refactor: clean up SSE transport variant

[DaleSeo]

The SSE transport support was removed in v1.5.0, but the enum variant, error path, match arms, and dedicated test were still hanging around and complicating the transport logic. This PR removes Transport::SSE entirely and cleans up the related code.

[apollo-librarian]

✅ Docs preview ready

The preview is ready to be viewed. View the preview

File Changes

0 new, 1 changed, 0 removed

* (developer-tools)/apollo-mcp-server/(latest)/config-file.mdx

Build ID: 2b77fd6116071d4543676779
Build Logs: View logs

URL: https://www.apollographql.com/docs/deploy-preview/2b77fd6116071d4543676779

[github-actions]

⏭️ Changeset check skipped via label

[claude]

Review Summary

Clean refactoring that fully removes the deprecated SSE transport variant and associated infrastructure. The changes are thorough and include proper test coverage to prevent SSE from being used.

Findings

[Consider] The boxing of auth::Config in Transport::StreamableHttp (changing from Optionauth::Config to Option<Boxauth::Config>) appears to be an optimization to reduce the enum's memory footprint. While this is generally good practice per Chapter 2.3 (clippy's large_enum_variant lint), it's worth confirming this was intentional and not an accidental change, as it wasn't mentioned in the PR description.

Test Coverage Assessment

Good test coverage:

• New test sse_transport_is_rejected_at_parse_time validates SSE is rejected during YAML parsing
• Tests for stdio and streamable_http transports confirm they parse correctly
• Removed test start_sse_server_returns_unsupported_error is no longer needed since SSE fails at parse time

Following Chapter 5 best practices, the test names are descriptive and focused on single behaviors.

Final Recommendation

Approve with suggestions - The code is solid and ready to merge. The auth boxing change is worth confirming as intentional.

---

Reviewed by Claude Code Sonnet 4.5

[andrewmcgivery]

Just want to double check the intent behind this Box. :)

[DaleSeo]

Without it, you will run into this linting issue:

error: large size difference between variants
  --> crates/apollo-mcp-server/src/server.rs:59:1
   |
59 | /  pub enum Transport {
60 | |      /// Use standard IO for server <> client communication
61 | |      #[default]
62 | |      Stdio,
   | |      ----- the second-largest variant carries no data at all
...  |
65 | |/     StreamableHttp {
66 | ||         /// Authentication configuration
67 | ||         #[serde(default)]
68 | ||         auth: Option<auth::Config>,
...  ||
84 | ||         host_validation: HostValidationConfig,
85 | ||     },
   | ||_____- the largest variant contains at least 356 bytes
86 | |  }
   | |__^ the entire enum is at least 360 bytes
   |
   = help: for further information visit https://rust-lang.github.io/rust-clippy/rust-1.92.0/index.html#large_enum_variant
   = note: `-D clippy::large-enum-variant` implied by `-D warnings`
   = help: to override `-D warnings` add `#[allow(clippy::large_enum_variant)]`
help: consider boxing the large fields or introducing indirection in some other way to reduce the total size of the enum
   |
68 -         auth: Option<auth::Config>,
68 +         auth: Box<Option<auth::Config>>,
   |

error: could not compile `apollo-mcp-server` (lib) due to 1 previous error
warning: build failed, waiting for other jobs to finish...
error: could not compile `apollo-mcp-server` (lib test) due to 1 previous error

[claude]

Review Summary

This is a clean refactoring that successfully removes all SSE transport remnants from the codebase. The approach of rejecting SSE at parse time (via serde deserialization) rather than at runtime is a better design.

Findings

No new issues found. The previous review correctly identified the auth boxing change, which follows clippy's large_enum_variant lint guidance (Chapter 2.3).

Test Coverage Assessment

Excellent test coverage:

• Parse-time validation for all transport types with descriptive test names following Chapter 5.1 best practices
• Proper removal of the obsolete start_sse_server_returns_unsupported_error test

Final Recommendation

Approve - The refactoring is well-executed and improves code clarity by removing dead code paths.

---

Reviewed by Claude Code Sonnet 4.5

[mabuyo]

Minor coment on docs addition

[mabuyo]

Might be helpful to add a version number here if there's a large chance folks are still using older versions! "no longer supported by Apollo MCP Server as of vx.x.x"

[claude]

Review Summary

This refactoring successfully removes all SSE transport code following its deprecation in v1.5.0. The latest commit adds appropriate version context to the deprecation notice in documentation.

Findings

No new issues found beyond what was previously identified in earlier reviews regarding the auth Config boxing optimization.

Test Coverage Assessment

Test coverage is excellent - the new tests ensure SSE is rejected at parse time, and the removal of the obsolete start_sse_server_returns_unsupported_error test is appropriate since SSE now fails during deserialization rather than at runtime.

Final Recommendation

Approve - The refactoring is well-executed, removes dead code paths cleanly, and improves maintainability. The auth boxing change follows Rust best practices per Chapter 2.3 (large_enum_variant lint).

---

Reviewed by Claude Code Sonnet 4.5