Closed Carifio24 closed 1 year ago
Base: 85.78% // Head: 79.31% // Decreases project coverage by -6.47%
:warning:
Coverage data is based on head (
ef2e465
) compared to base (08bda43
). Patch coverage: 39.24% of modified lines in pull request are covered.
:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
This PR contains improvements to the 3D scatter viewer from @victoriaono and myself. While we had previously opened #16, I've decided to split that into multiple PRs that each deal with changes to a specific exporter, just to keep things from getting unwieldy. This PR contains the 3D scatter-specific changes from #16, as well as some additional changes.
The features added to the 3D scatter viewer are:
There are a few issues, as well as a few choices that I think are worth mentioning.
sizeref
parameter that can be passed to a Cone trace) based on "the minimum 'time' to travel across two successive x/y/z positions at the average velocity of those two successive positions" - see the description of thesizeref
parameter here. In my testing, this parameter can get really small if there are cones close together. Since we don't seem to be able to change this (see issue 3613 over in the plotly repo), I've opted to make each cone its own trace for fixed-color layers as well.To warn users that the Plotly output may differ from what they see in glue, a warning message is displayed before exporting, which is attached to the (new) setting
SHOW_WARN_PLOTLY_3D_GRAPHICS_DIFFERENT
. Personally I think it's better to show one catch-all "things might be different" message rather than separate messages for each component that might differ (vectors, etc.).I'll add some tests once we see whether or not there's more to be done with this.