This is the code for my personnal blog, see this first article and guide for details: https://codecurrents.blog/article/2023-07-06.
New articles, written either as Markdown files (*.md) or Jupyter Notebooks (*.ipynb), must be created in the raws
directory. Their filename must be a date in ISO format (i.e.: YYYY-MM-DD
), followed by the extension (e.g.: 2023-09-16.ipynb
).
PNG images can be included in the raws/media
subdirectory; they need to start with the relevant article's date, followed by .png
(e.g.: 2023-09-16_Figure1.png
).
Raw articles can be processed with make publish
; only raws that have changed since the last execution of make publish
will be processed. This process converts Notebooks to Markdown, converts LaTeX to images, extracts briefs, and retargets media files. The output is saved under web/articles
, web/briefs
and web/media
.
make publish
also updates the RSS feed, under web/static/rss.xml
.
The website can be launched with make serve
during development (for testing). On my server, I use a systemd service file with the following content:
[Unit]
Description=Blog
After=network.target
[Service]
User=michel
WorkingDirectory=/home/michel/blog
ExecStart=/home/michel/.pyenv/shims/poetry run python /home/michel/blog/web/main.py
Restart=on-failure
[Install]
WantedBy=multi-user.target
The repo has changed somewhat since its creation. Its dependencies are now managed with poetry
and the Python version with pyenv
. Both were installed through my package manager (pacman
) but instructions vary per operating system.
pdflatex
is needed during the conversion of raw articles. On Arch linux I needed to install the following libraries to make it work (they might not all be required, but I hate debugging LaTeX and would rather install too many than to few):
sudo pacman -S texlive-binextra texlive-latex texlive-latexextra texlive-plaingeneric texlive-mathscience
I have disabled the proofreading using ChatGPT feature.