Skip to content
This repository was archived by the owner on May 10, 2024. It is now read-only.

Conversation

@wesm
Copy link
Member

@wesm wesm commented Jul 9, 2016

The toolchain issue is not tractable on an ad hoc basis. I suggest we use a consistent gcc toolchain across the CI environments for Parquet and Arrow, and we can post dev builds for human consumption / testing via conda-forge (where the RHEL devtoolset is there and already working well). Sound good @xhochy?

see also conda-forge/staged-recipes#992

@xhochy
Copy link
Member

xhochy commented Jul 9, 2016

Sounds fine.

One of the things I really like about the manylinux1 PEP is that it provides a clear spec of a common base system and also a tool to check that you actually adhere to this system.

@wesm
Copy link
Member Author

wesm commented Jul 9, 2016

Agreed. FWIW, I'm definitely supportive of having manylinux1 wheels as a primary means of obtaining a working Python stack when things are more "production ready"

@wesm
Copy link
Member Author

wesm commented Jul 9, 2016

+1

@asfgit asfgit closed this in cce6e2d Jul 9, 2016
@wesm wesm deleted the PARQUET-656 branch July 9, 2016 16:26
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants