Trouble using glossary (name contains german umlaut)

Question

I am struggling using the glossary package and probably have spent one ore maybe two full days finding a hint why it is not working. Now I think it's time to ask the professional tex users over here for some help!

Following minimal example makes latex complain ! Paragraph ended before \@skipmath was complete.

\documentclass[12pt]{scrreprt}
\usepackage[ngerman,english]{babel}

\usepackage[utf8]{inputenc}                     % no difference whether used or not

%\usepackage[xindy={language=german,codepage=duden-utf8}]{glossaries}   % no difference
\usepackage{glossaries}
\makeglossaries

\newglossaryentry{aquivalenzklassenbildung}{
    name={Äquivalenzklassenbildung},
    plural={Äquivalenzklassenbildungen},
    description={ist ein Verfahren, etc.}
}    

%----------------- document -----------------------------
\begin{document}

\selectlanguage{ngerman}

\printglossary


\chapter{Introduction}

\Gls{aquivalenzklassenbildung} is used... and latex reports ``! Paragraph ended before \\@skipmath was complete''.
The same goes for \Glspl{aquivalenzklassenbildung}...

\end{document}

The next thing you need to know is about my tex setup (I guess): I am using TexLive installed from Debian testing repository (currently version 2020.20210202-2)

In order to generate PDFs I usually execute commands like following:

latex example.tex
makeglossaries example
pdflatex example.tex

As mentioned I have tried lots of things and I can not remember everything (to be honest I only can remember very little). I remember that I tried out various things with inputenc as well as encoding the file that contains the glossary entries in ASCII but nothing had the desired effect.

Maybe I need some options for glossaries like \usepackage[xindy={language=german,codepage=duden-utf8}]{glossaries} but not for xindy instead for makeglossaries? I really have no clue how to make german umlauts work within glossary names (btw. not using umlauts for entry names does not seem to be an reasonable workaround for me).

In some posts people where using xindy instead of makeglossaries. This is one thing I have not tried out, yet.

Maybe someone of you has an idea? I'd appreciate every advice you can serve. Even a "use xindy with following options" would be appreciated :-)

Thanks in advance,

Chris

Answer 1

glossaries uses the mfirstuc package to uppercase the first letter.

Sadly this can't properly handle Umlauts.

Apart from simply avoiding the uppercasing commands like \Gls, you could also try to use the expl3 command \text_uppercase_first:n instead:

\documentclass[12pt]{scrreprt}
\usepackage{mfirstuc}
%----------------- document -----------------------------
\begin{document}

\makefirstuc{\emph{hallo}}
%\makefirstuc{äqui} %errors

\ExplSyntaxOn
\cs_set_eq:NN\makefirstuc\text_titlecase_first:n
\ExplSyntaxOff

\makefirstuc{\emph{hallo}}
\makefirstuc{äqui}


\end{document}

Answer 2

Alternatively to the answer of Ulrike Fischer you could try to switch the method of sorting by using the sort key. Works for me. And btw: xindy is no replacement of \makeglossaries instead it is another sort algorithm you can use.

\documentclass[12pt]{scrreprt}
\usepackage[ngerman,english]{babel}
\usepackage[utf8]{inputenc} 

\usepackage[sort=standard]{glossaries}
\newglossaryentry{aquivalenzklassenbildung}{
    name={Äquivalenzklassenbildung},
    plural={äquivalenzklassenbildungen}, % beware the small ä
    description={ist ein Verfahren, etc.},
    sort={aquivalenzklassenbildung},
}   

%----------------- document -----------------------------
\makeglossaries
\begin{document}
    \selectlanguage{ngerman}
    \printglossary
    
    \chapter{Introduction}
    \Gls{aquivalenzklassenbildung} is used as well as \Glspl{aquivalenzklassenbildung} 
\end{document}