Copyright (c) 2020, 2021, DCSO Deutsche Cyber-Sicherheitsorganisation GmbH
DCSO Portal Python Software Development Kit (SDK) helps you create application which need to interact with the DCSO Portal.
Documentation: https://dcso.github.io/dcso-portal-python-sdk/dcso/portal/
This library is, when installed, pure Python. We use only 3rd party tools for installation and documentation generation.
- Python v3.7 or greater (Python v3.6 as best effort)
- Requirements:
- for distribution, see
requirements_dist.txt
- for distribution, see
Requirements can be installed using, for example, pip
:
$ pip install -r requirements_dist.txt
We strongly advise to use a Virtual Environment for testing or for development. This makes sure your local Python environment stays clean.
$ python3 -m venv venvdir
$ source ./venvdir/bin/activate
(venvdir)$ pip install -r requirements.txt
This section explains how to install the DCSO Portal Python SDK. You can get a copy and source on GitHub.
$ pip install -U dcso-portal-python-sdk
Installation is done using setup.py
:
$ pip install .
or
$ python3 setup.py install
It is advised, when developing or testing, to use a Virtual Environment (see Prerequisites in this document).
Documentation is available as HTML in the docs/dcso/portal
folder.
It is possible to generate it using:
$ ./gendocs.py
Alternatively, it is also possible to have the documentation service by the pdoc3 web server. This is mostly useful for working on the documentation since it will rebuild automatically:
$ PYTHONPATH="./lib" pdoc --http : --html \
--template-dir support/doc_templates dcso.portal
open http://localhost:8080/dcso.portal/
See pdoc --help
for more information.
The folder examples
contains code which can be used to learn how to
use this SDK.
When running the examples, within the root of the repository, use
the PYTHONPATH
variable:
$ PYTHONPATH="lib" python3 examples/basic.py
The portal_cli.py
, Portal Command-Line Intereface, show how to create an
interactive client.
It will prompt for username, password, and if needed, also the one-time password.
The result is showing all permissions the user has, and whether the user
has TDH Coordinator access.
The exec_graphql.py
shows how to connect using a Machine Token (also known as
API Tokene), and execute some GraphQL.
The SDK itself and its tests stays as pure Python as possibly can be.
An exception is the Setuptools requirement for installation or any tools
for packaging.
We use the dcso
package namespace, which can be recognized lacking the
__init__.py
. This helps having other Python project which can then
install in the same location.
Directly using the unittest
package from command line:
$ PYTHONPATH="lib" python3 -m unittest discover -s ./lib/dcso/ -vv
Or using the wrapper script, to run all tests:
$ python3 test.py
Right click on the lib/
folder in the root of the repository and
select 'Mark Directory as > Sources Root'. This will make sure PyCharm
add the code to its PYTHONPATH
.
DCSO Portal Python SDK is make available for downloading from GitHub as source distributions are made available with each release.
We also upload to the Python Package Index (PyPI) using twine
, and an API token.
It is a good idea to start with a fresh Virtual Environment when creating a new release.
- Install requirements:
pip install -r requirements_dist.txt
- Create a supporting branch which contains the version number to be releases,
for example,
release/1.0.0-beta3
- Update
lib/dcso/portal/_version.py
and update the__version__
variable - Generate documentation:
./gendocs.py
- Create source distribution:
python setup.py sdist --formats=zip
(will be created in sub folderdist
)
With the source distribution created as ZIP-archive, we can now upload to PyPI using the Twine module.
python3 -m twine upload dist/dcso_portal_python_sdk-1.0.0b3.zip
The twine
module is installed using pip install -r requirements_dist.txt
.
Twine will ask you for credentials:
- Use
__token__
as username - As Password, enter the PyPI API Token for the
dcso-portal-python-sdk
project (note that this includes thepypi-
prefix)