add kubernetes namespace labels

[nbucci]

PR Type

Enhancement, Documentation

---

Description

• Added support for Kubernetes namespace labels in Drupal projects configuration
• Removed ignore_changes block from kubernetes_namespace resource, allowing external label modifications to cause drift
• Updated drupal_projects_list variable to include optional kubernetes_namespace_labels field
• Removed helm values outputs
• Updated CHANGELOG and README to reflect these changes
• Bumped version to 0.13.0 with breaking changes noted

---

Changes walkthrough 📝

	Relevant files
Enhancement	main.tf Add Kubernetes namespace labels support                                    main.tf Added labels field to the namespace_list local variable Updated kubernetes_namespace resource to include labels Removed ignore_changes lifecycle block for namespace labels +2/-6      variables.tf Update Drupal projects variable with namespace labels        variables.tf Added kubernetes_namespace_labels optional field to the drupal_projects_list variable +1/-0
Documentation	CHANGELOG.md Update CHANGELOG with new version and breaking changes      CHANGELOG.md Added entry for version 0.13.0 Documented new support for Kubernetes namespace labels Noted breaking changes regarding removal of ignore_changes and helm values outputs +14/-0    README.md Update README with new variable and removed outputs            README.md Removed template provider from requirements Updated drupal_projects_list input to include kubernetes_namespace_labels Removed outputs for helm values +1/-7

---

💡 PR-Agent usage:
Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

[sparkfabrik-ai-bot]

PR Reviewer Guide 🔍

(Review updated until commit 2ed0951)

⏱️ Estimated effort to review: 2 🔵🔵⚪⚪⚪

🧪 No relevant tests

🔒 No security concerns identified

⚡ Key issues to review Potential Drift Removal of ignore_changes block for kubernetes_namespace resource may cause unintended drift if labels are modified externally. Default Value The new kubernetes_namespace_labels variable is set with an empty map as default. Consider if this is the intended behavior or if a set of default labels should be provided.

[sparkfabrik-ai-bot]

PR Code Suggestions ✨

Latest suggestions up to 2ed0951

Category	Suggestion	Score
Best practice	Add validation for Kubernetes label format compliance Consider adding a validation block for the kubernetes_namespace_labels variable to ensure that label keys and values conform to Kubernetes naming conventions. variables.tf [48] kubernetes_namespace_labels = optional(map(string), {}) +validation { + condition = can([for k, v in var.kubernetes_namespace_labels : regex("^([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9]$", k) && regex("^(([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9])?$", v)]) + error_message = "Kubernetes labels must consist of alphanumeric characters, '-', '_' or '.', and must start and end with an alphanumeric character." +} Suggestion importance[1-10]: 9 Why: This suggestion adds important input validation, preventing potential issues with non-compliant Kubernetes labels.	9
	Add a conditional check to avoid setting empty label sets Consider adding a conditional check to only set labels if they are not empty. This can prevent creating unnecessary empty label sets in Kubernetes. main.tf [89] -labels = each.value.labels +labels = length(each.value.labels) > 0 ? each.value.labels : null Suggestion importance[1-10]: 7 Why: This suggestion improves resource efficiency and adheres to best practices, but it's not addressing a critical issue.	7
Possible issue	Provide a default value for labels to prevent potential null errors Consider adding a default empty map for labels in case p.kubernetes_namespace_labels is null. This can prevent potential errors if the labels are not provided. main.tf [50] -labels = p.kubernetes_namespace_labels +labels = coalesce(p.kubernetes_namespace_labels, {}) Suggestion importance[1-10]: 8 Why: This suggestion addresses a potential runtime error, which is important for code stability and robustness.	8
Enhancement	Simplify the namespace assignment using the coalesce function Consider using the coalesce function instead of the ternary operator for setting the namespace value. This can make the code more concise and easier to read. main.tf [49] -namespace = p.kubernetes_namespace == null ? "${p.project_name}-${p.gitlab_project_id}-${p.release_branch_name}" : p.kubernetes_namespace +namespace = coalesce(p.kubernetes_namespace, "${p.project_name}-${p.gitlab_project_id}-${p.release_branch_name}") Suggestion importance[1-10]: 6 Why: The suggestion improves code readability and conciseness, but it's not addressing a critical issue or major bug.	6

---

Previous suggestions

Suggestions up to commit 1854687

Category	Suggestion	Score
Best practice	Add validation for Kubernetes label format in the variable definition Consider adding a validation block to ensure that the kubernetes_namespace_labels map only contains valid Kubernetes label key-value pairs. Kubernetes has specific requirements for label keys and values. variables.tf [48] kubernetes_namespace_labels = optional(map(string), {}) +validation { + condition = can([for k, v in var.kubernetes_namespace_labels : regex("^([a-z0-9A-Z][-a-z0-9A-Z_.]*)?[a-z0-9A-Z]$", k) && length(k) <= 63 && regex("^(([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9])?$", v) && length(v) <= 63]) + error_message = "Kubernetes labels must be valid: keys and values must be 63 characters or less, begin and end with an alphanumeric character, and contain only alphanumerics, '-', '_' or '.'." +} Suggestion importance[1-10]: 9 Why: Adding validation for Kubernetes label format is crucial for preventing errors and ensuring compliance with Kubernetes naming conventions, significantly improving code robustness.	9
	Use a dynamic block to conditionally include labels in Kubernetes metadata Consider adding a conditional expression to only include the labels block if there are actually labels to apply. This can help avoid creating empty label blocks in Kubernetes. main.tf [87-89] metadata { name = each.value.namespace - labels = each.value.labels + dynamic "labels" { + for_each = length(each.value.labels) > 0 ? [1] : [] + content { + labels = each.value.labels + } + } } Suggestion importance[1-10]: 8 Why: This suggestion enhances resource efficiency by avoiding empty label blocks, which is a valuable optimization for Kubernetes resources.	8
Enhancement	Use the merge() function for combining default and provided labels Consider using the merge() function instead of the ternary operator for combining the default empty map with the provided labels. This approach is more idiomatic in Terraform and allows for easier extension in the future. main.tf [50] -labels = p.kubernetes_namespace_labels == null ? {} : p.kubernetes_namespace_labels +labels = merge({}, p.kubernetes_namespace_labels) Suggestion importance[1-10]: 7 Why: The suggestion improves code readability and maintainability by using a more idiomatic Terraform approach, though the functional difference is minimal.	7

[Monska85]

Suggested change

	## [0.13.0] - 2024-11-04
	## [0.13.0] - 2024-11-04
	[Compare with previous version](https://github.com/sparkfabrik/terraform-google-gcp-cloud-native-drupal-resources/compare/0.12.1...0.13.0)

[Monska85]

LGTM 👍

[sparkfabrik-ai-bot]

Persistent review updated to latest commit 2ed0951