way.
> 
> AFAIK the upper case of (german) szligature is SS and this character is
> supposed to be 223 in ec-encoding.  So the proposed changes are:
> 
> tex/context/base/enco-ec.tex
>         \definecharacter SS 223
In t1enc.def (LaTeX) there is:
\DeclareTextSymbol{\SS}{T1}{223}
\DeclareTextSymbol{\ss}{T1}{255}
\ss is the ß letter. It seems that \SS is the uppercase(ß) -> SS letter
(two letters in unicode). It seems that T1 encoding uses a single letter
for SS. We should consider using this one.
The problem is: \SS and \sz are already defined. See below.

> tex/context/base/enco-fde.tex
>         \installcompoundcharacter "s {\sz}
>         \installcompoundcharacter "z {\sz}
>         \installcompoundcharacter "Z {\SS}
>         \installcompoundcharacter "S {\SS}
"Z should stay SZ!

We have:
  \installcompoundcharacter "s {\SS}
  \installcompoundcharacter "z {\SS}
  \installcompoundcharacter "Z {SZ}
  \installcompoundcharacter "S {SS}

This works perfectly. If T1 encoding has (stupidly?) one character for
SS (two normal captial s; there is non such letter in in UNICODE 3.01 by
the way), we should define something \SZ or is this too confusing.
\SS should produce ß because I don't want to change hundreds of old
documents because of this incompatible unneccessary change.

I think we should produce (as we do currently):
"s -> ß
"z -> ß
"S -> SS  = uppercase "s
"Z -> SZ  = uppercase "z for the rare case that ß becomes SZ as in
MA"ZE vs. MA"SE
MASZE vs. MASSE
Maße  vs. Masse

This is how it is done in LaTeX, isn't it?

> >From the "Die amtliche Regelung der deutschen Rechtschreibung" §25 E3:
>   Bei Schreibung mit Großbuchstaben schreibt man SS, zum Beispiel:
>   Straße -- STRASSE
We do this using Stra"se and STRA"SE, Ma"se und MA"SE, Ma"ze und MA"ZE,
where's the problem?

Hraban wrote:
> Uwe Koloska wrote:
> > AFAIK the upper case of (german) szligature is SS and this character
> [True. The sometimes found spelling "SZ" is very old fashioned and only
> allowed, if "SS" causes confusion.]
I didn't know that it was even used :-) SZ or Eszett is by the way also
a wrong name the ß ligature comes from a long s (unicode 017F LATIN
SMALL LETTER LONG S) plus a normal small s (0073 LATIN SMALL LETTER S),
so the other name "sharp s" (scharfes s) is more apropiate, but
\sz=\SS=ß is the ConTeXt name. Note: \ss is plain TeX and was changed
into \SS because \ss became "sans-serif". Someone (Uwe?) suggested to
name it \sz, so we have (font-ini.tex):

%D \macros
%D   {ss, SS, sz}
%D
%D We are going to redefine \type{\ss} but for those wo still
%D want to have access to the german \SS, we save it's value in
%D \type{\SS}. Ok, I should have used \type{\sf} insead of
%D \type{\ss} in the first place.

\ifx\undefined\SS \let\SS=\ss \fi
\ifx\undefined\sz \let\sz=\ss \fi

Hans Hagen wrote:
> >AFAIK the upper case of (german) szligature is SS and this character is
> >supposed to be 223 in ec-encoding.  So the proposed changes are:
> >
> >tex/context/base/enco-ec.tex
> >       \definecharacter SS 223
> 
> \definecharacter sz 255
> \definecharacter SS 223
> 
> So, \sz remains 255 ?
Suggest: Leave it as it is, since it seems to be working well!
Or has someone any problems?

> >tex/context/base/enco-fde.tex
> >        \installcompoundcharacter "s {\sz}
> >       \installcompoundcharacter "z {\sz}
> >       \installcompoundcharacter "Z {\SS}
> >       \installcompoundcharacter "S {\SS}
> 
> Shoudl we have a \SZ ?
I don't know as long as "Z -> SZ and uppercase("z) -> SZ is not
hyphenated we don't need it, do we?
Especially how should it be mapped? \sz is ß which is already very
confusing, \SZ would be SS, and \SS then SZ?

Tobias