Skip to content

Conversation

@jrast
Copy link
Owner

@jrast jrast commented Aug 7, 2023

No description provided.

@BrianPugh
Copy link
Collaborator

I know you're reworking, but just commenting that I was able to install the prebuilt macos binary for arm:

Downloading littlefs_python-0.6.2-cp311-cp311-macosx_11_0_arm64.whl (150 kB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 150.4/150.4 kB 3.5 MB/s eta 0:00:00
Installing collected packages: littlefs-python

@jrast
Copy link
Owner Author

jrast commented Aug 7, 2023

Yeah, currently the Apple Silicon wheels are built, but it seems like overly complicated... I want to simplify and streamline the build process.

@BrianPugh
Copy link
Collaborator

in my python template, I have a long, but not-too-complicated pipeline. It particularly long so that:

  1. Everything can run in parallel.
  2. Each OS has it's own job. Other workflows like to have a bunch of steps that run if: matrix.os == 'ubuntu-latest' and such, but I think it actually ends up making the workflow harder to read and more error-prone. Many platforms have their own quirks and require their own step tweaks.

If you are interested in using Poetry for this project, I can help set that up.

@jrast
Copy link
Owner Author

jrast commented Aug 8, 2023

@BrianPugh thanks for the inputs. Can you confirm if the wheels created by the new pipeline (contained in the following artifacts: https://github.com/jrast/littlefs-python/actions/runs/5786354962?pr=46#artifacts) also work?
With the new pipeline the builds are run in parallel for the different python versions which leads to a runtime of about 15 minuts which is acceptable.

I use poetry in other (private) projects but as of now I prefere the current setup for this project.

@BrianPugh
Copy link
Collaborator

just tested littlefs_python-0.6.2.dev10+gf95a12b-cp311-cp311-macosx_11_0_arm64.whl and it installed fine!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants