Skip to contents

allonboard is early, and the most useful help right now is people trying it on their own data and saying what happened. You do not need to know the internals to do that.

Try it

# install.packages("remotes")
remotes::install_github("allboa/aobview")
library(aobview)
view(your_sf_object)

Get started has examples to copy. Some things worth trying:

  • Polar data, in the south or the north, in longitude and latitude or already projected.
  • A remote Cloud Optimized GeoTIFF read with terra::rast("/vsicurl/...") (the server must allow CORS for the browser to fetch tiles).
  • A CRS you care about, with crs =.
  • Large vector data, to see where it slows down.

Report what you find

Open an issue in the repository that fits, with the code you ran, your platform, and a screenshot if it is about drawing:

Warnings about tile budgets or tiles that cannot be projected are left noisy on purpose while the planner’s trade-offs are evaluated; reports of when they appear are welcome.

Contribute

Read the brief first: AGENTS.md in the design repository. It is written for coding agents, and people are welcome to follow the same flow. In short:

  • Work starts from an issue with acceptance criteria; comment to claim it.
  • One issue, one branch, one pull request that says how the criteria were checked.
  • Rendering changes come with a screenshot, tested in EPSG:3031 first.
  • Keep dependencies lean and package sources ASCII.

Much of the day-to-day work in these repositories is done by AI coding agents working from issues, with the maintainer setting direction and making the gate decisions. Human contributions follow the same flow and are very welcome.

Known gaps

  • Package names are placeholders, and nothing is on CRAN.
  • On macOS, CRAN’s gdalraster binary cannot find its PROJ database, so raster views need gdalraster from conda-forge.
  • Seams and the antimeridian in some projections are open work.