New bokeh-static package

This may not be the right place for this issue, but we can move it as appropriate. I was considering again today my annoyances with getting static plots (in particular so i can readily share static views of notebooks on github)

Additionally, firefox is now available as a conda package (https://github.com/conda-forge/firefox-feedstock). This will provide us much nicer rendering than what phantomjs spits out by default.

I’m imagining a workflow like:

conda install bokeh-export   # install selenium, pillow, phantomjs, firefox, geckodriver

# In a notebook
from bokeh_export import (
    initialize_driver,
    notebook_show as show
)
initialize_driver('firefox')  # set-up a single driver instance for faster rendering, pick your driver. use this to pass in options that are passed into a typical webdriver setup

p = figure().....
show(p)  # does the work of Ipython.display.Image(export_png(p)) with driver globally scoped to be the one setup

[bryevdv]

We are definitely interested to look at kaleido, it’s just a matter of someone finding the time. Assuming it works as well as the complicated selenium process we have how, I’d be in favor of just taking it on as a hard dependency since (AFAIK) it is widely available on pip and conda.

[joelostblom]

@birdsarah @bryevdv You might have seen it already, but I just stumbled across the recently announced Kaleido library from plotly, which is a new library for static export of web-based graphics, with the specific goal to make this functionality widely available.

The goal of the Kaleido project is to make static image export of web-based visualization libraries as universally available and reliable as it is in matplotlib and ggplot2.

I am excited about this because I would love to see better support for this functionality across several libraries. While I don’t know how this would fit with bokeh technically, it might be something to consider instead of developing a separate library for static export. In the blog post linked above, they mention briefly how to integrate it with other libraries:

While the development of Kaleido has been motivated by the needs of the plotly.py community, we know we’re not unique in facing these challenges. We’ve designed the C++ and Python portions of Kaleido using a basic plugin architecture (Kaleido plugins are called Scopes) with the goal of making it possible to support image export for other web-based visualization libraries. If you’re interested in adding support to Kaleido for another web-based visualization library, check out the Scope (Plugin) Architecture wiki page and let us know how we can help!

Also relevant for #9169