Changelog¶

Unreleased¶

Added an experimental “compiler” which converts a SankeyDefinition into a complete recipe for the Sankey diagram, which can be quickly and simply applied to a dataset later. This enables (somewhat) faster calculation of Sankey diagrams against many datasets, but more importantly makes it feasible to do the calculations in other languages/contexts, since all the complexity and interpretation of the SankeyDefinition has been done up-front. An experimental JavaScript library is included which can interpret the compiled specs.

sankeyview is now called floWeaver!
There is a new top-level interface to creating a Sankey diagram, the floweaver.weave() function. This gives more flexibility about the appearance of the diagram, and lets you save the results in different formats (other than showing directly in the Jupyter notebook), while still being simple to use for the most common cases.
No longer any need for from sankeyview.jupyter import show_sankey; use floweaver.weave() instead.
New way to specify link colours using floweaver.CategoricalScale and floweaver.QuantitativeScale, replacing hue and related arguments to show_sankey. See Colour-intensity scales for examples.