-
Notifications
You must be signed in to change notification settings - Fork 43
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
PlanResourceChange dirty refresh on empty maps #2047
Comments
|
Ah, this actually reproes in TF after the first apply we get and after the refresh it changes to
|
Ok it does show it but with
|
Some potential ways around this:
|
@t0yv0 pointed out the culprit is likely normalizeNullValues: https://github.com/hashicorp/terraform-plugin-sdk/blob/70fb6b9b15e8e5fc73f424e24084c28fedd1e013/helper/schema/grpc_provider.go#L1565 |
as suggested by @t0yv0 pulumi/pulumi#16146 fixes this |
Possibly related? #1106 |
Not sure if 1106 is very related, it's read returning null when we want it not to, here it's the reverse. The workaround I would like to try here is not likely to solve 1106 though I can test easily. |
Got pulumi/pulumi-aws#4013 to get this fixed for AWS. Curious if other providers satisfied as well. |
Related to #2047 Pulling out the tests from #2065 so that the changes in behaviour in the other PR are more visible and easy to review. This adds tests for refreshes on collections for: - list/set/map - RPC/non-PRC - null/empty/non-empty collections - nil/non-nil cloud values for the nil collections - top-level collections/nested collections - Properties with a value override in Create/ ones without - this is a behaviour observed in AWS and GCP labels and some other providers, mainly seen around maps. The goal of the tests is to: - Ensure we do not regress on behaviour in PRC vs non-PRC - Ensure we do not regress on behaviour in PRC as we change the implementation in #2073 / #2065 I've flattened the matrix in order to allow us to annotate/skip individual test cases.
This is no longer an issue with the new refresh behaviour in the CLI. |
What happened?
On master with PlanResourceChange enabled it looks like empty maps now refresh badly.
Example
TestAccBucketPy
https://github.com/pulumi/pulumi-gcp/actions/runs/9215462858/job/25354722934
Output of
pulumi about
.
Additional context
No response
Contributing
Vote on this issue by adding a 👍 reaction.
To contribute a fix for this issue, leave a comment (and link to your pull request, if you've opened one already).
The text was updated successfully, but these errors were encountered: