Script batch update caller workflows #6
Conversation
| @@ -0,0 +1,178 @@ | |||
| #!/usr/bin/env python | |||
| """ | |||
| Whenever a new reusable workflow is created or updated, we want to open a PR in every repository from which we want to call it. | |||
There was a problem hiding this comment.
I don't think it will always be the case that we want to strictly describe the workflows (using caller-templates) in which a reusable workflow is used. As the git ref of a reusable workflow can be specified e.g. as main, consuming workflows will get updates.
This change seems to be about updating the context in which shared workflows are used which might vary depending on the shared workflow, though in the case of the stale PR workflow this probably does make some sense.
| session = boto3.Session() | ||
| secrets_client = session.client("secretsmanager") | ||
| # using the scala_libs_pr_bumps secret for now but this should have its own secret | ||
| secret_value = secrets_client.get_secret_value(SecretId="builds/github_wecobot/scala_libs_pr_bumps") |
There was a problem hiding this comment.
We should avoid further use of GitHub API keys for weco-bot. Is the intention to automate running this script?
It's be useful to have some more context about what this is trying to achieve. |
What does this change?
Since we're planning to use more of these reusable github workflows, it'd be useful to be able to open PRs to update the "caller" repositories
How to test
Try it with a safe repo?
How can we measure success?
Have we considered potential risks?