How to call commands from within

Concept

Commands defined with Litr should be callable with Litr.

To call a command defined through Litr itself you could write it as you would in your shell:

[commands.build]
script = "cmake ..."

[commands.run]
script = [
  "litr build",
  "./build/start"
]

Easy to use and reason about
No extra logic needed, it would just work
- To be exact, this will work no matter what. Question is if this should be the documented way

What if a command calls itself?
- Is the exit condition up to the user?
- Can this be detected and avoided?
What if a command calls its parent command?
- Does not need to be obvious on nested commands
- This can not be validated as it could also be obfuscated or inside a file that gets called.
Slow
- As this will go through the defined shell, the config will be parsed again to execute a command
- Remark: Slow currently means 5ms

Another way could be with a special syntax.

# Syntax is not final
[commands.build]
script = "cmake ..."

[commands.run]
script = [
  "[[build]]",
  "./build/start"
]

Full control of execution
- This will therefore prevent recursive calls or calling a parent command
Fast
- Command can be called by reference without re-parsing the configuration

Special syntax that needs to be documented and explained to the user
Every script needs to be parsed for the special syntax
No control over parameter
- If not defined with more special syntax to make it possible