10000 [MRG+1] MAINT make it possible to use wheelhouse-uploader by ogrisel · Pull Request #4028 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content

[MRG+1] MAINT make it possible to use wheelhouse-uploader #4028

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 5, 2015

Conversation

ogrisel
Copy link
Member
@ogrisel ogrisel commented Dec 29, 2014

wheelhouse-uploader makes it possible to leverage our CI infrastructure
(travis and appveyor) to automate the generation of binary packages
for OSX and windows (both .exe and .whl packages).

Adding those lines in setup.py make is easy for the release manager
to collect all the prebuilt binary packages (more than 10) that match a specific release
and upload them all at once to PyPI as was done for the 0.15.2 release:

https://pypi.python.org/pypi/scikit-learn/0.15.2

The cloud containers that are used to store the build artifacts of the
CI workers are configured in the setup.cfg file.

More details at: https://github.com/ogrisel/wheelhouse-uploader

The module is not imported unless you execute the following release command:

python setup.py sdist fetch_artifacts upload_all

wheelhouse-uploader makes it possible to leverage our CI infrastructure
(travis and appveyor) to automate the generation of binary packages
for OSX and windows (both .exe and .whl packages).

The cloud containers that are used to store the build artifacts of the
CI workers are configured in the setup.cfg file.

More details at: https://github.com/ogrisel/wheelhouse-uploader
@ogrisel
Copy link
Member Author
ogrisel commented Dec 29, 2014

@amueller you might be interested in this.

@amueller
Copy link
Member
amueller commented Jan 9, 2015

I'll have a look, looks awesome :)

@ogrisel
Copy link
Member Author
ogrisel commented Jan 14, 2015

@GaelVaroquaux this is the PR I talked you about at the coffee machine. It should not be too intrusive as it does not import anything if the custom commands are not found in the command line.

@amueller
Copy link
Member

Don't you need to add wheelhouse-uploader to the travis config for this to work?
And is there travis OS X integration now?
Other than that, LGTM.

@ogrisel
Copy link
Member Author
ogrisel commented Jan 22, 2015

wheelhouse-uploader is already registered in our appveyor.yml config (it builds whl packages for Windows for both .dev versions on master and for tagged versions).

For OSX, it is configured on a separate repo https://github.com/MacPython/scikit-learn-wheels dedicated to building OSX wheels for scikit-learn on (pre-)release tags.

We could do it also in master at some point but this should not impact the change in this PR.

@ogrisel ogrisel changed the title [MRG] MAINT make it possible to use wheelhouse-uploader [MRG+1] MAINT make it possible to use wheelhouse-uploader Jan 22, 2015
@amueller
Copy link
Member

Fair enough :)
[ps your reviews on some of the bug-fix PRs would be very welcome ;)]

@ogrisel ogrisel added this to the 0.16 milestone Feb 25, 2015
@ogrisel
Copy link
Member Author
ogrisel commented Feb 25, 2015

@a 8000 rjoly any comment on this?

@ogrisel
Copy link
Member Author
ogrisel commented Feb 27, 2015

@arjoly @jnothman could anyone of you have a quick look at this? I would like to leverage it for the 0.16 beta release.

ogrisel added a commit that referenced this pull request Mar 5, 2015
[MRG+1] MAINT make it possible to use wheelhouse-uploader
@ogrisel ogrisel merged commit 8dbe3f8 into scikit-learn:master Mar 5, 2015
@ogrisel
Copy link
Member Author
ogrisel commented Mar 5, 2015

Merged to make it easier to prepare the 0.16 beta release.

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.

2 participants
0