Changelog

0.15.0

  • Make user agent header compliant with Wikimedia Foundation User-Agent policy - PR 590

0.14.1

  • pageid for images is deterministic - PR 549

0.14.0

  • Add WikipediaImage, ImagesDict, and imageinfo API - PR 525

0.13.0

  • Pass parameters to the underlying htttp client - PR 520

0.12.0

  • Add API for coordinates, random, search, and geosearch - PR 504

0.11.0

  • Introduce async support - AsyncWikipedia - `PR 479`_

0.10.2

  • Apply security best practices - PR 447

  • Increase CLI test coverage - PR 449

  • Increase code coverage - PR 451

0.10.1

0.10.0

  • Introduce command line tool - wikipedia-cli - PR 402

  • Kwargs are overwriting default API params (breaking) - PR 408

  • Handle HTTP errors, add retries (breaking) - PR 417

  • Split __init__.py into smaller files - PR 422

0.9.0

  • Drop support for Python 3.9 - PR 378

  • Add support for Python 3.14 - PR 378

  • Update dependencies

0.8.1

  • More user friendly error messages to make ChatGPT users happy

0.8.0

  • Allow specifying language variant - Issue 209

  • Allow passing additional parameters for API calls.

  • This breaks the API since variant is now the third parameter.

0.7.3

  • Regenerate stable documentation

0.7.2

  • Add support for Python 3.13

  • Regenerate stable documentation

0.7.1

  • Drop support for Python 3.8

0.6.9

  • Update dependencies

  • Add tests for more platforms

0.6.0

  • Make user agent mandatory - Issue 63

  • This breaks the API since user_agent is now the first parameter.

0.5.8

  • Adds support for retrieving all sections with given name - Issue 39

0.5.4

  • Namespace could be arbitrary integer - Issue 29

0.5.3

  • Adds persistent HTTP connection - Issue 26

    • Downloading 50 pages reduced from 13s to 8s => 40% speed up

0.5.2

0.5.1

  • Adds tox for testing different Python versions

0.5.0

  • Allows modifying API call parameters

  • Fixes Issue 16 - hidden categories

  • Fixes Issue 21 - summary extraction

0.4.5

  • Handles missing sections correctly

  • Fixes Issue 20

0.4.4

  • Uses HTTPS directly instead of HTTP to avoid redirect

0.4.3

  • Correctly extracts text from pages without sections

  • Adds support for quoted page titles

api = wikipediaapi.Wikipedia(
    language='hi',
)
python = api.article(
    title='%E0%A4%AA%E0%A4%BE%E0%A4%87%E0%A4%A5%E0%A4%A8',
    unquote=True,
)
print(python.summary)

0.4.2

  • Adds support for Python 3.4 by not using f-strings

0.4.1

  • Uses code style enforced by flake8

  • Increased code coverage

0.4.0

  • Uses type annotations => minimal requirement is now Python 3.5

  • Adds possibility to use more parameters for request. For example:

api = wikipediaapi.Wikipedia(
    language='en',
    proxies={'http': 'http://localhost:1234'}
)

  • Extends documentation

0.3.4

0.3.3

0.3.2

0.3.1

  • Removing WikipediaLangLink

  • Page keeps track of its own language, so it’s easier to jump between different translations of the same page

0.3.0

  • Rename directory from wikipedia to wikipediaapi to avoid collisions

0.2.4

  • Handle redirects properly

0.2.3

  • Usage method page instead of article in Wikipedia

0.2.2

0.2.1

0.2.0

  • Use properties instead of functions

  • Added support for property Info

0.1.6

  • Support for extracting texts with HTML markdown

  • Added initial version of unit tests

0.1.4

  • It’s possible to extract summary and sections of the page

  • Added support for property Extracts