hashicorp / terraform-provider-bitbucket

This Terraform provider is archived per our provider archiving process: https://terraform.io/docs/internals/archiving.html
https://registry.terraform.io
Mozilla Public License 2.0
31 stars 106 forks source link

Bitbucket API: "username" deprecated #36

Open kmoe opened 5 years ago

kmoe commented 5 years ago

As of 29th April 2019, the username field in the Bitbucket API is deprecated:

https://developer.atlassian.com/cloud/bitbucket/bitbucket-api-changes-gdpr/#removal-of-usernames-from-user-referencing-apis

The Bitbucket Terraform Provider currently uses usernames to specify the following resource attributes in Terraform configuration:

bitbucket_respository.owner

Used in requests to the /2.0/repositories/{username} endpoint.

As of 18th June 2019, this still works.

Recommendation

Deprecate bitbucket_repository.owner with a warning and add new attribute bitbucket_repository.owner_uuid. Either owner or owner_uuid must be specified.

bitbucket_default_reviewers.owner

Used in requests to the /2.0/repositories/{username}/{repo_slug}/default-reviewers endpoint .

As of 18th June 2019, this still works.

Recommendation

Deprecate bitbucket_default_reviewers.owner with a warning and add new attribute bitbucket_default_reviewers.owner_uuid. Either owner or owner_uuid must be specified.

bitbucket_default_reviewers.reviewers

Used in requests to the /2.0/repositories/{username}/{repo_slug}/default-reviewers/{target_username} endpoint, and responses from GET /2.0/repositories/{username}/{repo_slug}/default-reviewers.

Only UUIDs, and not usernames, are returned when requesting the default reviewers.

Recommendation

Specifying default reviewers via username is now broken. See #37 for details.

bitbucket_hook.owner

Used in requests to the /2.0/repositories/{username}/{repo_slug}/hooks endpoint.

As of 18th June 2019, this still works.

Recommendation

Deprecate bitbucket_hook.owner with a warning and add new attribute bitbucket_hook.owner_uuid. Either owner or owner_uuid must be specified.

kmoe commented 5 years ago

@cwood This is the result into my investigation into the new acceptance test failure in #35. Would you mind taking a look and letting me know if you agree with the analysis?