Image Build Fails with Custom Package

What I’m trying to do: I’m creating a hierarchical bubble chart (circle packing visualization) for my Anvil dashboard and need to import the circlify library to calculate circle positions.

What I’ve tried and what’s not working: I’ve successfully added circlify to my app’s dependencies (tried both v0.15.1 and v0.15.0), and the package almost installs and then takes forever (>15-20 min) and eventually fails on image build. I am using the data science base package, and just replicated the issue on a tutorial app. Reading other forum posts it seems that the issue was resolved for others?

Assembling config…Done
Loading base image…Done
Installing custom packages… Using Python 3.10.18 environment at: .env Resolved 1 package in 189ms Prepared 1 package in 23ms Installed 1 package in 20ms + circlify==0.15.1 Package install took 606 ms.
Saving image…

So I have been testing this as well based on your post and here’s what I found:

  1. If you only try to install circlify by itself, the build completes fully
  2. If you try to install any other package along with circlify, it will fail:

So the issue isn’t with the Data Science Image (I used the Minimal image to test), but something to do with how anvil is interacting with installing circlify

I’m having the same issue on an app that was previously working.
I think it’s a more general issue - I created a new app to see what the issue was. I used Python 3.1 and the Standard package. At this stage I have absolutely no code - just trying to build the environment.
Then it showed a “pip” and “urllib3” vulnerability error. I installed the latest pip and urllib3 packages and the vulnerability error went away.
But the build is failing:

Hi folks,

Thanks for letting us know about this. We are aware of an issue that is preventing some new package installations from completing, and we’re investigating. We will post here as soon as we have more information for you. Apologies for the inconvenience in the meantime.

Thanks for your patience - this issue should now be resolved. Please post here if you are still seeing any issues installing custom packages.

1 Like