Set of tools to render LaTeX documents in multiple languages using Jinja2 Python library and some conventions inspired by jekyll-multiple-languages-plugin.
So you have a LaTeX document that you want to internationalize. The first step
is convert your project in a template. Using latex-ji18n
command line tool,
you can render YAML files into LaTeX Jinja2 templates.
You need a pdflatex
binary available at your system PATH. Then run:
pip install latex-ji18n
Next structure of directories is the workflow that imposes this utility working with it:
├── _config
│ ├── data.yml
│ ├── layout.yml
│ ├── style.yml
│ └── _private (optional)
│ ├── data.yml
│ └── layout.yml
├── dist
│ ├── en.pdf
│ ├── es.pdf
│ └── fr.pdf
├── _i18n
│ ├── _private (optional)
│ │ └── es.yml
│ ├── en.yml
│ ├── es.yml
│ └── fr.yml
└── src
├── template.tex
├── assets.jpg
└── references.bib
The process is simple, you write your replacements in src/template.tex
Jinja2 template file with \BLOCK{}
and \VAR{}
syntax, write your
data or options inside _i18n/
and _config/
folders YAML files,
run latex-ji18n
to compile the PDFs and you will see each one for each
language in dist/
folder.
_private/
: Directories that stores all the sensitive data that you don't want to include, for example, sharing your repositorie with GIT. This folders are totally optional. Inside each one, you must keep the same files structure that in_i18n/
and_config/
directories.data.yml
: Data that you want to include in the context, at the root of the dictionary, common to all languages. All the fields stored here are included at the root of the context. You can't include the fieldslayout
norstyle
in the root of the context for data files.layout.yml
: Designed to store some layout options for the template. Fields stored here are available for the context in thelayout
dictionary.style.yml
: Designed to store style options for the template. Fields stored here are available for the context in thestyle
dictionary._i18n/{language}.yml
: Overrides data for the context. Designed to localize your data output. An output file will be created for every{language}.yml
file, with the name of{language}.pdf
.
The contexts are created updating a dictionary iterating over files in next order:
config/data.yml
config/layout.yml
config/style.yml
config/_private/data.yml
config/_private/layout.yml
config/_private/style.yml
_i18n/{language}.yml
_i18n/_private/{language}.yml
Context data is dumped following next rules:
- All the data located in
data.yml
files is dumped at the root of the context. - Data located in
layout.yml
files are dumped into alayout
variable at the root of the context. - Data located in
style.yml
files are dumped into astyle
variable at the root of the context. - If
src/
directory contains.bib
files, database entries located at this files will be available ordered by entry type at_bibdb
variable at the root of the context.