Starting December 3, join live sessions with database experts and the Microsoft product team to learn just how easy it is to get started
Learn moreGet certified in Microsoft Fabric—for free! For a limited time, get a free DP-600 exam voucher to use by the end of 2024. Register now
I have published a custom package (python) in an environment. What is the correct way to update it.
I keep on enhancing the package through out the DEV cycle, but it seems I need to
delete the old package (package_one-0.0.1-py3-none-any.whl) first
=> publish the changes
=> reupload the enhanced package
=> republish the same package (package_one-0.0.1-py3-none-any.whl) again (takes a whileeeeeeee).
What is an optimize way to do this (assuming I am doing it wrongly) and if possible from VS Code (desktop).
Is there a seamless way to DEV a package, upload, test in workspace (can't execute notebook in VS Code Synapse extension currently) and DEV again till PROD, without wasting so much time for file republishing?
Thank you in advance.
HI @smpa01,
Have you checked if any versions of this package or other similar feature package existed in the public libraries? If that is the case, I'd like to suggest you add them in the public libraries and easy maintenance.
For the custom package, current there not include any automation on this. Perhas you can try to submit an idea to add support on this package in the public libraries.
Regards,
Xiaoxin Sheng
When you are saying public library you are referring to libraries/packages from PyPi. But they are all internally developed packaged that can't be published to PyPi. But we need to able to seamlessly use them in a custom environment (with CRUD operation of the lifecycle of the package) for our notebooks.
HI @smpa01,
AFAIK, current there not exist official methods to automate manage custom packages.
For your scenario, I think it should better submit an idea about add official support to integrate with platforms(e.g. git) to help user simply manage the custom packages on the environment.
Regards,
Xiaoxin Sheng
Starting December 3, join live sessions with database experts and the Fabric product team to learn just how easy it is to get started.
Check out the November 2024 Fabric update to learn about new features.
User | Count |
---|---|
15 | |
12 | |
9 | |
9 | |
4 |