Skip to main content

A flexible pairwise tool written in Python.

Project description

https://badge.fury.io/py/covertable.svg https://github.com/walkframe/covertable/actions/workflows/python.yaml/badge.svg https://codecov.io/gh/walkframe/covertable/branch/master/graph/badge.svg https://img.shields.io/badge/code%20style-black-000000.svg https://img.shields.io/badge/License-Apache%202.0-blue.svg

Requirements

  • Python: 3.3 or later.

    • Tested with 3.7, 3.11

Installation

$ pip install covertable

Usage

Just import covertable and call make function.

>>> from covertable import make, sorters, criteria

>>> machine_list = ['iphone', 'pixel']
>>> os_list = ['ios', 'android']
>>> browser_list = ['FireFox', 'Chrome', 'Safari']
>>> # list input and output
>>> make(
...     [machine_list, os_list, browser_list],  # list factors
...     length=2,  # default: 2
...     sorter=sorters.random,  # default: sorters.hash
...     criterion=criteria.simple,  # default: criteria.greedy
...     seed=100,  # default: ''
...     pre_filter=lambda row: not(row[1] == 'android' and row[0] != 'pixel') and not(row[1] == 'ios' and row[0] != 'iphone'),  # default: None
... )
[
  ['pixel', 'android', 'Safari'],
  ['iphone', 'ios', 'Chrome'],
  ['iphone', 'ios', 'Safari'],
  ['pixel', 'android', 'Chrome'],
  ['pixel', 'android', 'FireFox'],
  ['iphone', 'ios', 'FireFox']
]


>>> # dict input and output
>>> make(
...     {'machine': machine_list, 'os': os_list, 'browser': browser_list},  # dict factors
...     length=2,  # default: 2
...     tolerance=3,  # default: 0
...     post_filter=lambda row: not(row['os'] == 'android' and row['machine'] != 'pixel') and not(row['os'] == 'ios' and row['machine'] != 'iphone'),  # default: None
... )
[
  {'machine': 'pixel', 'browser': 'Chrome', 'os': 'android'},
  {'machine': 'pixel', 'browser': 'FireFox', 'os': 'android'},
  {'machine': 'iphone', 'os': 'ios', 'browser': 'Chrome'},
  {'os': 'ios', 'browser': 'FireFox', 'machine': 'iphone'}
]

Options

covertable.make function has options as keyword argument.

All options are omittable.

length

Number of factors to be covered. (default: 2)

Obviously the more it increases, the more number of combinations increases.

sorter

Combinations depend on the order of spreading all over the rows.

You can choice a sorter from the following:

sorters.random:

This makes different combinations everytime. (fastest)

sorters.hash:

This makes combinations depending on hash of the pair and seed. (default)

  • It receives seed and useCache options.

    • seed option decides the order of storing from unstored pairs, therefore it outputs the same result every time when number of factors and seed are the same.

    • useCache option decide if using cache of hash or not. (default: true)

      • It is around 10% faster than setting useCache off.

criterion

criteria.simple:

This extracts any pairs that can be stored into the processing row.

criteria.greedy:

This attempts to make most efficient combinations. (default)

  • These combinations are not always shorter than simple criterion.

  • It receives tolerance option.

pre_filter

This means a function to filter beforehand.

It receives an argument row as object type.

When the function returns False, the row combination will not be registered.

  • If factors type is Array, you should specify an index at the subscript like row => row[1] < 6.

  • If factors type is Object, you should specify a key at the subscript like row => row['month'] < 6

post_filter

This means a function to filter later.

The usage is the same as preFilter, only the difference is the timing of the call. It will delete rows not matched this function at the last.

For this reason, the final test cases may not satisfy the factors coverage.

Development

# preparation
$ python3 -m venv venv
$ source venv/bin/activate
(venv) $ pip install -r dev_requirements.txt

# testing
(venv) $ pytest

Publish

(venv) $ python setup.py sdist bdist_wheel
(venv) $ twine upload --repository pypi dist/*

More info

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

covertable-2.1.0.tar.gz (7.9 kB view details)

Uploaded Source

Built Distribution

covertable-2.1.0-py3-none-any.whl (7.1 kB view details)

Uploaded Python 3

File details

Details for the file covertable-2.1.0.tar.gz.

File metadata

  • Download URL: covertable-2.1.0.tar.gz
  • Upload date:
  • Size: 7.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.10

File hashes

Hashes for covertable-2.1.0.tar.gz
Algorithm Hash digest
SHA256 e123e2206fc02b8a7b997ec8922f13d8fa1ba436ff5e8de522a5fa9fc72f4d41
MD5 bfb6322b8c8836f6c5aa264e1a4afda9
BLAKE2b-256 4819b3a67b537c66ce07a38d1038ee34f851a583847ec8a24dbb9a9aa9a0eebd

See more details on using hashes here.

File details

Details for the file covertable-2.1.0-py3-none-any.whl.

File metadata

  • Download URL: covertable-2.1.0-py3-none-any.whl
  • Upload date:
  • Size: 7.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.10

File hashes

Hashes for covertable-2.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 44c986d000ea3c41789e876da29ac94d091a275c78b3a110edbdb38768631fe7
MD5 d9ea270265684a198afb4d1b63874d9d
BLAKE2b-256 cb9320f3fb904ee02daad036b3d2cee6cc1185755c27817aa4b39ec8fe10a3ea

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page