Basically, is there any way to use the future outside of a Pulumi resource instantiation?

not sure what you mean - the future is a standard Python-ism and you can

await

them in an

async

function

the bridge between async and non-async functions are that you can use the values returned by async functions as inputs into Pulumi resources

But can I use the values returned by async functions outside of a Pulumi resource? Ex:

async def get_restapi(name):
  result = await apigateway.get_rest_api(name=name)
  return result.id

if get_restapi('test-jenkins') == None:
  rest_api = apigateway.RestApi(name, name=name)

Obviously this wouldn't work but that's the functionality I'm looking for

You’re assuming that

apigateway.get_rest_api

returns None when the rest API isn’t found, but I think it’s more likely that it’ll throw an exception instead

If you did want to do this sort of control flow, though, you could move the

if get_restapi

logic inside its own async function and then

await get_restapi

to await the future.

I'll give that a shot, thanks

I'm still too unfamiliar with async functions to do what you said but I was able to get what I wanted using boto3 like so:

apigwclient = boto3.client('apigateway')
existingapigws = apigwclient.get_rest_apis().get('items')

for gw in existingapigws:
  name = gw.get('name')
  if name in names:
    names.remove(name)

So I'm just removing gateways that already exist from my list of ones to create before the main loop

Another question: does Pulumi have a way to import resources like Terraform does?

not yet

at least officially, but there is a hack

Interesting. Thanks for the link, that will probably come in handy later

Hello people, I'm wondering what's the puprose of

props

in

pulumi.ComponentResource()

, documentation says the following : props (Optional[dict]) – An optional list of input properties to use as inputs for the resource. , When checking Node.JS docs for `new ComponentResource()`it doesn't have props and it says : unused . Component resources do not communicate or store their properties with the Pulumi engine. I'm a little bit confused, did I miss something ?

Use get functions to read the existing resource and bring its parameters into the stack, like this:

 export const database = gcp.sql.DatabaseInstance.get('test-database', 'name-in-gcp');

Is there a way to do the same in Python? Trying to come up with a feasible solution for migrating Terraform-controlled resources into a Pulumi stack

do the same get?

ah right, i dont think they are implemented?

at least looking at this

doesn't look like it. I'm now debating trying to learn JS lol

lol, yeah, i'm not going that route for sure 😉

Well thank you again for the link, that worked in JS. Now I'm going to look into using JS to create the state file but then switching the project to Python using the same state file...

i wonder if that will work xD

i wonder if you can just create a similar resource and just edit it into the existing state file

I was looking into that earlier but it seems like more effort and very error-prone

lol it worked 😛

Given the latest changes in terraform 0.12 and the lack of support of provider/engine side python code in pulumi, when will pulumi’s python support add the ability to have callbacks/hooks to python user code during the execution of a pulumi up/refresh/destroy?

you can run code on

pulumi up

for sure

not sure why would you need to run code on refresh though