wboayue / rust-ibapi / 26457394198

92%

Build Type

push

github

Committed by web-flow

Commit Message

display_groups: decode_display_group_updated proto-only at floor 213 (PR-C5) (#638)

* display_groups: decode_display_group_updated proto-only at floor 213 (PR-C5)

Gate is PROTOBUF_REST_MESSAGES_3 (213), the connection floor since #632.
Collapse the PR-A dual-format wrapper: receiver flipped &mut ResponseMessage
→ &ResponseMessage via require_proto(); drop log::warn import + decode_proto_or_text
plumbing; decode_display_group_updated_proto is now the single decode path.

Add DisplayGroupUpdatedResponse builder in new src/testdata/builders/display_groups.rs.
Extract inline tests to sibling *_tests.rs files (rule 8/9): decoders_tests.rs,
stream_decoders_tests.rs, sync_tests.rs, async_tests.rs. Migrate every text-frame
fixture to proto_response(IncomingMessages::DisplayGroupUpdated, ...encode_proto()).
Add test_decode_display_group_updated_rejects_text_framing regression test.

Mark ResponseMessage::len #[allow(dead_code)] (last production caller was the
text branch deleted above. Matches PR-C2/C4 treatment of next_long / next_bool.

Update plans/legacy-text-protocol-cleanup.md to record the cleanup and drop
the now-stale display-groups callsite refs in the helper-deletion checklist.
EOF
)

* display_groups: /simplify follow-ups for PR-C5

- Drop the message_type() precheck in decode_display_group_updated (the
  StreamDecoder dispatcher already filters by message_type, so the branch
  was unreachable; aligns with C2/C3/C4 precedents).
- Drop the now-redundant _wrong_message_type test (the unreachable branch)
  and the wrapper happy-path tests (_proto + _rejects_text_framing cover
  the full surface).
- DisplayGroupUpdatedResponse.request_id: Option<i32> -> i32 with
  TEST_REQ_ID_FIRST default. Matches sibling builders; Option<T> in builders
  is reserved for genuinely-optional wire fields (cf. WshEventDataRequestBuilder).
- Drop narration comments in sync_tests/async_tests/stream_decoders_tests
  and refine the dead_code rationale on ResponseMessage::le... (continued)

Coverage Stats

21 of 25 new or added lines in 2 files covered. (84.0%)

4 existing lines in 1 file now uncovered.

19400 of 21030 relevant lines covered (92.25%)

27.31 hits per line