Description

This Python library provides the user with some tools useful to automate coding of repetitive tasks.

Installation and usage

To install this package use :

% pip install pyreptasks

If you want to install a specific version x.x.x, use:

% pip install pyreptasks==x.x.x

There is not need of other Python packages to correctly run pyreptasks.

To import all the classes and methods of the package in your Python projects:

from pyreptasks import *

You may find examples of usage in the folder examples.

Developing

If you want to contribute, to ensure that you have all the needed dependencies installed in your local environment, install the requirements described in requirement_files/develop_requirements.txt. To do so, you can download the file or just copy it into a develop_requirements.txt file then do

% pip install -r develop_requirements.txt

This project uses black to format code. We support pre-commit to ensure this formatting has been run. If you have installed the requirements, you should have these packages already installed.

Codecov is set up for this project, so for any push or pull request you may check the code coverage report at https://app.codecov.io/gh/tsenovilla/pyreptasks.

Testing

This project uses pytest to run the tests defined in the folder tests. Tests ensure that future releases will not break the former functionality unexpectedly. If some functionality is updated, the tests must be updated accordingly. If a test fails, there are two possibilities: either the tests are not up to date (no problem, it is just an oversight, just make them be coherent with the new version) or something in the code has changed but it should not have been modified. It is up to the developper to determine the actual situation.

To run the tests locally, install the test requirements described in requirement_files/tests_requirements.txt. To do so, you can download the file or just copy it into a tests_requirements.txt then do

% pip install -r test_requirements.txt

To run all the tests, go to the repository where you have stored the folder tests and run:

% pytest

In the case that you would like to run only a specific test, select it individually, for example:

% pytest tests/test_switch.py

Versions log

Version 1.0.0

• New module switch.py: Contains the class Switch. Switch objects may be used to create friendly switch-case structures within a Switch object. This provide the programmer with the advantage of reusing the switch-case along the code if it is needed several times without copy/pasting it. As the object's configuration may be changed along the code, the same Switch object might be used for different switch-case structures sharing some part of their cases. Moreover, this tool might be used to define switch-case structures in Python versions 3.7, 3.8 and 3.9, where the match-case structure was not yet included into the language (version 3.10).

Version 2.0.0

• Update of module switch.py: From this version, integer_switch is not a parameter available for the class Switch anymore. This parameter was used in a Switch object to indicate that the switch had to contain only integer keys. The parameter is deleted due to it was not actually useful.

• Switch examples and tests updates to fit with the new version.

• Other changes:

  • Adoption of the use of setuptools_scm (https://github.com/pypa/setuptools_scm/) package to manage version control instead of versioneer, used in the previous version. This also allows us to get rid of some configuration and version control files, in order to get a simpler package.

  • Adoption of pyproject.toml introduced in the standard PEP 518.

  • Creation of release.yaml to automate releases when a new tag is pushed to GitHub.

Version 2.0.1

• Bug fix: The version 2.0.0 changed the folder's name containing the source code, named pyreptasks, to src, without creating a pyreptasks package inside. This change created a conflict when importing the package, as its name was not anymore pyreptasks but src. This new version solves that oversight by creating a Python package pyreptasks inside the src folder.

• Deletion of the file setup.py as it is not needed anymore with the introduction of pyproject.toml

• The former file requirements.txt might have been misunderstood, as it does not contain the actual requirements to run pyreptasks, but the requirements to make correctly a git commitment. Due to up to this version there is not need of any other Python package to run pyreptasks, this file is now named develop_requirements.txt, and it is included in the new folder requirement_files, with the file containing the packages needed for testing (tests_requirements.txt). If pyreptasks needs another Python package to run in the future, the requirements.txt file would be created again to carry out its actual rol.

• As setup.py is not available anymore, the workflow release.yaml have been updated accordingly.

Version 2.1.0

• New module data_formatter.py: This module introduces the class DataFormatter. An object created with this class allows to define fields that should respect certain format. There are several options to customize the format of each field. Then, the object is able to convert some data string into the format specified by the fields, as well as to ensure that the data respects the fields constraints, otherwise it will throw an exception. Everything is done in a few lines, enabling the developper to skip coding many data controls and format conversions.

Version 3.0.0

• Updates in the module switch.py:

  • The Switch class does not need anymore a list of conditions and switch values. From this version, the switch configuration must be transmitted to the object via a dictionary containing the pairs {conditions : actions}. This makes easier to configure the switch-case structure.

  • Switch attributes cannot be accesed in a direct way from this version. This update protects the programmer against the possibility of changing the values by error, and so getting an unexpected execution of the Switch. New methods have been included allowing the programmer to modify the attributes.

• The workflow ci.yaml has been updated to run the tests in every Python version from 3.7, to ensure the library is compatible with all of them.

• Updates in the module data_formatter.py: As this module uses a Switch object, the code has been updated to fit with the new version.

• Tests and examples have been also updated to fit with the new version.

Version 3.1.0

• Updates in the module data_formatter.py:

  • The attribute fields cannot be accesed in a direct way from this version. This update protects the programmer against the possibility of changing the values by error, and so getting an unexpected execution of the DataFormatter object. A new method field_format has been added to show the format of a specific field introduced as input.

  • The method add_field has a new argument called required. As its name suggests, this field tells the formatter that the field must contain some characters to be valid.

  • Bug fixs:

    1. The arguments length_range and value_range of the add_field method could be not well-formed in previous versions and the method did not check them correctly. In particular, the method checked that they were lists of length 0 or 2, but did not check their content. From this version, the list must be formed of actual range bonds.

    2. Fields did not accept empty strings even if the length range specified that length zero is accepted. This was produced because native methods such isalpha or int() does not support empty strings.

• Tests and examples have been also updated to fit with the new version.

Workflows

• CI (Continuous Integration): This workflow runs the test files contained in tests using pytest in each Python version from 3.7, ensuring that the library might be used in all of them. It also runs code coverage to ensure the whole code is being used.

• Pre-commit: This workflow ensures that all Python files respect PEP 8 formatting, via black.

• Build distribution: This workflow ensures that each time a push/pull request is made, the project is in a packageable state. Furthermore, if a tag is pushed, the workflow publishes the package on PyPI and Conda.

Other

Author: Tomás Senovilla Polo

Email : tspscgs@gmail.com

License: pyreptasks is available under the MIT license. See LICENSE.txt for more information.

If you have in mind some functionality deserving to be included in pyreptasks, let me know it! I will be happy to include it in future releases 😄