.travis.yml

# Use a docker image for access to Ubuntu 16.04 (more details below)
sudo: required
dist: trusty
services:
  - docker

language: python
python:
  - "3.6"  # This value may be referred to as $TRAVIS_PYTHON_VERSION

before_install:
  # Use Ubuntu 16.04 instead of 14.04 to avoid the error:
  # "ImportError: dlopen: cannot load any more object with static TLS"
  # https://github.com/pytorch/pytorch/issues/2083
  # https://github.com/RobotLocomotion/director/issues/467
  # In short, Ubuntu 16.04 has a larger default allotment of TLS slots.
  - docker pull ubuntu:16.04

  # Check if miniconda is already accessible; if not, download and install it
  - chmod +x ./.travis/maybe_install_miniconda.sh
  - ./.travis/maybe_install_miniconda.sh

  # Add miniconda to PATH
  - export PATH="$HOME/miniconda/bin:$PATH"

  # Set conda settings
  - conda config --set always_yes yes  # For all conda commands, choose 'yes'
  - conda config --set changeps1 no  # Do not show the name of the active env

install:
  # Create conda environment
  - conda env create -f environment.yml
  - source activate metal

  # Add metal/ to PYTHONPATH
  - source add_to_path.sh

  # Run linter/style checks
  - make dev
  - make check

script:
  - nosetests
  - runipy tutorials/Basics.ipynb
  - runipy tutorials/Class_Balance.ipynb
  - runipy tutorials/Multitask.ipynb
  - runipy tutorials/Visualization.ipynb

notifications:
  email:
    recipients:
      - bradenjh@cs.stanford.edu
    on_success: change # default: change [change, always, never]
    on_failure: change # default: always