TODO / FIXME

• better error handling instead of unwrap, esp. on docker commands -> bubble up errors and display then with color_eyre?

• check experimental packages for leap (arm64) (see: bootstrap_leap.sh)

• update from Leap to Spring

• check some use cases from here: https://docs.eosnetwork.com/docs/latest/node-operation/api-node/ can we fulfill them?

• embed deploy_eos_image.py in the binary so we can run "kudune build-image" from anywhere

• use IndexMap and the indexmap feature on configparser to ensure we do not mess up the config file order also ensure we're keeping comments from the config file

• make sure we can generate a decent documentation

• try to follow guidelines from https://clig.dev

README

Here is a list of workflows that should be enabled by the kudune utility and how they can be performed. This should serve both as design document and end-to-end testing of the binary to assess its usefulness / ease-of-use

By default, we run nodeos in a docker container with the name eos_container

build a base image to be used for creating containers with all the tools installed

kudune build-image                # use a default Ubuntu base image
kudune build-image wackou:devbox  # use a provided image

build a new container from scratch

kudune destroy    # ensure that we don't have a lingering docker container
kudune start-node

build a new container with a given config file

kudune destroy    # ensure that we don't have a lingering docker container
kudune start-node --config <CONFIG_FILE.INI>

set our own default config instead of nodeos default

in particular, we want to expose the http port to all listeners, not only localhost

kudune set-config default

you can set specific values like so:

kudune set-config http-server-address=0.0.0.0:8888 chain-state-db-size-mb=65536 contracts-console=true