Emulate all color options exposed by Plotly Express

[tpvasconcelos]

This is a meta-issue to track and discuss the implementation of all color options exposed by Plotly Express.

This discussion started in #215 but I decided to open a new issue to discuss and track the more general goal of implementing of all color options exposed by Plotly Express, and make sure ridgeplot follows the same conventions laid out by the Plotly Express interface.

Here are some relevant references:

• https://plotly.com/python/colorscales/
• https://plotly.com/python/discrete-color/
• https://plotly.com/python/builtin-colorscales/
• https://plotly.com/python/reference/

---

Notes

Plotly uses color_continuous_scale in Plotly Express functions and colorscale for its object API (e.g. marker.colorscale in go.Scatter)

From the Plotly docs:

legends are visible representations of the mapping between colors and data values. Legend markers also change shape when used with various kinds of traces, such as symbols or lines for scatter-like traces. Legends are configurable under the layout.legend attribute. Legends are the discrete equivalent of continuous color bars

Currently our built-in support for legends is also quite basic. Maybe this is something we could expand on in the future?

Also from the docs:

Most Plotly Express functions accept a color argument which automatically assigns data values to discrete colors if the data is non-numeric. If the data is numeric, the color will automatically be considered continuous.

Is there something we can do to emulate this color behaviour? There seems to be some conceptual overlap between Plotly's color and our colormode...

Tasks:

General

• More comprehensive docs for all coloring options #279
• Use active template's to infer default colorscale #236
• Stop using the term "midpoint" in our docs. The technical term would be "interpolant"
• Take a look at _plotly_utils.basevalidators.ColorscaleValidator to see if we can use it to replace some of our own custom logic.

Continuous color scales

• Allow continuous color scale values to be integers too (for 0 and 1). This should be a simple fix to the type annotations.
• Add support for named CSS colors #229
• Continuous colorscales as lists of CSS colors #230
• Plotly Express allows passing a continuous color scale by name just like ridgeplot already does. However, Plotly treats the input as case-insensitive strings, while I don't think we do... (e.g., they allow both 'Inferno' and 'inferno'). (REF1) (REF2)
• Rename colorscale to color_continuous_scale. Maybe we could deprivate colorscale and support both for a long while?
• What is the difference between ridgeplot.list_all_colorscale_names() and px.colors.named_colorscales() (and ColorscaleValidator().named_colorscales)? Why did we use a different name? Document this?
• Implement range_color #276
• Better, easier, and integrated support for color bars #277
• Do we need to do anything special to support cyclical color scales? Maybe they are already supported? Are they included in our ridgeplot.list_all_colorscale_names()? Etc...
• Same as above but for diverging color scales... See next point too 👇
• See: Ideas for more coloring options and coloring-related aids #278

Discrete color scales

• Implement color_discrete_sequence #275
• Implement color_discrete_map #274

[tpvasconcelos]

All tasks have either been completed or turned into issues/discussions... 🚀