[yashdotdev13]

Summary

Improve handling of whitespace in comma-delimited application names in CredhubEnvironmentRepository.

Changes

• Trim leading and trailing whitespace from application names before deduplication.
• Add a regression test for application names containing leading and trailing whitespace.
• Make test helper path normalization platform-independent.

Testing

• CredhubEnvironmentRepositoryTests — 13 tests passed.
• Full spring-cloud-config-server module tests passed successfully.

[ryanjbaxter]

I dont see how this fixes #1873. That issue is about comma separated list of application names, not white spaces

[yashdotdev13]

Thanks for pointing this out. After revisiting the original issue, I realize that my PR only addresses whitespace handling and does not solve the core requirement of supporting multiple comma-separated application names.

I'll revisit the implementation to handle each application name individually and update the PR accordingly. Sorry for the misunderstanding.

[ryanjbaxter]

Then the PR description should be updated since it does not fix the issue

[yashdotdev13]

You're right. I'll keep this PR focused on whitespace normalization in CredHub application names rather than addressing #1873. I'll update the PR title and description accordingly. Thanks for the clarification!

[yashdotdev13]

hey @ryanjbaxter , ha've updated the PR title and description and removed the reference to #1873. This PR now focuses solely on whitespace normalization in CredHub application names. Thanks for your feedback and guidance!

[yashdotdev13]

Addressed all three review comments and pushed the changes. Tests are passing. Could you please take another look? Thanks!

[yashdotdev13]

Updated to reuse the normalized values and avoid duplicating the logic. Thanks!

[ryanjbaxter]

Use normalize(String commaDelimitedItems, String defaultItem)

[yashdotdev13]

Just to clarify, should the Environment metadata include the default application/profile values from the normalized lists, or should it preserve the original metadata behavior while reusing the normalized values for property source lookup?