Skip to content

Please note that nbsmoke is still undergoing heavy development, so you should not rely on it, and various features are likely to change.

nbsmoke can be installed via pip install nbsmoke or conda installed from the pyviz channel (conda install -c pyviz nbsmoke) or conda-forge channel (conda install -c conda-forge nbsmoke`).

Please see https://github.com/pyviz/nbsmoke/blob/v0.2.8/README.rst for more information.

Changes since the previous release (v0.2.7):

  • Shorter lint tracebacks
  • Added --nbsmoke-lint-onlywarn: Report lint failures as warnings
  • Added --nbsmoke-lint-debug: Allow easier investigation (and debugging) of lint failures. nbsmoke lint currently uses nbconvert to convert ipynb to py, then runs pyflakes on that. You can now see the py source that was actually flake checked (pyflakes-reported line numbers correspond to this source). (Note: unhandled magics e.g. from holoviews can cause spurious flakes. I have separate changes that attempt to improve the situation. Meanwhile, this option is also helpful for debugging that situation.)
  • Only supporting html storage in python 3.

Full list of changes since the previous release: v0.2.7...v0.2.8

Assets 2
May 1, 2019
Improved lint output and debugging

@ceball ceball released this Jun 1, 2018 · 4 commits to master since this release

This is the first release of nbsmoke, a pytest plugin to test jupyter notebooks. Please note that nbsmoke is still undergoing heavy development, so you should not rely on it, and various features are likely to change.

nbsmoke can be installed via pip install nbsmoke or conda install -c conda-forge nbsmoke).

Please see https://github.com/pyviz/nbsmoke/blob/v0.2.7/README.rst for more information.

Assets 2
You can’t perform that action at this time.