Lily-Assistant

Set of scripts, git hooks etc. helping us to stick to the high quality standards imposed on our code, tests, documentation, and the way we work with git

Installation

Just copy and execute the below snippet while being in the ROOT directory of your project (where one has direct access to .git, src, tests)

pip install -e git+https://github.com/cosphere-org/lily-assistant.git#egg=lily-assistant

in order to install a particular version of the lily-assistant please use the following:

pip install -e git+https://github.com/cosphere-org/lily-assistant.git@1.0.0#egg=lily-assistant

After the installation of Lily-Assistant one should initialize it (while being in the root of one's project) by running:

lily_assistant init <name_of_src_dir>

Where naturally the <name_of_src_dir> should be replaced by the name of the src folder.

The above operation will install newest git hooks (./.git/hooks directory) and perform so preliminary checks. Finally at the end it will info you about the necessity of adding include .lily/lily_assistant.makefile at the top of your Makefile.

config.json

After running init CLI command lily-assistant will create in the root of the project:

• .lily fodler
• inside which one will find config.json file.

Config file takes the following form:

{
    "name": "lily-assistant",
    "src_dir": "lily_assistant",
    "repository": "https://github.com/cosphere-org/lily-assistant.git",
    "version": "1.0.6",
    "last_commit_hash": "f53dce5eb37d3cf77e33a9d3fbbae35bf9bb1a8d"
}

It serves as a centralized placed for finding all needed meta information regarding the repository itself.

Please notice that "version" and "last_commit_hash" are filled automatically by the upgrade_version_<X> commands covered below.

Makefile commands

Lily-Assitant exposes various helpful Makefile commands:

• make install - for setting up virtualenv and installing all requirements.txt and text-requirements.txt
• make lint - when executed it will run the linter against the tests and source folders
• make test tests=<path to test directory / file> - running selected tests
• make test_all - running all tests
• make inspect_coverage - loads in Chrome browser the html coverage report allowing one to find all lines that are missing coverage etc.
• make upgrade_version_patch - perform PATCH (0.0.X) version update (together with git tag, git push and update of config.json)
• make upgrade_version_minor - perform MINOR (0.X.0) version update (together with git tag, git push and update of config.json)
• make upgrade_version_major - perform MAJOR (X.0.0) version update (together with git tag, git push and update of config.json)

IDE and Testing

Lily-Assitant assumes that one uses py.test for testing therefore if you're triggering your tests to be run by IDE either point them to make test_all or make test test=<path to test directory / file> or use directly the command rendered in the .lily/lily_assistant.makefile

Lily-Assitant Commands

Lily-Assitant itself exposes plenty of super helpful commands. One can inspect them by running:

lily_assistant --help

Required project structure

On each commit attempt Lily-Assitant asserts if the stucture of the project is correct. It probes for the following:

├── <project_name>/
├── tests/
├── .git/
├── env.sh
├── Makefile
├── pytest.ini
├── README.md
├── requirements.txt
├── test-requirements.txt
└── setup.py

Only projects that are fulfilling this structure will pass initial Lily-Assitant checks.

Development (Lily-Assitant itself)

If one is interested in contributing to Lily-Assitant, please run the following to install its all dependencies:

make install

It's assumed that as per python standards the above command will be executed while being in some sort of virtualenv.

Reference

• Naming convention error codes: https://github.com/PyCQA/pep8-naming#plugin-for-flake8

• Docstring convetion error codes: http://www.pydocstyle.org/en/2.1.1/error_codes.html

• General error codes: http://pycodestyle.pycqa.org/en/latest/intro.html#error-codes