This page contains information about GEMSEO development and how to contribute to it. The source code of GEMSEO is available on gitlab, this is the place where code contributions shall be submitted. Note also that it is required to accompany any contribution with a Developer Certificate of Origin, certifying that the contribution is compatible with GEMSEO software licence.
We aim to have industrial quality standards for software, in order to:
have a good and running software,
be confident about it,
facilitate collaborative work with the team,
facilitate distribution to our partners.
To meet these goals, we use best practices described below, these practices are not optional, they are fully part of the development job.
First time setup:
Then on Linux
tox -e dev
or, on Windows
tox -e dev-win
Run the tests
tox -e py27,py37
tox -e py27-win,py37-win
We use tox for handling the environments related to the development, be it for coding, testing, documenting, checking … This tool offers a simplified and high level interface to many ingredients used in development, while providing reproducible and isolated outcomes that are as much independent as possible of the platform and environment from which it is used.
All the settings of tox are defined in the file
It contains the descriptions of the environments:
version of Python to use,
packages to install,
environment variables to set or pass from the current environment,
commands to execute.
All the directories created by tox
are stored under
.tox next to
.tox contains the environments
in directories named after the environments.
conda create -n tox python=3.8 pip conda activate tox pip install tox-conda conda deactivate conda activate tox
The last two commands are necessary to have the tox executable available in the just created environment.
The MATLAB Python API is not defined as a dependency of GEMSEO, it has to be installed manually in the Anaconda environment. The Python API usually needs to be built and installed since it is not done by default during the MATLAB installation.
For testing with tox,
set the environment variable
to point to the path to a
pip installable version of the MATLAB Python API,
with eventually a conditionnal dependency on the Python version:
export MATLAB_PYTHON_WRAPPER="<path or URL to MATLAB Python API package> ; python_version<'3.9'"
Like the MATLAB Python API, the pSeven one shall be installed manually in the Anaconda environment.
For testing with tox,
set the environment variable
to point to the path to a
pip installable pSeven Python API.
Set the environment variable
for the pSeven license.
How to use tox¶
The environments created by tox and their usage are described in the different sections below. In this section we give the common command line usages and tips.
Create the environment named env and run its commands with:
tox -e env
The first invocation of this command line may take some time to proceed, further invocations will be faster because tox shall not create a new environment from scratch unless, for instance, some of the dependencies have been modified.
You may run (sequentially) more than one environment with:
tox -e env,env2,env3
Recreate an existing environment with:
tox -e env -r
This may be necessary if an environment is broken or if tox cannot figure out that a dependency has been updated (for instance with dependencies defined by a git branch).
conda activate .tox/env
An Anaconda environment created by tox has no Anaconda name, thus conda cannot activate it by its name as usual.
Activating environments may be useful for instance
to investigate a particular issue that happens
in a specific environment and not others.
You may modify an activated environment
just like any other anaconda environment,
in case of trouble just recreate it.
Be aware that the environment variables defined in
will not be set with a manually activated environment.
Show available environments with:
Use a double
-- to pass options to an underlying command,
tox -e env -- ARG1 --opt1
Not all the environments allow this feature, see the specific topics below for more information.
On Windows, the environment names shall be suffixed with -win. This is a limitation of tox.
Create the development environment:
tox -e dev
tox -e dev-win
This will create an environment with:
GEMSEO installed in editable mode,
all the GEMSEO dependencies,
tools used for development (debugging, code checking and formatting)
git settings (see Git)
With an editable installation, GEMSEO appears installed in the development environment created by tox, but yet is still editable in the source tree.
You do not need to activate this environment for contributing to GEMSEO.
We use the pep8 convention. The formatting of the source code is done with isort and black. The code is systematically checked with flake8 and on demand with pylint. A git commit shall have no flake8 violations.
Except for pylint, all these tools are used:
either automatically by the git hooks when creating a commit,
or manually by running tox -e style.
Use tox -e pylint to run pylint.
- String formatting
Loggers shall be defined at module level and named after the module with:
LOGGER = logging.getLogger(__name__)
This means that logger names track the package/module hierarchy, and it’s intuitively obvious where events are logged just from the logger name.
- Error messages
Error messages will be read by humans: they shall be explicit and valid sentences.
We use the gitflow for managing git branches. For the daily work, this basically means that evolutions of GEMSEO are done in feature branches created from the develop branch and merged back into it when finished.
When a commit is being created, git will perform predefined actions:
remove the trailing whitespaces,
fix the end of files,
check toml, yaml and json files are well formed,
check that no big file is committed,
check bad symbolic links,
check or fix some of the python docstrings formatting,
fix the Python import order,
fix the Python code formatting,
check for Python coding issues (see Coding Style),
check the commit message (see Commit message),
check for forbidden
check for misused
check or fix license headers
Those actions will eventually modify the files about to be committed. In this case your commit is denied and you have to check that the modifications are OK, then add the modifications to the commit staged files before creating the commit again.
We use conventional commits for writing clear and useful git commit messages. The commit message should be structured as follows:
<type>(optional scope): <description> [optional body] [optional footer(s)]
<type> defines the type of change you are committing
feat: A new feature
fix: A bug fix
docs: Documentation only changes
style: Changes that do not affect the meaning of the code
refactor: A code change that neither fixes a bug nor adds a feature
perf: A code change that improves performance
test: Adding missing tests or correcting existing tests
build: Changes that affect the build system or external dependencies
ci: Changes to our CI configuration files and scripts
(optional scope) provide additional contextual information and is contained within parentheses
<description> is a concise description of the changes, imperative, lower case and no final dot
[optional body] with the motivation for the change and contrast this with previous behavior
[optional footer(s)] with information about Breaking Changes and reference issues that this commit closes
Commit message examples:
feat(study): open browser when generating XDSM
fix(scenario): xdsm put back filename arg
Commit best practices¶
The purpose of these best practices is to ease the code reviews, commit reverting (rollback changes) bisecting (find regressions), branch merging or rebasing.
- Write atomic commits
Commits should be logical, atomic units of change that represent a specific idea as well as its tests. Do not rename and modify a file in a single commit. Do not combine cosmetic and functional changes in a single commit.
- Commits history
Try to keep the commit history as linear as possible by avoiding unnecessary merge commit. When possible, prefer rebasing over merging, git can help to achieve this with:
git config pull.rebase true git config rerere.enabled true
- Rework commit history
You may reorder, split or combine the commits of a branch. Such history modifications shall be done before the branch has been pushed to the main repository.
Avoid commits that break tests, only push a branch that passes all the tests for py27 and py38 on your machine.
Testing is mandatory in any engineering activity, which is based on trial and error. All developments shall be tested:
this gives confidence to the code,
this enables code refactoring with mastered consequences: tests must pass!
Tests writing guidelines¶
Follow the Arrange, Act, Assert, Cleanup steps by splitting the testing code accordingly. Limit the number of assertions per test functions in a consistent manner by writing more test functions. Use the pytest fixtures features. Tests shall be independent, any test function shall be executable alone.
- Skipping under Windows
Use the pytest marker like:
@pytest.mark.skip_under_windows def test_foo():
- Validation of images
For images generated by matplotlib, use the
image_comparison()decorator provided by the matplotlib testing tools. See
tests/post/dataset/test_surfaces.pyfor an example.
For Python 2.7, run the tests with:
tox -e py27
tox -e py38 -- --last-failed --step-wise
Run the tests for several Python versions with for instance (on Linux):
tox -e py27,py38
-win to the names of the test environments,
tox -e py27-win,py38-win
For a selected python version (for instance Python 3.8), get the coverage information with:
tox -e py38 -- --cov --cov-report=term
See pytest-cov for more information.
Generating the doc¶
The documentation is written with sphinx. On Linux, generate the documentation with:
tox -e doc
-win to the names of the this environment.
Pass options to
tox -e doc -- -vv -j2
Check the links in the generated documentation with:
tox -e doc-linkchecker
doc-linkchecker does not work on windows.
Documenting classes, functions, methods, attributes, modules, etc… is mandatory. End users and developers shall not have to guess the purpose of an API and how to use it.
Use the Google Style Docstrings format for documenting the code. This Example Google Style Docstrings shows how to write such docstrings. Older docstrings use the legacy epydoc docstrings format which is visually dense and hard to read. They will be overhauled progressively.
For functions and methods, write type hints with inlined comments as shown in Example Google Style Docstrings (this is compatible with both Python 2.7 and 3.6+). The type hints are used when generating the functions and methods documentation, they will also be used gradually to check and improved the code quality with the help of a type checker like mypy.
Functions and methods arguments shall use standard duck typing.
In practice, use
List when appropriate,
Mapping instead of
use only the value types with no container.
Return types shall match exactly the type of the returned object.
Type hinting may cause circular imports,
if so, use the special constant
False by default
True when type checking:
from typing import TYPE_CHECKING if TYPE_CHECKING: from gemseo.api import create_discipline
Use semantic linefeeds by starting a new line at the end of each sentence, and splitting sentences themselves at natural breaks between clauses, a text file becomes far easier to edit and version control. You can have a look at the current page’s source for instance.
Have a look to the uncertainty module for an example of proper code documentation.
We use semantic versioning for defining the version numbers of GEMSEO. Given a version number MAJOR.MINOR.PATCH, we increment the:
MAJOR version when we make incompatible API changes,
MINOR version when we add functionality in a backwards compatible manner, and
PATCH version when we make backwards compatible bug fixes.
Use pyperf to create valid benchmark, mind properly tuning the system for the benchmark (see the docs).
The Python standard library provides a profiler, mind using it with controlled system like for benchmarking. The profiling data could be analyzed with one of these tools: