#14 download and load cldr data

[wkyoshida]

Description:

• Added dependency to the cldr-annotations-derived-full npm package for the emoji data
• Added step to the Jupyter Notebook to download the latest CLDR data
• Added quick initial code just to illustrate the ability to load CLDR data (further updates/changes incoming)
• Fixed minor in-source doc for gen_autosuggestions()

Some notes:

• I noticed that Unicode actually appears to keep the cldr-json data as separate npm packages 🙌, which is awesome
  • Separate packages seems convenient so Scribe is able to only get the data that it needs
  • Using npm, Scribe could also easily simply reference whatever the tagged latest version of the package is. Getting newest data is a matter of npm install
  • cldr-annotations-derived-full appeared to be the package Scribe would need for the emoji work, but can look into if other packages make sense instead, e.g. one of the other cldr-annotations-* packages
  • Can also add multiple packages if needed (in the package.json file); npm install would get the latest for all
• Had to wrestle a bit with how to pick up the npm command inside Jupyter, but running the nvm shell setup seems to do the trick:
  • . $HOME/.nvm/nvm.sh (btw if nvm is used for node and npm version management, this is often what is run anyways whenever a new shell is opened)
  • Without the above, Jupyter does not have npm in the $PATH due to how the shell process works in Jupyter
  • Been a while since I've used Jupyter; so if there's a better way to pick up binaries/commands from inside Jupyter, let me know 🙌

Quick PR to at least get and load the CLDR data. Processing logic incoming. Let me know any thoughts ✌️

[andrewtavis]

Looks really great so far, @wkyoshida! Good thinking on using the NPM packages for this 😊