Skip to content

Use manylinux1 for linux wheels - #10873

Merged
eteq merged 1 commit into
astropy:masterfrom
astrofrog:use-manylinux1
Oct 16, 2020
Merged

eteq merged 1 commit into
astropy:masterfrom
astrofrog:use-manylinux1

Conversation

@astrofrog

Copy link
Copy Markdown
Member

I made this fix already for v4.0.3 but I think we might want to do it for v4.1 - basically manylinux2010 wheels require pip 19 or later which is maybe a little too restrictive, so I think for now and until we make a conscious decision to move to manylinux2010 we should use manylinux1.

Does anyone have a good argument for using manylinux2010?

@pllim pllim left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it is reasonable.

I am only afraid if it is manylinux2020 . 🔥

@pllim

pllim commented Oct 15, 2020

Copy link
Copy Markdown
Member

Not sure what is failing this time...

@saimn saimn left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fine for me. We should consider switching to manylinux10 in the future, but it's probably better to wait a little longer.

@Cadair

Cadair commented Oct 16, 2020 •

Copy link
Copy Markdown
Member

I think our minimum supported pip can be driven our minimum Python version, as the "latest stable pip is bundled with CPython". I think therefore we should be able to move to manylinux2010 when we drop 3.6. (Not that I can find a reference for what version was bundled with 3.7.0)

@eteq eteq left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I agree it's a good idea to do this.

My biggest concern is that this impacts people using docker since the "canonical" way to do this would be to use the appropriate CentOS, which is well past end-of-life now. But I don't think it's a true blocker, and @Cadair's "last pip at time of oldest python" seems like a good standard that puts this not too far off.

@eteq

eteq commented Oct 16, 2020

Copy link
Copy Markdown
Member

Looks like we're all in accord on this being the right thing to do for 4.1 so I will merge and backport now.

@eteq
eteq merged commit 6dbff70 into astropy:master Oct 16, 2020
eteq added a commit that referenced this pull request Oct 16, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants