Description
Note: This is a temporary issue just for bug discussion / tracking. Should be closed after this iteration.
As of 7/7/2022 this is my current groupings of the Getting Started bugs.
Python Extension install / Jupyter Extension install
• Don't show the "Notebooks might need to be restarted" prompt after Python Extension is installed (or at all?)
• Install kernels from the marketplace only lists Jupyter
• Extension dependency issues when building python dev containers
• Don't notify the user to reload after installing the extension
• Impossible to run any notebook without Python extension when RAW isn't supported
Python / pip install
• Change our "Install Python" command to use the Python contributed command
• Install Python shows up if in remote mode
• Selecting install from prompt when missing pip fails because pip is not available
• Zmq Not supported and requiring Python for Jupyter (5897 dupe?)
IPyKernel Install
• Consider adding a More Info button to the modal dialog prompt requiring users to install ipykernel
• installing ipykernel looks successful after hitting permission denied
• Change prompt text that requires users to upgrade ipykernel to a newer version
• Reword ipykernel installation message
• Unclear why "Ipykernel setup required for this feature" is showing up for new setups and trying to run-by-line
• Identify kernels that have ipykernel installed to kernels picker.
• Version 2 of "I installed ipykernel 6, but I still can't use run-by-line"
• I installed ipykernel 6 to try run-by-line, but I don't want to have to restart the kernel.
• I misread the modal dialog about installing ipykernel 6. I thought it was telling me to install with pip.
• Option to change kernel sends me down a rabbit hole
Performance
• building of wheel taking a long time
• Mitigate slow kernel discovery (API adoption, placeholder kernels, measurements)
Initial Suggested Kernel
• 2.x python kernel suggested but no longer supported
• Automatically disconnect local conda kernel and switch to dummy kernel. (Probably fixed)
• Auto picked kernel after running 'Connect to Github' is .NET (remote)
• Allow a machine to configure the default Python interpreter version
• Preferred Python Notebook Kernel
• Losing selected kernel when installing python extension
• Jupyter notebooks not using Python in current path
• Show available kernels rather than languages in Create New Blank Notebook UI
Notifications / Prompts
• If Python is missing server auto starter tries to run and pops up the "Needs Python Extension" modal
• Wall of notifications
• Notification overload when installing ipykernel
Remote
• Remote jupyter server failures are not diagnosable
• ModuleNotFound errors on remote could indicate that files are not copied to the remote machine
• Consider treating Notebooks in remote repositories like they are in untrusted workspaces
Jupyter Starting
• Cannot resolve starting jupyter if active interpreter doesn't support jupyter
• Kernels using 32-bit Python environments installed via conda does not work
• Jupyter interpreter can be invalid and break things
• Ship our own Jupyter to eliminate configuration problems
Environment setup / configuration
• Investigate a method to create or suggest a virtual environment with ipykernel + needed packages for a given notebook
• Create a 'blessed' environment that has known modules installed
• Consider adding a blessed basic environment to install with the Jupyter extension
Basic Initial Workflow (everything works, something very basic in work flow is confusing)
• Set new notebook folder location
Tutorials / Docs / Walkthroughs
• "Run by line" tutorial is hard to navigate
• Walkthrough text is not localized
• 'Select a kernel' walkthrough step button does nothing
• Jupyter Notebook Tutorial
Aggregate / Grouping issues
• Making it easy to get started with the Jupyter extension
• Proposal for proxy kernels (as it's more of a broad work item)
• Notebook getting started experience challenges/questions
• Improvements to getting started experience for DS
Not "Getting Started" issues
• If I open an ipynb and then select a kernel for it, opening a Python terminal should activate the same environment