Automatically Generating Two Dimensional Reveal.js Documents Using Jupyter Notebooks

One of the things I finally got round to exploring whilst at the Reproducible Research Using Jupyter Notebooks curriculum development hackathon was the ability to generate slideshows from Jupyter notebooks.

The underlying slideshow presentation framework is reveal.js. This uses a 1.5(?) dimensional slide geometry, so slides can transition left to right, or you can transition down to subslides off a single slide.

This got me wondering… could I use a notebook/script to generate a reveal.js slideshow that could provide a convenient way of navigating automatically generated slideshows made up from automatically generated data2text slides?

The 1.5/two dimensional component would mean that slides could be structured by topic horizontally, with subtopic vertically downwards within a topic.

A quick test suggests that this is absolutely doable…

import IPython.nbformat as nb
import IPython.nbformat.v4.nbbase as nb4

test.cells.append(nb4.new_markdown_cell('# Test slide1',metadata={"slideshow": {"slide_type": "slide"}}))
test.cells.append(nb4.new_markdown_cell('# Test slide2',metadata={"slideshow": {"slide_type": "slide"}}))
test.cells.append(nb4.new_markdown_cell('Slide2 extra content line 1\n\nSlide2 extra content line 2'))
test.cells.append(nb4.new_markdown_cell('# Test slide3',metadata={"slideshow": {"slide_type": "slide"}}))
test.cells.append(nb4.new_markdown_cell('Slide3 fragment 1',metadata={"slideshow": {"slide_type": "fragment"}}))
test.cells.append(nb4.new_markdown_cell('Slide3 fragment 2',metadata={"slideshow": {"slide_type": "fragment"}}))
test.cells.append(nb4.new_markdown_cell('# Slide4',metadata={"slideshow": {"slide_type": "slide"}}))
test.cells.append(nb4.new_markdown_cell('Slide4 extra content line 1\n\nSlide4 extra content line 2'))
test.cells.append(nb4.new_markdown_cell('# Slide4 subslide',metadata={"slideshow": {"slide_type": "subslide"}}))


#Generate and render slideshow
!jupyter nbconvert $nbf --to slides --post serve

Let the fun begin…:-)

PS here’s a second pass:

def addSlideComponent(notebook, content, styp=''):
    if styp in ['slide','fragment','subslide']: styp={"slideshow": {"slide_type":styp}}
    else: styp={}
    notebook.cells.append(nb4.new_markdown_cell(content, metadata=styp))

addSlideComponent(test,'# Test2 slide1','slide')
addSlideComponent(test,'# Test slide2','slide')
addSlideComponent(test,'Slide2 extra content line 1\n\nSlide2 extra content line 2')
addSlideComponent(test,'# Test slide3','slide')
addSlideComponent(test,'Slide3 fragment 1','fragment')
addSlideComponent(test,'Slide3 fragment 2','fragment')
addSlideComponent(test,'# Slide4','slide')
addSlideComponent(test,'Slide4 extra content line 1\n\nSlide2 extra content line 1')
addSlideComponent(test,'# Slide4 subslide','subslide')


Author: Tony Hirst

I'm a Senior Lecturer at The Open University, with an interest in #opendata policy and practice, as well as general web tinkering...

%d bloggers like this: