constellation/docs/screencasts
2023-09-26 13:08:22 +02:00
..
docker deps: update ubuntu:22.04 Docker digest to aabed32 2023-09-26 13:08:22 +02:00
.gitignore docs: embedd asciinema casts (#1154) 2023-03-11 00:13:32 +01:00
generate-readme-svg.sh cli: change generate-config flag to update-config flag (#1897) 2023-06-28 12:47:44 +00:00
generate-screencasts.sh docs: embedd asciinema casts (#1154) 2023-03-11 00:13:32 +01:00
README.md docs: update screencasts to demo node groups (#2243) 2023-08-16 13:50:31 +02:00
window-frame.svg docs: embedd asciinema casts (#1154) 2023-03-11 00:13:32 +01:00

Screencast / Asciinema

Asciinema is used to automatically generate terminal session recordings for our documentation. To fully automate this we use scripts that utilize expect to interface with different CLI tools, and run them inside a container.

Usage

mkdir -p constellation
./generate-screencasts.sh
sudo chown -R $USER:$USER ./constellation
cd constellation && constellation iam destroy
cd .. && rm -rf ./constellation

This will:

  • build the container
  • run the expect based scripts
  • copy recordings into the assets folder of our docs

To replay the output you can use asciinema play recordings/verify-cli.cast.

Include the generated screencast into our docs using the AsciinemaWidget:

import AsciinemaWidget from '../../src/components/AsciinemaWidget';

<AsciinemaWidget src="/constellation/assets/verify-cli.cast" fontSize={16} rows={20} cols={112} idleTimeLimit={3} preload={true} theme={'edgeless'} />

Then re-build and locally host the docs.

Styling

There are three different locations were styling is applied:

  1. The prompt is styled using ANSI escape codes. More explanation and the actual color codes can be found in Dockerfile.
  2. Player dimensions are passed to the AsciinemaWidget when it's embedded in the docs. Check the asciinema-player for a full list of options.
  3. Everything else is styled via CSS. This includes the option to build a custom player theme.

GitHub README.md

The GitHub README.md doesn't support embedding the JavaScript asciinema-player, therefore we generate an svg file for that use case.

# Make sure to install the converter.
# https://github.com/nbedos/termtosvg
pip3 install termtosvg

# Generate SVG. This takes ~10min, since it actually creates a cluster in GCP.
./generate-readme-svg.sh