I have a big ansible project where I have separated my files into playbooks, vars files, an inventory folder and roles. These files make extensive use of jinja templates.
For my use case, I need to be able to access the variables in these files outside ansible. So far, I have included copy tasks to dump the variable I need to templated out vars files. Then, in my app I read those with the python yaml module without issue since they are free of jinja templates. The variables I am interested in are the variables in the vars section of a playbook as well as those in vars_files. Such variables, get merged with role defaults.
I am now trying to remove those copy tasks and instead use the Ansible API to get the final value for the variables. This will also have the benefit of fully respecting the precedence rules baked into Ansible. So far, this is what I have come up with.
#!/usr/bin/env python
import code
import os
import readline
import rlcompleter
os.environ["ANSIBLE_CONFIG"] = "ansible/ansible.cfg"
import ansible.constants as C
from ansible.inventory.manager
import InventoryManager
from ansible.parsing.dataloader import DataLoader
from ansible.vars.manager import VariableManager
from ansible.template import Templar
from ansible.playbook import Playbook
loader = DataLoader()
inventory = InventoryManager( loader=loader, sources=["data/worlds/active/inventory/configs"] )
variable_manager = VariableManager(loader=loader, inventory=inventory)
p = Playbook.load( "ansible/plays/instances/main.yml", loader=loader, variable_manager=variable_manager)
readline.parse_and_bind("tab: complete") code.InteractiveConsole(locals=globals()).interact()
I don’t know how to proceed. I am supposed to actually run the tasks in the playbook to manage to get the variables? Can I somehow achieve what I want with the Ansible API without running any task? I don’t care about set_fact or dynamic creation/alteration of the variables.