!!! note In most cases you won't need to build this package locally.

Unless you're doing development work on the pytextrank library itself, simply install based on the instructions in "Getting Started".

To set up the build environment locally:

python3 -m pip install -r requirements-dev.txt

We use pre-commit hooks based on pre-commit and to configure that locally:

pre-commit install
git config --local core.hooksPath .git/hooks/

The pre-commit hooks use typing and mypy for type checking. To run these tests specifically:

mypy pytextrank/*.py

The pre-commit hooks use pylint for code checking. To run these tests specifically:

pylint pytextrank/*.py

The pre-commit hooks use codespell to check for spelling errors. To run these tests specifically:

codespell pytextrank/*.py *.md docs/*.md

The pre-commit hooks use bandit to check for security issues. To run these tests specifically:

bandit pytextrank/*.py

This project uses pytest and coverage for unit test coverage. Source for unit tests is in the tests subdirectory.

To run the unit tests:

coverage run -m pytest tests

To generate a coverage report and (providing you have the access token) upload it to the codecov.io reporting site:

coverage report
bash <(curl -s https://codecov.io/bash) -t @.cc_token

Test coverage reports can be viewed at https://codecov.io/gh/DerwenAI/pytextrank

To generate documentation pages, this project uses:

• MkDocs
• makedocs-material
• pymdown-extensions
• MathJax
• Jupyter
• nbconvert
• mknotebooks
• Selenium
• Chrome
• Flask

You will also need to download ChromeDriver for your version of the Chrome browser, saved as chromedriver in this directory.

Source for the documentation is in the docs subdirectory.

To build the documentation:

./bin/nb_md.sh
./pkg_doc.py docs/ref.md
mkdocs build

To preview the generated microsite locally:

./bin/preview.py

Then browse to http://localhost:8000 to review the generated documentation.

To package the generated microsite for deployment on a Flask/WSGI server:

tar cvzf ptr.tgz site/

To update the release on PyPi:

./bin/push_pypi.sh

You can use grayskull to generate a conda-forge recipe:

grayskull pypi pytextrank
mv pytextrank/meta.yaml ./