Skip to content

Latest commit

 

History

History
447 lines (345 loc) · 14.9 KB

README.rst

File metadata and controls

447 lines (345 loc) · 14.9 KB
PyPI version License Python versions supported Format

Continuous integration test status Continuous integration test coverage Documentation status

Description

This package provides engineering-related classes and functions, including:

  • A waveform class that is a first-class object:

    >>> import copy, numpy, peng
    >>> obj_a=peng.Waveform(
    ...     indep_vector=numpy.array([1, 2, 3]),
    ...     dep_vector=numpy.array([10, 20, 30]),
    ...     dep_name='obj_a'
    ... )
    >>> obj_b = obj_a*2
    >>> print(obj_b)
    Waveform: obj_a*2
    Independent variable: [ 1, 2, 3 ]
    Dependent variable: [ 20, 40, 60 ]
    Independent variable scale: LINEAR
    Dependent variable scale: LINEAR
    Independent variable units: (None)
    Dependent variable units: (None)
    Interpolating function: CONTINUOUS
    >>> obj_c = copy.copy(obj_b)
    >>> obj_a == obj_b
    False
    >>> obj_b == obj_c
    True

    Numerous functions are provided (trigonometric, calculus, transforms, etc.) and creating new functions that operate on waveforms is simple since all of their relevant information can be accessed through properties

  • Handling numbers represented in engineering notation, obtaining their constituent components and converting to and from regular floats. For example:

    >>> import peng
    >>> x = peng.peng(1346, 2, True)
    >>> x
    '   1.35k'
    >>> peng.peng_float(x)
    1350.0
    >>> peng.peng_int(x)
    1
    >>> peng.peng_frac(x)
    35
    >>> str(peng.peng_mant(x))
    '1.35'
    >>> peng.peng_power(x)
    EngPower(suffix='k', exp=1000.0)
    >>> peng.peng_suffix(x)
    'k'
  • Pretty printing Numpy vectors. For example:

    >>> from __future__ import print_function
    >>> import peng
    >>> header = 'Vector: '
    >>> data = [1e-3, 20e-6, 30e+6, 4e-12, 5.25e3, -6e-9, 70, 8, 9]
    >>> print(
    ...     header+peng.pprint_vector(
    ...         data,
    ...         width=30,
    ...         eng=True,
    ...         frac_length=1,
    ...         limit=True,
    ...         indent=len(header)
    ...     )
    ... )
    Vector: [    1.0m,   20.0u,   30.0M,
                         ...
                70.0 ,    8.0 ,    9.0  ]
  • Formatting numbers represented in scientific notation with a greater degree of control and options than standard Python string formatting. For example:

    >>> import peng
    >>> peng.to_scientific_string(
    ...     number=99.999,
    ...     frac_length=1,
    ...     exp_length=2,
    ...     sign_always=True
    ... )
    '+1.0E+02'

Interpreter

The package has been developed and tested with Python 2.7, 3.5, 3.6 and 3.7 under Linux (Debian, Ubuntu), Apple macOS and Microsoft Windows

Installing

$ pip install peng

Documentation

Available at Read the Docs

Contributing

  1. Abide by the adopted code of conduct

  2. Fork the repository from GitHub and then clone personal copy [1]:

    $ github_user=myname
    $ git clone --recurse-submodules \
          https://github.com/"${github_user}"/peng.git
    Cloning into 'peng'...
    ...
    $ cd peng || exit 1
    $ export PENG_DIR=${PWD}
    $
  3. The package uses two sub-modules: a set of custom Pylint plugins to help with some areas of code quality and consistency (under the pylint_plugins directory), and a lightweight package management framework (under the pypkg directory). Additionally, the pre-commit framework is used to perform various pre-commit code quality and consistency checks. To enable the pre-commit hooks:

    $ cd "${PENG_DIR}" || exit 1
    $ pre-commit install
    pre-commit installed at .../peng/.git/hooks/pre-commit
    $
  4. Ensure that the Python interpreter can find the package modules (update the $PYTHONPATH environment variable, or use sys.paths(), etc.)

    $ export PYTHONPATH=${PYTHONPATH}:${PENG_DIR}
    $
  5. Install the dependencies (if needed, done automatically by pip):

  6. Implement a new feature or fix a bug

  7. Write a unit test which shows that the contributed code works as expected. Run the package tests to ensure that the bug fix or new feature does not have adverse side effects. If possible achieve 100% code and branch coverage of the contribution. Thorough package validation can be done via Tox and Pytest:

    $ PKG_NAME=peng tox
    GLOB sdist-make: .../peng/setup.py
    py27-pkg create: .../peng/.tox/py27
    py27-pkg installdeps: -r.../peng/requirements/tests_py27.pip, -r.../peng/requirements/docs_py27.pip
    ...
      py27-pkg: commands succeeded
      py35-pkg: commands succeeded
      py36-pkg: commands succeeded
      py37-pkg: commands succeeded
      congratulations :)
    $

    Setuptools can also be used (Tox is configured as its virtual environment manager):

    $ PKG_NAME=peng python setup.py tests
    running tests
    running egg_info
    writing peng.egg-info/PKG-INFO
    writing dependency_links to peng.egg-info/dependency_links.txt
    writing requirements to peng.egg-info/requires.txt
    ...
      py27-pkg: commands succeeded
      py35-pkg: commands succeeded
      py36-pkg: commands succeeded
      py37-pkg: commands succeeded
      congratulations :)
    $

    Tox (or Setuptools via Tox) runs with the following default environments: py27-pkg, py35-pkg, py36-pkg and py37-pkg [3]. These use the 2.7, 3.5, 3.6 and 3.7 interpreters, respectively, to test all code in the documentation (both in Sphinx *.rst source files and in docstrings), run all unit tests, measure test coverage and re-build the exceptions documentation. To pass arguments to Pytest (the test runner) use a double dash (--) after all the Tox arguments, for example:

    $ PKG_NAME=peng tox -e py27-pkg -- -n 4
    GLOB sdist-make: .../peng/setup.py
    py27-pkg inst-nodeps: .../peng/.tox/.tmp/package/1/peng-1.0.11.zip
    ...
      py27-pkg: commands succeeded
      congratulations :)
    $

    Or use the -a Setuptools optional argument followed by a quoted string with the arguments for Pytest. For example:

    $ PKG_NAME=peng python setup.py tests -a "-e py27-pkg -- -n 4"
    running tests
    ...
      py27-pkg: commands succeeded
      congratulations :)
    $

    There are other convenience environments defined for Tox [3]:

    • py27-repl, py35-repl, py36-repl and py37-repl run the Python 2.7, 3.5, 3.6 and 3.7 REPL, respectively, in the appropriate virtual environment. The peng package is pip-installed by Tox when the environments are created. Arguments to the interpreter can be passed in the command line after a double dash (--).

    • py27-test, py35-test, py36-test and py37-test run Pytest using the Python 2.7, 3.5, 3.6 and 3.7 interpreter, respectively, in the appropriate virtual environment. Arguments to pytest can be passed in the command line after a double dash (--) , for example:

      $ PKG_NAME=peng tox -e py27-test -- -x test_peng.py
      GLOB sdist-make: .../peng/setup.py
      py27-pkg inst-nodeps: .../peng/.tox/.tmp/package/1/peng-1.0.11.zip
      ...
        py27-pkg: commands succeeded
        congratulations :)
      $
    • py27-test, py35-test, py36-test and py37-test test code and branch coverage using the 2.7, 3.5, 3.6 and 3.7 interpreter, respectively, in the appropriate virtual environment. Arguments to pytest can be passed in the command line after a double dash (--). The report can be found in ${PENG_DIR}/.tox/py[PV]/usr/share/peng/tests/htmlcov/index.html where [PV] stands for 2.7, 3.5, 3.6 or 3.7 depending on the interpreter used.

  8. Verify that continuous integration tests pass. The package has continuous integration configured for Linux, Apple macOS and Microsoft Windows (all via Azure DevOps).

  9. Document the new feature or bug fix (if needed). The script ${PENG_DIR}/pypkg/build_docs.py re-builds the whole package documentation (re-generates images, cogs source files, etc.):

    $ "${PENG_DIR}"/pypkg/build_docs.py -h
    usage: build_docs.py [-h] [-d DIRECTORY] [-r]
                         [-n NUM_CPUS] [-t]
    
    Build peng package documentation
    
    optional arguments:
      -h, --help            show this help message and exit
      -d DIRECTORY, --directory DIRECTORY
                            specify source file directory
                            (default ../peng)
      -r, --rebuild         rebuild exceptions documentation.
                            If no module name is given all
                            modules with auto-generated
                            exceptions documentation are
                            rebuilt
      -n NUM_CPUS, --num-cpus NUM_CPUS
                            number of CPUs to use (default: 1)
      -t, --test            diff original and rebuilt file(s)
                            (exit code 0 indicates file(s) are
                            identical, exit code 1 indicates
                            file(s) are different)

Footnotes

[1]All examples are for the bash shell
[2]It is assumed that all the Python interpreters are in the executables path. Source code for the interpreters can be downloaded from Python's main site
[3](1, 2) Tox configuration largely inspired by Ionel's codelog

License

The MIT License (MIT)

Copyright (c) 2013-2019 Pablo Acosta-Serafini

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.