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

Conda updates #2009

Merged
merged 9 commits into from Nov 9, 2017

Conversation

Projects
None yet
3 participants
@richafrank
Member

richafrank commented Nov 8, 2017

Updated versions of conda tooling

Install newer miniconda - This fixes the failing appveyor build on master.

Removed python 3.4 builds

@richafrank richafrank requested review from freddiev4 and ernestoeperez88 Nov 8, 2017

@coveralls

This comment has been minimized.

coveralls commented Nov 8, 2017

Coverage Status

Coverage remained the same at 87.232% when pulling c86ab29 on conda-updates into b21cb4d on master.

@freddiev4

Awesome work @richafrank 😄 I just had some questions for you below!

package:
name: '{{ name|lower }}'
version: '{{ version }}'

This comment has been minimized.

@freddiev4

freddiev4 Nov 8, 2017

Contributor

Where does lower come from? Does that signify that when the packages are built then conda will read the name as lowercase?

This comment has been minimized.

@richafrank

richafrank Nov 8, 2017

Member

Where does lower come from?

It's built in to the jinja2 interpreter used to handle the templating here.

Does that signify that when the packages are built then conda will read the name as lowercase?

Yea, the name variable (defined above statically as "parameterized") will be lowered before being used as the name of the package. Given that static name, lowering is a noop here, but this is all autogenerated by conda skeleton.

@@ -31,6 +31,7 @@ def main(env, do_upload):
"--python", env['CONDA_PY'],
"--numpy", env['CONDA_NPY'],
"--skip-existing",
"--old-build-string",

This comment has been minimized.

@freddiev4

freddiev4 Nov 8, 2017

Contributor

👍 for adding this arg

@@ -78,7 +68,8 @@ install:
- sed -i "s/scipy==.*/scipy==%SCIPY_VERSION%/" etc/requirements.txt
- conda info -a
- conda install conda=4.1.11 conda-build=1.21.11 anaconda-client=1.5.1 --yes -q
- conda install conda=4.3.30 conda-build=3.0.28 anaconda-client=1.6.3 --yes -q

This comment has been minimized.

@freddiev4

freddiev4 Nov 8, 2017

Contributor

Are there any many main differences/reasons for using these newer versions, aside from fixing the build, that we might care about?

This comment has been minimized.

@richafrank

richafrank Nov 8, 2017

Member

There are definitely new features and bug fixes. Many are optional, like the new build strings and changes in how one pins dependencies (like the numpy C-API), which I haven't tried out. You noticed the consolidated recipes - I regenerated a bunch of ours.

The reason I updated in this PR was to fix our toolchain, in that the old miniconda could no longer work with the package index or update to newer conda versions.

@@ -3,7 +3,6 @@ sudo: false
fast_finish: true
python:
- 2.7
- 3.4

This comment has been minimized.

@freddiev4

freddiev4 Nov 8, 2017

Contributor

Should we still run tests for Py34? (Looking at the CI builds, I'm only seeing Py27 and Py35)

This comment has been minimized.

@richafrank

richafrank Nov 8, 2017

Member

I was seeing build failures from missing dependencies in py34, but we're no longer running anything on py34, so it's ok to remove.

# See
# http://docs.continuum.io/conda/build.html
# for a list of environment variables that are set during the build process.

This comment has been minimized.

@freddiev4

freddiev4 Nov 8, 2017

Contributor

TIL you can move this into the .yaml files. That's cool!

This comment has been minimized.

@richafrank

richafrank Nov 8, 2017

Member

Yep, that's from the newer version of conda-build. I used conda skeleton to re-generate these recipes.

@freddiev4

This comment has been minimized.

Contributor

freddiev4 commented Nov 9, 2017

After reading through the responses, LGTM 👍

@richafrank richafrank merged commit 505c191 into master Nov 9, 2017

2 checks passed

continuous-integration/appveyor/pr AppVeyor build succeeded
Details
continuous-integration/travis-ci/pr The Travis CI build passed
Details

@richafrank richafrank deleted the conda-updates branch Nov 9, 2017

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