ipyslides 0.9.8

Live rich content slides in jupyter notebook

IPySlides

Create Interactive Slides in Jupyter/Voila with all kind of rich content.

Launch example Notebook

See PDF-Slides

Changelog

Content below assumes you have ls = LiveSlides().

0.9.8

• PDF printing is optimized. See PDF-Slides. You can hover over top right corner to reveal a slider to change view area while taking screenshot. Also you can select a checkbox from side panel to remove scrolling in output like code.
• You can now display source code using context manager slides.source.
• You can (not recommended) use browser's print PDF by pressing key P in jupyterlab but it only gives you current slide with many limitations, e.g. you need to collect all pages manually.

0.9.6

• Code line numbering is ON by default. You can set ls.code_line_numbering(False) to turn OFF.
• Add slides in for loop using slides.enum_slides function. It create pairs of index and slides.

PDF Printing (Tested on Windows)

• PDF printing is now available. Always print in full screen or set bbox of slides. Read instructions in side panel. PDF-Slides

0.9.5

• You can now give function/class/modules etc. (without calling) in write and source code is printed.
• Objects like dict/set/list/numpy.ndarray/int/float etc. are well formatted now.
• Any object that is not implemented yet returns its __repr__. You can alternatively show that object using display or library's specific method.

0.9.4

• Now you can set logo image using ls.set_logo function.
• LaTeX's Beamer style blcoks are defined. Use ls.block(...,bg='color'), or with few defined colors like ls.block_r, ls.block_g etc.
• @ls.slides no more support live calculating slides, this is to avoid lags while presenting.

0.9.3

• Add custom css under %%slide as well using ls.write_slide_css.
• Slides now open in a side area in Jupyterlab, so editing cells and output can be seen side by side. No more need of Output View or Sidecar.

0.9.1

• In Jupyterlab (only inline cell output way), you can use Ctrl + Shift + C to create consoles/terminals, set themes etc.
• Use Ctrl + Shift + [, Ctrl + Shift + ] to switch back and forth between notebooks/console/terminals and enjoy coding without leaving slides!

0.8.11

• All utilities commnads are now under LiveSlides class too, so you can use either ipyslides.utils.command or ls.command for command in write,iwrite,file2code etc.

0.8.10

• You can add two slides together like ls1 + ls2, title of ls2 is converted to a slide inplace.
• You can now change style of each slide usig **css_props in commands like @ls.slides, with ls.slide and with ls.title.
• A new command textbox is added which is useful to write inline references. Same can be acheived with slides.cite(...here=True).
• You can use ls.alert('text'), ls.colored('text',fg,bg) to highlight text.

0.8.7

• Support added for objects matplotlib.pyplot.Figure, altair.Chart, pygal.Graph, pydeck.Deck, pandas.DataFrame, bokeh.plotting.Figure to be directly in write command.
• write command now can accept list/tuple of content, items are place in rows.

0.8.5

• @ls.slides(...,calculate_now=True) could be used to calculate slides in advance or just in time. Default is True.
• You can now use ipyslides.utils.iwrite to build complex layout of widgets like ipywidgets, bqplot etc. (and text using ipyslides.utils.ihtml).

0.8.3

• You can now use ls.cite method to create citations which you can write at end by ls.write_citations command.
• ls.insert_after no longer works, use

@ls.slides(after_slide_number,*objs)
def func(obj):
    write(obj) #etc. for each obj in objs

decorator which is more pythonic way.

0.8.0 +

Note: All these points may not or only partially apply to earlier versions. So use stable API above version 8.

• Before this version, slides were collected using global namespace, which only allowed one presentation per notebook. Now slides are stored in local namespace, so no restriction on number of slides per notebook.
• To acheive local namespace, functions are moved under class LiveSlide and it registers magics too. So now you will be able to use %%slide, %%title magics. Now you will use context managers as follows

ls = LiveSlides()
ls.convert2slides(True)

with ls.title():
    ...
with ls.slide(<slide number>):
    ...
ls.insert_after(<slide number>,*objs, func)

• ipyslides.initialize() can write all above code in same cell.

Note: For LiveSlides('A'), use %%slideA, %%titleA, LiveSlides('B'), use %%slideB, %%titleB so that they do not overwite each other's slides.

• You can elevate simple cell output to fullscreen in Jupyterlab >= 3.

• with ls.slide content manager is equivalent to %%slide so make sure none of them overwrite each other.

• Auto refresh is enabled. Whenever you execute a cell containing %%title, with ls.title, %%slide, with ls.slide or ls.insert_after, slides get updated automatically.

• LiveSlides should be only in top cell. As it collects slides in local namespace, it can not take into account the slides created above itself.

Install

> pip install ipyslides>=0.8

For development install, clone this repository and then

> cd ipyslides
> pip install -e .

Editable Demo

See a Demo Notebook at Kaggle. You can edit it yourself.

For jupyterlab >= 3, do pip install sidecar for better presenting mode.

Content Types to Embed

You can embed anything that you can include in Jupyter notebook like ipywidgets,HTML,PDF,Videos etc.,including jupyter notebook itself!

Note: Websites may refuse to load in iframe. Note: You can embed one intsnace of slides ls1' inside other instance ls2' using ls2.insert_after(<N>,ls1.box). This is very cool.

IPython Display Objects

Any object with following methods could be in write command: _repr_pretty_, _repr_html_, _repr_markdown_, _repr_svg_, _repr_png_, _repr_jpeg_, _repr_latex_, _repr_json_, _repr_javascript_, _repr_pdf_ Such as IPython.display.<HTML,SVG,Markdown,Code> etc. or third party such as plotly.graph_objects.Figure.

Plots and Other Data Types (0.8.7+)

These objects are implemented to be writable in write command: matplotlib.pyplot.Figure, altair.Chart, pygal.Graph, pydeck.Deck, pandas.DataFrame, bokeh.plotting.Figure. Many will be extentended in future. If an object is not implemented, use display(obj) to show inline or use library's specific command to show in Notebook outside write.

Interactive Widgets

Any object in ipywidgets or libraries based on ipywidgtes such as bqplot,ipyvolume,plotly's FigureWidget can be included in iwrite command. Text/Markdown/HTML inside iwrite is made available through ihtml command.

Full Screen Presentation

• Jupyterlab 3.0+ has full screen eneabled from any view:

  • Direct output of cell can be turned into fullscreen.
  • Linked output view using Create New Output View command is very useful, so you can edit and see output right there.
  • If you have installed Jupyterlab-Sidecar, slides automatically open in it.

• Use Voila for full screen prsentations. Your notebook remains same, it is just get run by Voila, may not work as expected.

• Slides in Jupyter Lab are theme aware in Inherit theme mode, so theme of slides changes based on editor theme.

Very thankful to Python-Markdown which enabled to create write command as well as syntax highliting.