Glossaries package not working with Marathi language - TeX

TopAnswers TeX

Meta

Databases

TeX

Code Golf

APL

C++

.net

db<>fiddle

Java

*nix

PHP

PowerShell

Python

Rust

टेक्-मराठी

Typst

Web Client Dev

Web Server Dev

Glossaries package not working with Marathi language

add tag

निरंजन

```
\documentclass{article}
\usepackage{fontspec}
\usepackage{babel}
\babelprovide[main,maparabic]{marathi}
\setmainfont[Renderer=Harfbuzz,Script=Devanagari]{Shobhika}
\usepackage[acronym]{glossaries}
\makeglossaries
\newglossaryentry{संज्ञा}{name=संज्ञा,description={स्पष्टीकरण}}
\newacronym{संक्षेप}{संक्षेप}{उलगडा}

\begin{document}
\gls{संज्ञा} व \acrshort{संक्षेप}

\printglossary[type=\acronymtype,title=संक्षेपसूची]

\printglossary[title=संज्ञासूची]
\end{document}
```

LuaLaTeX compiles this code successfully, but running `makeglossaries` and then rerunning LuaLaTeX fails. The log says -

```
A funny symbol that I can't read has just been (re)read.
Just continue, I'll change it to 0xFFFD.
```

The output is also messed up. How to get things working correctly? 

Additionally if I use `maparabic` parameter to `\babelprovide` command, it converts the page numbers into Devanagari ones, but then `makeglossaries` rejects all the glossary entries in the document. `.glg` file reports -

```
!! Input index error (file = test-package.glo, line = 1):
   -- Illegal page number १.
```

Top Answer

samcarter

You need some unicode aware index tool. Although no longer developed, you could try your luck with `xindy`:


```
% !TeX program = txs:///arara
% arara: lualatex: {synctex: on, interaction: nonstopmode, shell: yes}
% arara: makeglossaries
% arara: lualatex: {synctex: on, interaction: nonstopmode, shell: yes}
\documentclass{article}
\usepackage{fontspec}
\usepackage{babel}
\babelprovide[main,mapdigits]{marathi}
\setmainfont[Renderer=Harfbuzz,Script=Devanagari]{Shobhika}
\usepackage[acronym,xindy,automake]{glossaries}
\makeglossaries
\newglossaryentry{संज्ञा}{name=संज्ञा,description={स्पष्टीकरण}}
\newacronym{संक्षेप}{संक्षेप}{उलगडा}

\begin{document}
\gls{संज्ञा} व \acrshort{संक्षेप}

\printglossary[type=\acronymtype,title=संक्षेपसूची]

\printglossary[title=संज्ञासूची]
\end{document}  
```

![image.png](/image?hash=c1f3ded9d6fdde904242f6e9275cd51623dba58f9789d66f514fa12d95193e36)

Actually `mapdigits` and `maparabic` are both required for the cross-engine compatibility (mapdigits won't work with XeLaTeX). I didn't mention `mapdigits` in my question as it didn't produce any errors.

arara is a handy tool if you need some special configuration for one document. You don't have to change the configuration of your editor, but instead just specify *in* the document how it should be compiled. Under the hood arara then calls the normal programs to compile the document

How to customize this in my editor? I use emacs. Your magic-comments don't work for me there. Using `glsnumbers=false` solved the error that I mentioned in the question, but it generates a new one on my system. Your screenshot looks perfectly fine, but I need Devanagari numerals in place of the Arabic ones.

Is this the output you are looking for? (sorry, can't read it, so no idea if it makes sense): ``` % !TeX program = txs:///arara % arara: lualatex: {synctex: on, interaction: nonstopmode, shell: yes} % arara: makeglossaries % arara: lualatex: {synctex: on, interaction: nonstopmode, shell: yes} \documentclass{article} \usepackage{fontspec} \usepackage{babel} \babelprovide[main,mapdigits]{marathi} \setmainfont[Renderer=Harfbuzz,Script=Devanagari]{Shobhika} \usepackage[acronym,xindy,automake]{glossaries} \makeglossaries \newglossaryentry{संज्ञा}{name=संज्ञा,description={स्पष्टीकरण}} \newacronym{संक्षेप}{संक्षेप}{उलगडा} \begin{document} \gls{संज्ञा} व \acrshort{संक्षेप} \printglossary[type=\acronymtype,title=संक्षेपसूची] \printglossary[title=संज्ञासूची] \end{document} ```

1 Answer