Le Pressoir is a static site generator designed to create html-enriched books. Designed primarily by and for researchers in the humanities and social sciences, it enables the publication of data- and information-rich texts, including detailed metadata, structured bibliographies, footnotes, infra-textual tagging (to identify selected terms, link them to authorities and create editorial objects : index, glossary, cartography...), additional content (illustrations, videos, text complements...), automatic indexes, an annotation tool (with Hypothes. is), research queries, ...
Based on a modular publishing chain, this generator adapts to the needs and constraints of digital publishing, and can be used to produce enriched books, course materials, documentation and all types of digital publications.
Le Pressoir is based on free and open source tools. The code is available as open source under the GPLv3 license on a repository in Huma-Num's GitLab instance.
Technical principles
Technical and editorial choices have been established according to five general principles:
- the granularity of content and the fine structuring of data,
- the modularity of the editorial chain and the various formats,
- low-tech applied to formats and software, as a guarantee of the sustainability and durability of the chain and the content produced,
- the durability of data and its accessibility,
- free software, open source and open access.
In line with these principles, texts, metadata and bibliographic references are edited in markdown, yaml and bibtex formats respectively, from which the Pressoir produces static html files. Books can then be simply deposited on a server, or deployed via a software forge.
History
When the project began in 2018, the production script was written in bash, using the following software and languages: Pandoc (html content generation), XSLT (html enrichment), BaseX and XQuery (index production). It was then implemented in a Python script for internal use by the two collections involved in its development: Les Ateliers de [sens-public] and “Parcours Numériques” (for the Presses de l'Université de Montréal). During 2024, it was finally distributed as a Python package accessible to all and usable beyond the collections for which it had originally been designed, and in 2025, accessible through UV Python packages manager.