fix(deps): bump cmf-sdk-go v0.0.5 to v0.0.7 and handle breaking changes by paras-negi-flink · Pull Request #3333 · confluentinc/cli

Paras Negi (paras-negi-flink) · 2026-04-24T19:07:22Z

Release Notes

Breaking Changes

None. (CLI surface unchanged. The SDK's own Go API has breaking changes — internal callers updated in this PR.)

New Features

None.

Bug Fixes

Restores confluent flink compute-pool list/describe/create on Confluent Platform, which had been broken by an upstream cmf-sdk-go v0.0.6 schema regression that caused ComputePool.status to fail to decode.

Checklist

What

This PR bumps github.com/confluentinc/cmf-sdk-go from v0.0.5 → v0.0.7 and adapts the CLI to two breaking type changes the new SDK introduces. Targets Confluent Platform / CP Flink (CMF on-prem).

ComputePool.Status: *ComputePoolStatus → *map[string]interface{}
- The typed status struct was replaced by an untyped map. Phase is no longer a struct field — it's an entry in the map.
- Added extractComputePoolPhase(pool) helper in internal/flink/command_compute_pool.go, used by the 4 consumer sites (convertSdkComputePoolToLocalComputePool, _create_onprem, _describe_onprem, _list_onprem).
- A non-string phase value (server/schema contract violation) is logged at debug level via log.CliLogger.Debugf — matches the existing pkg/featureflags/feature_flags.go pattern.
PostEnvironment.Name: string → *string
- Writes use &name. Reads use the SDK-generated nil-safe GetName() accessor (pkg/flink/cmf_rest_client.go, test/test-server/flink_onprem_handler.go).
Drive-bys:
- Fixed a pre-existing godoc typo on UpdateEnvironment ("Create an environment" → "updates an existing environment").
- Fixed pre-existing gci import-ordering violation in pkg/schemaregistry/schema_test.go (from fix: use --normalize flag value when registering schemas #3332) — was blocking lint on this branch.

Blast Radius

Scope is limited to on-prem Flink commands that consume ComputePool.Status or PostEnvironment.Name: confluent flink compute-pool {list,describe,create} and confluent flink environment {create,update}.
If something goes wrong:
- Compute-pool commands could show empty/wrong Phase (helper returns "" for any non-string value; debug log fires for unexpected types).
- Environment create/update could submit an empty/nil name if the helper logic regresses; current code reads from a required CLI positional, so the field is always populated in practice.
No changes to Confluent Cloud commands, no changes to other Flink commands (statements, applications, savepoints, catalogs).
Reverting is a single-commit revert: switching back will re-introduce the v0.0.6 decode bug, so the recommended rollback is to pin the SDK to v0.0.5 and reverse the type adaptations.

References

Jira: CF-3528 — CLI: bump cmf-sdk-go to v0.0.7
SDK PR (released as v0.0.7): confluentinc/cmf-sdk-go#66
Upstream OpenAPI fix: confluentinc/cp-flink-cmf#1306

Test & Review

Environment

Repo: confluentinc/cli
Branch: bump-cmf-sdk-v0.0.7
SDK: github.com/confluentinc/cmf-sdk-go v0.0.7
CMF: tested locally against cmf-service (kubectl port-forward svc/cmf-service 9090:80)

Automated

go build ./... clean
go vet ./... clean
go test ./internal/flink/... passes
go test ./pkg/flink/... — only pre-existing flake TestInteractiveOutputControllerTestSuite/TestCloseTableViewOnUserInput (unrelated to SDK; UI-input test)

Manual CLI validation

Attached in the comment below.

confluent-cla-assistant · 2026-04-24T19:07:36Z

🎉 All Contributor License Agreements have been signed. Ready to merge.
_{Please push an empty commit if you would like to re-run the checks to verify CLA status for all contributors.}

Copilot

Pull request overview

This PR updates the CLI’s CMF integration to use github.com/confluentinc/cmf-sdk-go v0.0.7 and adapts the on-prem Flink commands/tests to two upstream breaking model changes (ComputePool status becoming an untyped map; PostEnvironment name becoming a pointer).

Changes:

Bump cmf-sdk-go from v0.0.5 → v0.0.7 (and update go.mod/go.sum accordingly).
Add extractComputePoolPhase helper to read status["phase"] from the new untyped ComputePool.Status and use it in on-prem compute-pool list/describe/create + local conversion.
Update environment create/update paths (client + test server + commands) to use PostEnvironment.Name as *string via &environmentName writes and GetName() reads; add unit tests for phase extraction.

Reviewed changes

Copilot reviewed 10 out of 11 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
test/test-server/flink_onprem_handler.go	Adjust test-server models to match SDK pointer/name and status-map changes.
pkg/flink/cmf_rest_client.go	Switch PostEnvironment name reads to `GetName()`; fix UpdateEnvironment godoc.
internal/flink/command_environment_create.go	Set `PostEnvironment.Name` as `*string`.
internal/flink/command_environment_update.go	Set `PostEnvironment.Name` as `*string`.
internal/flink/command_compute_pool.go	Add `extractComputePoolPhase` and update local conversion to use it.
internal/flink/command_compute_pool_list_onprem.go	Use `extractComputePoolPhase` instead of typed status access.
internal/flink/command_compute_pool_describe_onprem.go	Use `extractComputePoolPhase` instead of typed status access.
internal/flink/command_compute_pool_create_onprem.go	Use `extractComputePoolPhase` for Phase output.
internal/flink/command_compute_pool_test.go	Add table-driven tests for `extractComputePoolPhase`.
go.mod / go.sum	Dependency bump for `cmf-sdk-go`.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-24T19:11:16Z

 			Name:         sdkComputePool.GetMetadata().Name,
 			Type:         sdkComputePool.GetSpec().Type,


In the human output table, Name and Type are taken from the input (sdkComputePool) while CreationTime/Phase come from the server response (sdkOutputComputePool). This can print incorrect values if the server defaulted/normalized fields; it’s also inconsistent with describe/list which print response data. Prefer using sdkOutputComputePool for Name and Type as well.

Suggested change

Name: sdkComputePool.GetMetadata().Name,

Type: sdkComputePool.GetSpec().Type,

Name: sdkOutputComputePool.GetMetadata().Name,

Type: sdkOutputComputePool.GetSpec().Type,

This is a valid concern.

Channing Dong (channingdong) · 2026-04-30T02:57:04Z

 			Name:         sdkComputePool.GetMetadata().Name,
 			Type:         sdkComputePool.GetSpec().Type,


This is a valid concern.

Channing Dong (channingdong) · 2026-04-30T02:59:17Z

 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
+
+	srsdk "github.com/confluentinc/schema-registry-sdk-go"


It looks like we have merging conflict for this file, can we fix that?

v0.0.7 introduces two breaking changes that needed CLI adaptation: 1. ComputePool.Status is now *map[string]interface{} (flattened from the earlier v0.0.6 nested map shape). Phase is no longer a typed struct field - callers read it from the untyped status map. Added extractComputePoolPhase helper used by the 4 compute-pool command sites; a non-string "phase" value (server/schema contract violation) is logged at debug level. 2. PostEnvironment.Name is now *string. Writes use &name; reads use the SDK-generated nil-safe GetName() accessor. Also fixes a pre-existing godoc typo on UpdateEnvironment.

…d phase-extraction test Addresses review feedback on #3333: - Read Name/Type from sdkOutputComputePool so create matches describe/list and the human/serialized outputs agree on what was actually persisted. - Add table-driven unit test for extractComputePoolPhase covering nil status, missing key, nil/string/empty/non-string phase values. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Channing Dong (channingdong) · 2026-04-30T05:20:15Z

+// A missing or nil value is treated as "phase not yet populated" and returns "". A
+// present value that isn't a string indicates a server/schema contract violation and
+// is logged at debug level.
+func extractComputePoolPhase(pool cmfsdk.ComputePool) string {


Are we expecting similar *map[string]interface{} as a general data struct for future CMF development?

We have an ongoing initiative to automate the CP CMF CLI source code generation, and your team doesn't need to spend time on the standard CRUDL feature development anymore, so we'd like to have the custom functions as few as possible.

Can we explore ways to make this function more generic so that we can try to generate it whenever we have a need to extract certain key from a map?

Agreed, it's realistically the only practical way to surface controller-driven values, and patterns like these will pop up from time to time.

Generalized to pkg/flink.GetMapField so auto-gen has a single helper to target.

….GetMapField cmf-sdk-go v0.0.7 represents 9+ K8s/FKO-style fields as map[string]any (Environment.Status, ApplicationInstanceStatus.Spec, FlinkApplicationAllOf .{Metadata,Spec,Status}, ComputePoolDefaults, etc.). The pattern will recur as more CMF resources surface controller-driven status. Lift the compute-pool-specific extraction into a generic, type-parameterized helper available CMF-wide. Replaces the local extractComputePoolPhase logic with a one-line wrapper. Tests moved alongside the helper and expanded to cover string, int64, nested map, and slice value types. The auto-gen tooling for the CP CMF CLI can target this signature directly when emitting accessors for untyped fields. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sonarqube-confluent · 2026-04-30T12:25:59Z

Quality Gate passed

Issues
5 New issues
1 Fixed issue
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube

Copilot AI review requested due to automatic review settings April 24, 2026 19:07

Paras Negi (paras-negi-flink) requested a review from a team as a code owner April 24, 2026 19:07

Copilot started reviewing on behalf of Paras Negi (paras-negi-flink) April 24, 2026 19:07 View session

Copilot AI reviewed Apr 24, 2026

View reviewed changes

airlock-confluentinc Bot force-pushed the bump-cmf-sdk-v0.0.7 branch from 33f07a3 to ba18b99 Compare April 24, 2026 19:11

Channing Dong (channingdong) reviewed Apr 30, 2026

View reviewed changes

Paras Negi (paras-negi-flink) and others added 2 commits April 30, 2026 09:04

airlock-confluentinc Bot force-pushed the bump-cmf-sdk-v0.0.7 branch from 9e95400 to 0b22a62 Compare April 30, 2026 03:59

Channing Dong (channingdong) reviewed Apr 30, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(deps): bump cmf-sdk-go v0.0.5 to v0.0.7 and handle breaking changes#3333

fix(deps): bump cmf-sdk-go v0.0.5 to v0.0.7 and handle breaking changes#3333
Paras Negi (paras-negi-flink) wants to merge 3 commits intomainfrom
bump-cmf-sdk-v0.0.7

Paras Negi (paras-negi-flink) commented Apr 24, 2026 •

edited

Loading

Uh oh!

confluent-cla-assistant Bot commented Apr 24, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 24, 2026

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Uh oh!

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Uh oh!

Paras Negi (paras-negi-flink) Apr 30, 2026

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Uh oh!

Paras Negi (paras-negi-flink) Apr 30, 2026 •

edited

Loading

Uh oh!

sonarqube-confluent Bot commented Apr 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		Name: sdkComputePool.GetMetadata().Name,
		Type: sdkComputePool.GetSpec().Type,

Conversation

Paras Negi (paras-negi-flink) commented Apr 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Release Notes

Checklist

What

Blast Radius

References

Test & Review

Uh oh!

confluent-cla-assistant Bot commented Apr 24, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Paras Negi (paras-negi-flink) Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Channing Dong (channingdong) Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Paras Negi (paras-negi-flink) Apr 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

sonarqube-confluent Bot commented Apr 30, 2026

Quality Gate passed

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Paras Negi (paras-negi-flink) commented Apr 24, 2026 •

edited

Loading

Paras Negi (paras-negi-flink) Apr 30, 2026 •

edited

Loading