Skip to content
task management & automation tool
Python

README.rst

README

https://travis-ci.org/pydoit/doit.png?branch=master https://ci.appveyor.com/api/projects/status/f7f97iywo8y7fe4d/branch/master?svg=true https://coveralls.io/repos/pydoit/doit/badge.png?branch=master Join the chat at https://gitter.im/pydoit/doit

doit - automation tool

doit comes from the idea of bringing the power of build-tools to execute any kind of task

Project Details

license

The MIT License Copyright (c) 2008-2015 Eduardo Naufel Schettino

see LICENSE file

developers / contributors

see AUTHORS file

install

doit is tested on python 3.3, 3.4, 3.5.

The last version supporting python 2 is version 0.29.

$ python setup.py install

dependencies

  • cloudpickle
  • pyinotify (linux)
  • macfsevents (mac)

Tools required for development:

  • git * VCS
  • py.test * unit-tests
  • mock * unit-tests
  • coverage * code coverage
  • epydoc * API doc generator
  • sphinx * doc tool
  • pyflakes * syntax checker
  • doit-py * helper to run dev tasks

development setup

The best way to setup an environment to develop doit itself is to create a virtualenv...

doit$ virtualenv dev
(dev)doit$ dev/bin/activate

install doit as "editable", and add development dependencies from dev_requirements.txt:

(dev)doit$ pip install --editable .
(dev)doit$ pip install --requirement dev_requirements.txt

Note

Windows developers: Due to a bug in wheel distributions pytest must not be installed from a wheel.

e.g.:

pip install pytest --no-use-wheel

See for more information:

tests

Use py.test - http://pytest.org

$ py.test

documentation

doc folder contains ReST documentation based on Sphinx.

doc$ make html

They are the base for creating the website. The only difference is that the website includes analytics tracking. To create it (after installing doit):

$ doit website

The website will also includes epydoc generated API documentation.

spell checking

All documentation is spell checked using the task spell:

$ doit spell

It is a bit annoying that code snippets and names always fails the check, these words must be added into the file doc/dictionary.txt.

The spell checker currently uses hunspell, to install it on debian based systems install the hunspell package: apt-get install hunspell.

profiling

python -m cProfile -o output.pstats `which doit` list

gprof2dot -f pstats output.pstats | dot -Tpng -o output.png

contributing

On github create pull requests using a named feature branch.

Something went wrong with that request. Please try again.