Repository navigation
Explain the coordinate systems in the user guide - #294
Conversation
|
Hi @manduinca thank you for the submission! Could you let us know if AI was used to make this PR and if so, to what extent? |
|
I'm also merging in the main branch to this PR so remember to |
It's better to compare them in a single figure and the symbols didn't match the docstrings of the conversion methods.
|
@leouieda yes, and to a large extent. I used Claude Code (an AI coding agent) for most of this PR: it wrote the text of the coordinate systems section and the code that draws the figure, following my direction, and the docs build was checked locally before I opened it. The AI policy came out after I submitted this, and under it this PR counts as mostly AI-made. If you'd rather close it, that's fine with me; your figure changes are worth keeping either way. This reply was also drafted with AI help. |
|
Hi @manduinca thanks for the disclosure. We can accept this one since we've already put in the time for reviewing and it was from before the new policy. Thank you! Going forward, we would prefer mainly human-authored PRs. |

Fixes #257
Adds a section at the top of the coordinate conversions page describing the four systems Boule converts between (geocentric Cartesian, geocentric spherical, geodetic and ellipsoidal harmonic) and what each coordinate means geometrically, before the examples.
The figure is generated in a
jupyter-executeblock with matplotlib, so there is no image file to keep in sync and no copyright to worry about. Every distance and angle in it is computed with Boule itself on anEllipsoidwith a flattening of 0.3; the text says so, since with a realistic flattening the systems are indistinguishable by eye. Left panel: Cartesian axes, r and φ′, h and φ. Right panel: the confocal ellipsoid through P with u, its foci, and β on the circumscribing circle.Built the page locally with Sphinx 7.2 and jupyter-sphinx 0.5; the new section renders without warnings.