Now that kaleido development has ceased and orca is deprecated, are there recommendations for generating static plots? Since kaleido is dependent on the Chromium browser, how much longer can we continue to use it before there is a major change to the Chromium browser that will impact kaleido? My company is not going to allow usage of unsupported packages and the only other option is to cease using plotly and migrate to matplotlib.
Thanks for any suggestions.
Need to up this even if this late.
Started having issues using plotly on Azure databricks… related to this:
Cause kaleido to explicitly fail if no chromium: by ayjayt · Pull Request #224 · plotly/Kaleido
EDIT: after some more digging found out I can bypass the issues with chromium by setting the rendering, this is needed in Azure Datarbricks only when using fig.show() in job runs after installing kaleido. Nowhere else…
fig.show(renderer='databricks')
I arrived here looking to see how to reduce the kaleido module size. I’m bumping into a slug size limitation on Heroku and kaleido is the biggest contributor. I just added a pdf generation feature so I’m sending web page graphics to server with to_json and then using to_image on server to create image files for pdf. I’m not actually having any functionality issues at this point.
I mainly selected Plotly for this ability to serialize plots and pass them back and forth from client to server. Would running the pdf generation with plotlyjs on a node server make more sense or are there similar issues there? (kaleido or otherwise)