v0.4.9

Enhancements to the cauldron.display.table() command that allow for simplified random sampling as part of the display process as well as display column formatting.

v0.4.8

• Support for Plotly 4.0 has been added.
• The standard dockerfile now uses Python 3.7 instead of Python 3.6.
• The Python 3.7 images now use Python 3.7.2 instead of 3.7.1.

Improved Library Management

Improves the management of libraries throughout the lifecycle of a notebook project, especially in relation to non-interactive execution.

Image display

This release adds a new display function cd.display.image() for including images within notebooks. Images must be stored in the assets directory of the project, or an explicitly specified custom asset directory.

Component Styling Enhancements

This release fixes component styling in the notebook display due to CSS collisions in components that didn't fully isolate their class scopes.

Plotly Resizing

This release updates the fluid sizing of Plotly plots to support the latest release of Plotly that has modified the dom structure.

Remote External Libraries

Improve support/handling of external directory libraries when running in a remote context.

Updated Dependencies

Updates notebook web dependencies to latest versions.

Expanded Compatibility

This release:

• Expands compatibility with Python 3.7,
• Updates package configuration,
• Fixes Plotly scaling issue on certain platforms,
• Adds shared project data to the cauldron.run_project return object,
• Increases unit testing of the batcher module.

Console Enhancements

This release introduces the following changes:

• #42 Adds cd.step.render_to_console() that allows for rendering strings with Jinja2 to the console from within steps.
• #47 Removes whitespace console notifications to reduce console noise when running steps.
• #52 Modifies server execution route to fix Python 3.7 having a new keyword async.