In this tutorial we cover the basics of creating and deploying a napari plugin using the cookiecutter template. We will build a plugin providing reader, widget and writer contributions, cover testing workflows, configuration for the napari hub and automatic deploys to PyPi.
Timestamps:
0:00 Introduction
3:11 What are we building
7:24 The cookiecutter
11:58 Configuring secrets for automated deploy
13:34 Touring package files
18:43 Reader
25:32 Widget
36:44 Writer
41:26 Testing
46:48 Precommit checks & preparing for PR
53:22 napari hub config
58:30 Merge & deploy
We assume you have:
- a GitHub account and working knowledge of `git`
- a PyPI account
- some previous exposure to napari and/or its plugins
- some experience with `conda` environments, and an environment built using this environment file: gist.github.com/DragaDoncila/dc3e197552cd0d3c4f8bee293c7764de
- a basic understanding of 2D+T imaging datasets and what image segmentation is
- downloaded some 2D+T data from the tracking challenge datasets: celltrackingchallenge.net/2d-datasets/
Useful links:
- the napari cookiecutter: github.com/napari/cookiecutter-napari-plugin
- magicgui docs: napari.org/magicgui/
- napari manifest reference: napari.org/plugins/manifest.html
- manifest contribution reference: napari.org/plugins/contributions.html
- plugin testing workshop: chanzuckerberg.github.io/napari-plugin-accel-workshops/workshops/january
- all the code we wrote today: github.com/DragaDoncila/tracking-challenge-demo
- plugin best practices: napari.org/plugins/best_practices.html
- setting up napari hub preview: github.com/chanzuckerberg/napari-hub/blob/main/docs/setting-up-preview.md
- configuring napari hub metadata: github.com/chanzuckerberg/napari-hub/blob/main/docs/customizing-plugin-listing.md