After coming back from holidays a few days ago, getting back to work is always exciting. It's especially exciting when something no longer works: It appears since the update to v1.1.0 of the databricks extension for VS Code, the notebook results are no longer rendered properly, making IDE development quite difficult.
Is anyone else facing this issue? Were you able to resolve it?
In the above screenshot you can see: A sample notebook run via 'Run File as Workflow on Databricks' on the topleft, the output in VS Code on the top right, and the associated workflow run in Databricks itself on the bottom. As we can see, the job was run correctly - the output simply didn't get displayed.
Expectation would have been a rendered notebook in VS Code as seen in the browser. Actual result is... well.. nothing.
We tested with a downgraded extension version 1.0.0 as well - same issue. We're runnning this in VS Code version 1.77.0.
Edit:
Only I - but unfortunately not my colleagues - have been able to produce this pandas error with all experimental features enabled in the settings.
Any idea which pandas version it's trying to fetch? All python envs on the machine are running at least Pandas 2.0.3
Edit 2: Checking the logs, I was able to find another thing:
When I'm submitting a new run i get this error about a missing notebookwrapper.py
It appears Databricks has addressed the issues. At least for my coworkers and me it appears to be working again.