Using conda with Travis CI

If you are already using Travis CI, using conda is a preferable alternative to using apt-get and pip to install packages. The Debian repos provided by Travis may not include versions of Python packages for all versions of Python, or may not be updated as quickly. Pip installing such packages may also be undesirable, as this can take a while, eating up a large chunk of the 50 minutes that Travis allows for each build. Using conda also lets you test the building of conda recipes on Travis.

Conda is language-agnostic, so it can be used for anything, not just Python, but the following guide shows how to use it to test a Python package on Travis CI.

The .travis.yml file

The following shows how to modify the .travis.yml file to use Miniconda for a project that supports Python 2.6, 2.7, 3.3, and 3.4.

NOTE: Please see the Travis CI website for information about the basic configuration for Travis.

language: python
python:
  # We don't actually use the Travis Python, but this keeps it organized.
  - "2.6"
  - "2.7"
  - "3.3"
  - "3.4"
install:
  - sudo apt-get update
  # We do this conditionally because it saves us some downloading if the
  # version is the same.
  - if [[ "$TRAVIS_PYTHON_VERSION" == "2.7" ]]; then
      wget https://repo.continuum.io/miniconda/Miniconda2-latest-Linux-x86_64.sh -O miniconda.sh;
    else
      wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh;
    fi
  - bash miniconda.sh -b -p $HOME/miniconda
  - export PATH="$HOME/miniconda/bin:$PATH"
  - hash -r
  - conda config --set always_yes yes --set changeps1 no
  - conda update -q conda
  # Useful for debugging any issues with conda
  - conda info -a

  # Replace dep1 dep2 ... with your dependencies
  - conda create -q -n test-environment python=$TRAVIS_PYTHON_VERSION dep1 dep2 ...
  - source activate test-environment
  - python setup.py install

script:
  # Your test script goes here

Additional Steps

If you wish to support a package that doesn’t have official Continuum builds, you can build it yourself, and add it to a Anaconda.org channel. You can then add

- conda config --add channels your_Anaconda_dot_org_username

to the install steps in .travis.yml so that it finds the packages on that channel.

Building a Conda Recipe

If you support official conda packages for your project, you may want to use conda build in Travis, so the building of your recipe is tested as well. We recommend that you include the conda recipe in the same directory as your source code. Then replace the following:

- python setup.py install

with

- conda build your-conda-recipe
- conda install your-package --use-local

For more information on building conda packages, see the conda build section and the example recipes in the conda-recipes repo.

AppVeyor

An alternative to using Travis CI with conda is AppVeyor, a continuous build service for Windows built on Azure.

You can see an example project for building conda packages on AppVeyor located at https://github.com/rmcgibbo/python-appveyor-conda-example.