Getting into monkeypatch land...

err, maybe I'll just try something 🙁 I'm going to have to override an entire module call I think

You can't override what's inside the module

You can override the

module "..." {.. }

itself

Or can I just do a single input

I think maybe?

lol

we have this huge block of data I need to change a single boolean inside

it's not a local

so I suspect I need to atleast override the entire input into the module

atmos terraform shell doesn't produce the files from the

generate

block 🥲

this is without any JIT TTL crap

also - it seems like it doesn't select the terraform workspace either

facing a few weird issues

will put in an issue after I get this thing fixed

Yep, you know the drill 😉

Didn't know about overrides files until yesterday

now I can just do whatever I want on the fly 😈

we were using the provider generation feature before

seems like it's the same sauce

oof I think I found a big bug with JIT, backend generation, and output stores

Not respecting workdir provisioning on apply/plan 😕

hooks:
        store-outputs:
          outputs:
            vpc_properties: .vpc_properties

ℹ Executing command: atmos terraform apply demo-vpc -auto-approve
 INFO  Running hooks event=before.terraform.apply
Fetching vpc_properties output from demo-vpc in dev

x Fetching vpc_properties output from demo-vpc in dev
# Error

**Error:** terraform output returned nil: failed to get terraform output for key
vpc_properties: failed to execute terraform output for component demo-vpc in
stack
dev: failed to generate backend file: open
/tmp/gitrepo.fFs98Y/components/terraform/demo-vpc/backend.tf.json: no such file
or
directory

I removed the hooks and it provisioned the workdir successfully

for some reason init works fine, I'm guessing because it skips something where it needs to run the terraform output command which also needs it to generate the backend.tf. I'm not sure.

I put a couple issues in and am throwing claude at them in my fork

It found a few logic errors with the store hooks, JIT components with generate using base_path AND the workdir. Then for some reason atmos uses the base_path instead of the workdir and fails because it's running apply on a mostly empty directory

Running a local redis container to be my atmos store

Big help

Doing a little more testing but I think this might be working

We're very AI friendly and understanding. 😃

turned out bigger than I thought

but I tested this against my repro, and my real codebase and it seems to handle all of the variables/features cleanly JIT, TTL, workdirs, atmos stores with different event types, terraform base dirs...

@Andriy Knysh (Cloud Posse) you should also review

(also, we should use #C07RMALBX7V for core changes)