Closed janosh closed 1 year ago
thanks @janosh !! It looks nice. I can see downstream people ask for color distinction for different oxi states, but as an incremental fix for the labels I think this is sufficient.
Sure happy to merge as is. We just need to await the next pymatgen release. Also, would be great if @mkhorton or @tschaume or whoever can invite me to the Percy project to green-light the visual changes causing CI to fail.
percy.io
New pymatgen
release to go along with this merge planned for later this week.
Closes #338.
This is a WIP to start gathering feedback. As suggested by @yang-ruoxi, this first step splits every oxi state into its own legend label without affecting the site colors. I.e. different oxi states still have the same color in the structure.
Suggestion: Make negative oxi states slightly darker, positive ones slightly brighter and combine different oxi states for the same element into 1 wider label.
This PR requires upstream changes in pymatgen: https://github.com/materialsproject/pymatgen/pull/2998 i.e. we can't put this into prod until pymatgen cuts a new release.
@mkhorton If you have a minute, could you let me know if there's any problem with making
StructureMoleculeComponent.get_scene_and_legend()
non-static? See e418cae.Here's what this PR looks like for the problematic structure reported by @ardunn on matsci.org.