fix(streaming): keep indexed snapshots compact

[jbeckwith-oai]

Summary

• treat streamed choice and tool-call indices as protocol identifiers while keeping snapshot arrays compact
• preserve deterministic index ordering and independent state for interleaved choices and tool calls
• reject negative indices before mutating stream state
• keep choice/tool-call lookup linear or constant-time and avoid repeated sorting for monotonic streamed appends

The behavior is implemented in the handwritten streaming accumulator; generated transport models remain unchanged.

Validation

• rebased onto origin/main at 2c4abe5e321053be32916d0df823a20f621a8d4c
• focused stream-index regressions on Ruby 3.3.12, 3.4.10, and 4.0.6: 8 runs / 40 assertions per version
• complete chat streaming and snapshot matrix: 42 runs / 200 assertions per version
• complete suite on Ruby 3.3.12, 3.4.10, and 4.0.6: 1,120 runs / 10,146 assertions per version
• bundle exec rake lint
• bundle exec rake build:gem
• independent two-pass review, thermo-nuclear maintainability review, and scoped Codex Security diff scan

[github-actions]

Castiron custom code

✅ No new custom-code files detected.

48 mixed files remain; 0 existing customizations changed.

Compared 851f53438f5a → 15c96822ca98. Generated baselines verified.

48 existing customizations unchanged

• lib/openai.rb
• lib/openai/client.rb
• lib/openai/models/chat/chat_completion_message.rb
• lib/openai/models/chat/chat_completion_message_function_tool_call.rb
• lib/openai/models/chat/completion_create_params.rb
• lib/openai/models/response_format_json_schema.rb
• lib/openai/models/responses/function_tool.rb
• lib/openai/models/responses/response.rb
• lib/openai/models/responses/response_create_params.rb
• lib/openai/models/responses/response_format_text_config.rb
• lib/openai/models/responses/response_format_text_json_schema_config.rb
• lib/openai/models/responses/response_function_tool_call.rb
• lib/openai/models/responses/response_function_web_search.rb
• lib/openai/models/responses/response_output_text.rb
• lib/openai/models/responses/tool.rb
• lib/openai/resources/beta/responses.rb
• lib/openai/resources/beta/threads.rb
• lib/openai/resources/chat/completions.rb
• lib/openai/resources/files.rb
• lib/openai/resources/responses.rb
• lib/openai/resources/vector_stores/file_batches.rb
• lib/openai/resources/vector_stores/files.rb
• lib/openai/resources/webhooks.rb
• rbi/openai/client.rbi
• rbi/openai/models/chat/chat_completion_message.rbi
• rbi/openai/models/chat/chat_completion_message_function_tool_call.rbi
• rbi/openai/models/chat/completion_create_params.rbi
• rbi/openai/models/response_format_json_schema.rbi
• rbi/openai/models/responses/response.rbi
• rbi/openai/models/responses/response_create_params.rbi
• rbi/openai/models/responses/response_function_tool_call.rbi
• rbi/openai/models/responses/response_function_web_search.rbi
• rbi/openai/models/responses/response_output_text.rbi
• rbi/openai/resources/chat/completions.rbi
• rbi/openai/resources/files.rbi
• rbi/openai/resources/responses.rbi
• rbi/openai/resources/vector_stores/file_batches.rbi
• rbi/openai/resources/vector_stores/files.rbi
• scripts/castiron/README.md
• scripts/castiron/custom_code_report.py

8 more in the full report.

A changed generated baseline means this report cannot reliably identify which handwritten lines changed.

Inspect the custom-code diff

Download the exact patch produced by this run (requires repository access):

gh run download 32446582790 --repo openai/openai-ruby \
  --name castiron-custom-code-32446582790-1 --dir /tmp/castiron-custom-code-32446582790-1
git apply --stat /tmp/castiron-custom-code-32446582790-1/custom-code.patch
cat /tmp/castiron-custom-code-32446582790-1/custom-code.patch

Or reproduce it from an SDK checkout containing the vendored reporter:

git fetch --no-tags origin 851f53438f5afc7f32e3ccbe866fd03bbade576a 15c96822ca9889eb5d921b692e591b8e0e855e88
python3 scripts/castiron/custom_code_report.py report \
  --base 851f53438f5afc7f32e3ccbe866fd03bbade576a \
  --head 15c96822ca9889eb5d921b692e591b8e0e855e88 --fetch --require-head-hash --public \
  --out /tmp/castiron-custom-code-15c96822ca98
cat /tmp/castiron-custom-code-15c96822ca98/custom-code.patch

This is the current full custom patch for mixed files, not an attribution of only the handwritten lines changed by this PR.

Full report and patch

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-08-21T04:21:53.232362Z	15c9682	New commits
🔒 Security Review	✅ Completed 2026-08-21T04:22:34.150111Z	15c9682	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[openai-sdks]

OkTest Summary

✅ 237/237 SDK tests passed in 9.913s for Ruby SDK PR #490.

Test results — 42 files

Test	Result	Time
tests/chat-completions-complex-body.test.ts	✅ Passed	180ms
tests/chat-completions-create.test.ts	✅ Passed	237ms
tests/chat-completions-stream.test.ts	✅ Passed	139ms
tests/files-content-binary.test.ts	✅ Passed	233ms
tests/files-create-multipart.test.ts	✅ Passed	142ms
tests/files-list-pagination.test.ts	✅ Passed	153ms
tests/initialize-config.test.ts	✅ Passed	183ms
tests/instance-isolation.test.ts	✅ Passed	233ms
tests/models-list.test.ts	✅ Passed	174ms
tests/responses-background-lifecycle.test.ts	✅ Passed	243ms
tests/responses-body-method-errors.test.ts	✅ Passed	430ms
tests/responses-cancel-timeout.test.ts	✅ Passed	222ms
tests/responses-cancel.test.ts	✅ Passed	264ms
tests/responses-compact-retries.test.ts	✅ Passed	229ms
tests/responses-compact.test.ts	✅ Passed	246ms
tests/responses-create-advanced-stream.test.ts	✅ Passed	142ms
tests/responses-create-advanced.test.ts	✅ Passed	184ms
tests/responses-create-disconnect.test.ts	✅ Passed	977ms
tests/responses-create-errors.test.ts	✅ Passed	257ms
tests/responses-create-malformed-api-responses.test.ts	✅ Passed	218ms
tests/responses-create-retries.test.ts	✅ Passed	312ms
tests/responses-create-stream-failures.test.ts	✅ Passed	162ms
tests/responses-create-stream-timeout.test.ts	✅ Passed	211ms
tests/responses-create-stream-wire.test.ts	✅ Passed	2.469s
tests/responses-create-stream.test.ts	✅ Passed	89ms
tests/responses-create-terminal-states.test.ts	✅ Passed	289ms
tests/responses-create-timeout.test.ts	✅ Passed	204ms
tests/responses-create.test.ts	✅ Passed	245ms
tests/responses-delete.test.ts	✅ Passed	253ms
tests/responses-input-items-errors.test.ts	✅ Passed	200ms
tests/responses-input-items-list.test.ts	✅ Passed	195ms
tests/responses-input-items-options.test.ts	✅ Passed	255ms
tests/responses-input-tokens-count-timeout.test.ts	✅ Passed	240ms
tests/responses-input-tokens-count.test.ts	✅ Passed	269ms
tests/responses-malformed-inputs.test.ts	✅ Passed	2.088s
tests/responses-not-found-errors.test.ts	✅ Passed	354ms
tests/responses-parse.test.ts	✅ Passed	197ms
tests/responses-retrieve-retries.test.ts	✅ Passed	287ms
tests/responses-retrieve.test.ts	✅ Passed	258ms
tests/responses-stored-method-errors.test.ts	✅ Passed	738ms
tests/retry-behavior.test.ts	✅ Passed	3.116s
tests/sdk-error-shape.test.ts	✅ Passed	355ms

View OkTest run #32446583564

_{SDK merge (2acfbada75c6) · head (15c96822ca98) · base (851f53438f5a) · OkTest (2b1bdfd25e98)}

[HAYDEN-OAI]

Found one introduced P2 reliability/scalability issue in the indexed streaming accumulator.

[HAYDEN-OAI]

One substantive P2 performance regression is documented inline.

[jbeckwith-oai]

Rebased onto current origin/main and completed a comprehensive re-review. The independent pass found one sibling performance gap: monotonically appended tool calls still triggered a full sort_by! after every chunk. The deterministic regression failed first with 127 sorts for 127 appends and now passes with zero; out-of-order/interleaved ordering remains covered. Exact head 32bf0169bb08e89687a41f241dcf01859f8e30ad is fully green. Local verification: chat streaming matrix 42 runs / 200 assertions on Ruby 3.3.12, 3.4.10, and 4.0.6; full suite 1,120 runs / 10,146 assertions on each; lint, types, rubyfmt, gem build, two-pass review, thermo review, and scoped security review all passed.