Skip to content

Latest commit

 

History

History
83 lines (62 loc) · 4.84 KB

developer_setup.md

File metadata and controls

83 lines (62 loc) · 4.84 KB
Raw

Developer/Contributor Setup

Repo organization

This repository is organized as follows:

  1. hamilton/ is code to orchestrate and execute the graph.
  2. tests/ is the place where unit tests (or light integration tests) are located.

How to contribute

  1. Checkout the repo. If external to Stitch Fix, fork the repo.
  2. Create a virtual environment for it. See python algo curriculum slides for details.
  3. Activate the virtual environment and install all dependencies. One for the package, one for making comparisons, one for running unit tests. I.e. pip install -r requirements*.txt should install all three for you.
  4. Make pycharm depend on that virtual environment & install required dependencies (it should prompt you because it'll read the requirements.txt file).
  5. brew install pre-commit if you haven't.
  6. Run pre-commit install from the root of the repository.
  7. Create a branch off of the latest master branch. git checkout -b my_branch.
  8. Do you work & commit it.
  9. Push to github and create a PR.
  10. When you push to github circle ci will kick off unit tests and migration tests (for Stitch Fix users only).

How to run unit tests

You need to have installed the requirements-test.txt dependencies into the environment you're running for this to work. You can run tests two ways:

  1. Through pycharm/command line.
  2. Using circle ci locally. The config for this lives in .circleci/config.yml which also shows commands to run tests from the command line.

Using pycharm to execute & debug unit tests

You can debug and execute unit tests in pycharm easily. To set it up, you just hit Edit configurations and then add New > Python Tests > pytest. You then want to specify the tests/ folder under Script path, and ensure the python environment executing it is the appropriate one with all the dependencies installed. If you add -v to the additional arguments part, you'll then get verbose diffs if any tests fail.

Using circle ci locally

You need to install the circleci command line tooling for this to work. See the unit testing algo curriculum slides for details. Once you have installed it you just need to run circleci local execute from the root directory and it'll run the entire suite of tests that are setup to run each time you push a commit to a branch in github.

Pushing to pypi

These are the steps to push to pypi. This is taken from the python packaging tutorial.

  1. Have an account & be granted the ability to push to sf-hamilton on testpypi & real pypi.

  2. Setup API tokens and add them to your ~/.pypirc.

  3. Run python3 -m pip install --upgrade build.

  4. Run python3 -m pip install --upgrade twine

  5. Run python3 -m build to build Hamilton. It should create things in dist/*.

  6. Push to test pypi - python3 -m twine upload --repository testpypi dist/*.

    Note: you cannot push the same version twice to test or real pypi. So it's useful to append -rcX to the version. Once you're happy, you can remove that; just remember to not check that in.

  7. Validate you can install from testpypi - follow the URL output.

  8. If you can, then push to real pypi. python3 -m twine upload dist/*

  9. Double check you can download and install what you just pushed in a fresh environment. A good thing to test is to run the hello world example.

Announcing your release

Now that you've pushed to pypi, announce your release! We plan to automate this, but for now...

  1. Create a github release (select auto-generate release for painless text generation). Create a tag that's called sf-hamilton-{version_string}. See 1.2.0 for an example.
  2. Announce on the #announcements channel in slack, following the format presented there.
  3. Thanks for contributing!

Pushing to Anaconda

These are the steps to push to Anaconda after you have built and pushed to PyPi successfully.

  1. Make sure you have conda installed with conda-build. See these instructions. Note: since it is common to have pyenv installed too -- conda and pyenv don't play nice. My suggestion is to run conda config --set auto_activate_base False to not set conda to be active by default.
  2. Make sure you have an Anaconda account and are authorized to push to anaconda.
  3. Log in to anaconda (e.g. conda activate && anaconda login).
  4. We have a script build_conda.sh that is a bash script that encapsulates the steps. For reference it roughly follows this documentation. Run it with bash build_conda.sh. It should "just work".
  5. Be sure to remove any files it creates afterwards so when you come to do a release again, you're not uploading the same files.