arabica

A sound and versatile pandoc LaTeX boilerplate to produce academic books (in PDF) utilising solely modular Markdown files, featuring YAML, KOMA-Script, BibLaTeX and CSL.

For typesetting an article please see the robusta boilerplate.

Thanks

The arabica boilerplate is inspired by the pandoc latex template eisvogel by Wandmalfarbe and was built with the help of the following sources and tools: References and Credits. For other projects like the one at hand please see the list of user contributed templates at the pandoc wiki and the list of pandoc boilerplates at the pandoc wiki.

Features

1. Simple Your writing process is as simple as possible because you only need to write Markdown files and let the strong defaults of pandoc and LaTeX work for you. (see ./controls/ and ./settings/)
2. Configurable Your book projects are as configurable as possible because you can set your metadata and variables via YAML files and then reuse them if you want to. (see ./settings/)
3. Modular Your book projects are as modular as possible because of strict separation of contents and presentation (style). Your content live in separate directory (see ./core/) where you can organise them intuitively. The pandoc template itself lives in ./controls/.
4. State-of-the-art technology stack and low tech debt at the same time.

Preview

Default title page	Default info page	Default example page

Examples

Go to ./examples/ to view a list of complete example book projects.

Install

Preconditions

1. Install TeX Live or any alternative LaTeX distribution.
2. Install pandoc.

Clone

1. Go to your templates directory: cd /path/to/your/templates/.
2. Clone (or download) this repository git clone https://github.com/qualiacode/arabica.git
3. Done!

Use

Optional: copy, paste, rename, delete

1. Copy the entire arabica boilerplate located at your own local templates directory and paste the entire arabica boilerplate to your directory of your own projects directory: cp /path/to/your/templates/arabica/* /path/to/your/projects/projectname/
2. Go to your project directory: cd /path/to/your/projects/projectname/
3. Delete all the files you do not need e.g. ./examples: rm -vrf examples/*

Set metadata

1. Go to ./settings/00_01_metadata.yaml
2. Set your (meta)data, the more the better.
3. It is a good idea to add at least one author, a title and a year.
4. Enclose your values ((meta)data) in single quotes: '()(meta)data)'.
5. If you want to leave a value undefined enter a tilde.
6. Example:

title: &title 'Short title'

subtitle: &subtitle ~

author: &author
  - 'Jane Roe'
  - 'John Doe'

year: '2018'

Set variables

1. You can leave the defaults and you will be fine most of the time.
2. If you want to change things, then the same rules as in Metadata settings apply.
3. Go to ./settings/00_01_settings.yaml.
4. Set your settings.
5. Example:

classoption:
  - 'twoside'
  - 'twocolumn'
  - 'parskip=full'

custom_title_page: false

...

toc: true

Collect

1. Do your research, collect your bibliographic data.
2. Go to ./bib/
3. Save your bibliographic data in Bib(La)TeX format (.bib) in the file ./bib/.references.bib.
4. Go to ./settings/00_01_settings.yaml.
5. Refer to your .bib file by adding the corresponding path of your .bib file as value to the corresponding key.
6. Example (default):

bibliography: './bib/references.bib'

7. Example entry in your .bib file:

@article{Cantor1984,
  langid = {german},
  title = {Ein Beitrag zur Mannigfaltigkeitslehre},
  volume = {1877},
  doi = {10.1007/978-3-7091-9516-1_3},
  journaltitle = {Über unendliche, lineare Punktmannigfaltigkeiten},
  date = {1984},
  pages = {25-44},
  author = {Cantor, Georg}
}

8. In this example your cite key for this bibliographic entry is Cantor1984.

Select

1. Go to CSL
2. Dowload your desired citation style language file e.g. chicago-author-date-de.csl.
3. Save your .csl fiile to: ./csl/
4. Go to ./settings/00_01_settings.yaml.
5. Refer to your .csl file by adding the corresponding path of your .csl file as value to the corresponding key.
6. Example (default):

csl: './csl/chicago-author-date-de.csl'

Write

1. Go to ./core/ and write you markdown files.
2. Organise them as you like but keep in mind that pandoc will append them sequentially, determined by the alpha-numeric ordering of your file names.
3. A good example:

01_introduction.md
02_mainpart.md
03_conclusion.md

Cite

1. Cite your sources by adding your refercences via [@citekey] to your markdown file(s) where ever you need them.
2. Example:

„Cantor once said: ‚Wenn zwei wohldefinirte Mannigfaltigkeiten ...‘ [@Cantor1984]“

Run

1. Execute pandoc with the following options:

pandoc                                                                      \
  -s                                                                        \
  -o ./output/projectname.pdf                                               \
  --data-dir=$HOME/path/to/your/projects/projectname/                       \
  --filter pandoc-crossref                                                  \
  --filter pandoc-citeproc                                                  \
  --pdf-engine=xelatex                                                      \
  --top-level-division=chapter                                              \
  --number-sections                                                         \
  --template $HOME/path/to/your/projects/projectname/controls/arabica.latex \
  ./core/*.md                                                               \
  ./settings/*.yaml                                                         \
  ./metadata/*.yaml

2. Done!

3. Alternatively you can omit the long path in the --template option by copying the arabica.latex template located at ./controls/arabica.latex to the default directory of your pandoc templates located at ~/.pandoc/templates/.

4. If you have copied the arabica.latex template into the directory of your default pandoc templates, then you can execute pandoc like this:

pandoc                                                \
  -s                                                  \
  -o ./output/projectname.pdf                         \
  --data-dir=$HOME/path/to/your/projects/projectname/ \
  --filter pandoc-crossref                            \
  --filter pandoc-citeproc                            \
  --pdf-engine=xelatex                                \
  --top-level-division=chapter                        \
  --number-sections                                   \
  --template arabica.latex                            \
  ./core/*.md                                         \
  ./settings/*.yaml                                   \
  ./metadata/*.yaml

5. For debugging puproses you can produce a .tex file by runing pandoc with

-o ./output/projectname.tex

5. Please note, though: the arabica boilerplate is designed to be used with the given directory structure respectively file structure indicated in BODYPLAN.md. Of particular importance are the metadata and settings files: ./metadata/00_01_metadata.yaml ./settings/00_02_settings.yaml.

6. The boilerplate should work even if you change its inner structure but keep in mind to change all the paths and variables accordingly. The arabica.template file may work as long as you provide some basic variables used in this template file.

Enjoy

1. Go to ./output/ and enjoy your results.

Other

1. To read out your current default template execute: pandoc -D latex > default.latex.

License

MIT License

Copyright (c) 2018 Martin Maga

Go to ./LICENSE.md.

References and Credits

1. Plot graphics: Gnuplot
2. Blind text: Jasper Van der Jeugt lorem-markdownum
3. Bug fix by Marco Torchiano @ StackExchange see Pandoc issue 1023.
4. Atom editor
5. Better Bib(La)TeX for Zotero
6. CSL
7. git
8. KOMA-Script
9. LaTeX
10. Markdown
11. Pandoc
12. Pandoc-citeproc
13. Pandoc-crossref
14. TeXLive
15. XeTeX
16. YAML
17. Zotero
18. zotero-citations
19. zotero-picker
20. zotfile.