Skip to main content

A dockerized build tool for paper compilation

Project description

texbld

Although we expect LaTeX compilation to be a declarative process (source to PDF), the compilations for large projects eventually require a large number of custom external programs and dependencies. For example, a compilation step might require running a script written in haskell, piping that output into pandoc, then putting everything into a LaTeX file for compilation with pdflatex. Good luck installing all of those programs (ESPECIALLY the pesky ghc dependencies) in a production system!

Furthermore, different LaTeX distributions will have ever so slightly different outputs (especially when working with biblatex), which is an issue for reproducibility.

The first take on these problems was mktex. Although it solves some dependency issues, it suffers from the various fragility and reproducibility issues that come with using pre-built docker images. Furthermore, because of its design, these images were forced to be monolithic, bloated, and ultimately inflexible. Each build should have exactly the dependencies that it requires and nothing more!

texbld aims to solve these problems by providing an environment where build images are fully reproducible and shareable. It uses docker for absolute system reproducibility and for usage across all platforms which it supports (MacOS, its native Linux, and Windows (untested)).

Image hashes are used to ensure that any docker image is completely immutable, preventing dependency modification issues.

Users can specify their build image in a simple TOML file (along with associated files) and upload them to github, from which it can be inherited and used by other people in their own projects.

Images can be inherited from packages in the local filesystem, GitHub, or Docker.

Installation

The project is live on pypi.

Various installation methods are described here.

NixOS

Check the nix expressions in the release branch and configure accordingly.

In order to build the master branch, run nix-build or nix build (to use flakes).

Setting Up This Project

NixOS

Running nix develop should set up everything. Note, however, that the resulting poetry virtual environment will be installed in ~/.cache/pypoetry/virtualenvs, so it's not completely nix-based.

Non-NixOS

This project uses poetry as its dependency manager. Simply run poetry install and poetry shell inside the project directory, and you should land in a virtual environment with all of your dependencies.

Testing

In order to run tests in the virtual environment, run pytest.

The Local Environment

The project configuration file should be in (project root)/texbld.toml, while local image configurations should be in $HOME/.config/texbld/packages.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

texbld-0.4.0.tar.gz (29.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

texbld-0.4.0-py3-none-any.whl (39.3 kB view details)

Uploaded Python 3

File details

Details for the file texbld-0.4.0.tar.gz.

File metadata

  • Download URL: texbld-0.4.0.tar.gz
  • Upload date:
  • Size: 29.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.13 CPython/3.10.5 Linux/5.18.13-200.fc36.x86_64

File hashes

Hashes for texbld-0.4.0.tar.gz
Algorithm Hash digest
SHA256 a24e7126d146893b5ce2763964cea96dca9f48fff367b8395bcc917f0011e3b5
MD5 2e5a3d0c979018c8964b48a14f5d4f3a
BLAKE2b-256 a8e6f89cad60d9498d4a958682a4dcd24c51f696e74202e9486edf9992bdf329

See more details on using hashes here.

File details

Details for the file texbld-0.4.0-py3-none-any.whl.

File metadata

  • Download URL: texbld-0.4.0-py3-none-any.whl
  • Upload date:
  • Size: 39.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.13 CPython/3.10.5 Linux/5.18.13-200.fc36.x86_64

File hashes

Hashes for texbld-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 82c791169b9f8ee666a39ace288173139a587298b40efdd1780763ca5a5a50fa
MD5 6d07cc1d47ff3ae858193ba593795c2c
BLAKE2b-256 0bee4d94f47f1291b6af367aebed33da4259cafbd2e5f30314d62259075cb578

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page