bored-charts

Build easy, minimal, PDF-able data reports with markdown and python.

Roadmap

• custom figures for individual reports using path parameters
• dynamic figure sizing using flexbox
• installable as a framework
• example project
• separate pyproject.toml for lib and example
• publish to pypi
• less boilerplate to register figure endpoints
  • @bored_router.chart decorator
  • allow decoration of functions that return figures directly
• make report_name path parameter optional
• altair figures as html
• matplotlib figures as png (drop mpld3)
• allow nested pages for grouping reports
• matplotlib figures as svg?
• support python 3.10, 3.11
• make plotting libraries optional
• pdf exports with selenium in headless mode
• ability to archive reports (export to static html, move to archive directory, still serve from archive directory, but can get rid of analysis—could just be archive endpoints for figures?)
• cli? (boredcharts init, boredcharts export [report], boredcharts dev, boredcharts serve)
• deploy to bored-charts-example.oliverlambson.com
• dashboard layout with tighter grid layout
• example with database
• example with authentication

Usage

See bored-charts/README.md (you're currently reading the development README).

Development

You'll need uv. (and node for linting/formatting, sorry)

make env  # you need uv & node for this
make test
make lint
make fmt

I recommend running the full example project to see how everything fits together:

cd examples/full
make dev

Built with

• fastapi
• htmx
• jinja
• tailwindcss

Alternatives

Inspired by evidence.dev, observable framework, and rill, but: (1) I wanted to use python for data analysis and charting, (2) I didn't want a static site, and (3) I didn't want to pay for deployment.

You could also achieve something similar with pandoc and nbconvert, see examples/_alternatives for more.