Running packaged deploys programmatically

[madisvain]

I would like to use the packaged deployments from within Python code to build a more dynamic deployer system which would have a UI.

How would I go about running the packaged deployments this way? By just calling the packaged deploy I get a exception:
type object 'State' has no attribute 'config'

Is there a way to do this and could an example be provided? In the documentation currently only how to use the CLI is extensively described.
https://docs.pyinfra.com/en/2.x/deploy-process.html

[phlummox]

In case it's useful – I noticed that there's some information on how to use PyInfra via its API here: https://docs.pyinfra.com/en/2.x/api/index.html

However, it's not entirely clear from this how you'd invoke a packaged deploy. In the CLI code, it looks like the following function is responsible for loading a deploy script:

pyinfra/pyinfra_cli/util.py

Lines 191 to 193 in 9256e6a

	def load_deploy_file(state: "State", filename):
	state.current_deploy_filename = filename
	_parallel_load_hosts(state, lambda: exec_file(filename), filename)

It's not entirely clear to me what the invoked _parallel_load_hosts function is doing, but if you look at it, it seems to be passing a pyinfra.api.state.State object to the current Context, and ultimately, compiling the packaged deploy and executing it with Python's builtin exec function (here).

(I have a somewhat similar problem, trying to execute a particular function programmatically – if I end up creating an issue for it, I'll link this one too, in case the answers might be of assistance to you too.)