- Installation
- Supported python
- Usage
- Additional information on using
- Development
- Releases and changes
- Reporting issues
- License and copyright
The tool is for now not published on PyPI. To install, download the latest wheel from the github releases, and install it using pip.
For example: pip install templatepython-0.0.1-py3-none-any.wh.
Note: this project is not yet distributed on PyPI.
We strongly recommend to run it in a dedicated virtual envionment, to avoid potential conflict with other python scripts and libraries installed on the system (python best practice). If you choose for this option, please see the venv documentation.
Any CPython version starting from python 3.10 is supported: when issusing the first version of the project, 3.8 and 3.9 were still supported by the python foundation, but since we're taking benefits from python 3.10 new features, it has been chosen to not support previous versions of the language.
No other python than CPython is officially neither supported not tested ... which doesn't means that it won't work with other implementations of the language.
...
Note
The entry points are configured in [project.script] inside pyproject.toml:
scriptname = "package.module:mainfunction"orscriptname = "package.module:Class.staticmdethod"
...
Development is managed using hatch. If you don't have it, it please go on the hatch installation web site.
Project is configured in pyproject.toml.
From the command line, inside the repository root directory:
-
To start an environment with all the required dependencies:
hatch shell- The script can be executed directly witing that shell (an executable is created)
-
To create a build: hatch build
All those requirements are enforced by configuration in pyproject.toml at the root of the repository.
The decision to change any item of this configuration is left to the main author of the package.
Code linting is performing using the ruff linter provided with hatch.
It has been configured so that all the non-experimental controls are enabled, with some specific rules diabled with good reason.
It shall be checked using hatch fmt.
All the methods and functions shall be typed, mypy strict checking is enabled.
Exceptions shall be documented using # type: ignore comment.
Typing of variables is optional. It becomes mandatory if required by mypy to verify typing of methods and functions.
This shall be checked using hatch run types:check.
All the public features in the code shall be documented. The numpy docstring convention shall be used, wich the exception that, thanks to typing, type documentation is not needed. Please refer to this example.
Documentation of private features is optional, and becomes required if requested by code linting.
It shall be checked using hatch fmt.
We target full decision coverage on all the supported python versions.
Unit testing with statement and decision coverage is executed by
hatch test --cover.
Unit testing on supported python version is executed by hatch test --all.
Make sure that the version number is up-to-date by either editing
__about__.py or by invoking hatch version accordingly.
The command hatch build builds the sdist and wheel packages. Those packages
will then eventually be attached to the github releases.
As soon as the version 0.1.0 is released, every development shall refer a github issue. Upgrade to version minor or version major whall be made via a change request.
Every release shall be developped in her own branch.
The main branch shall be updated, via a pull request, as the last step before publication of the build.
The build shall be created on the upgraded main branch and published as a github version. A corresponding tag shall be created.
Issues shall be reported via the project GIT repository's issues
Apart of those issues, the following limitations are known:
- ...
This is distributed under the terms of the BSD 3-Clause License. See LICENSE.md.
Copyright (C) 2004-Today Michaël Hooreman