remove get_responses_requests and get_responses_request_bodies to use in-place matcher

[aibrahim-oai]

No description provided.

[aibrahim-oai]

@codex fix it

[chatgpt-codex-connector]

Summary

• Added a response mock helper and used it to collect request bodies in unified exec tests for response assertions. codex-rs/core/tests/suite/unified_exec.rsL158-L163 codex-rs/core/tests/suite/unified_exec.rsL1230-L1256
• Aligned wiremock matcher imports in review and compact/test helpers to match current usage. codex-rs/core/tests/suite/review.rsL34-L40 codex-rs/core/tests/common/test_codex.rsL25-L33 codex-rs/core/tests/suite/compact_resume_fork.rsL25-L40

Testing

• ✅ just fmt
• ✅ just fix -p codex-core
• ❌ cargo test -p codex-core (failed: default_client::tests::test_create_client_sets_default_headers; models_manager::manager::tests::refresh_available_models_*; exec::tests::kill_child_process_group_kills_grandchildren_on_timeout; shell_snapshot::tests::linux_sh_snapshot_includes_sections; shell_snapshot::tests::try_new_creates_and_deletes_snapshot_file; unified_exec::tests::multi_unified_exec_sessions; unified_exec::tests::reusing_completed_process_returns_unknown_process; interrupted after long-running unified_exec::tests::unified_exec_persists_across_requests/unified_exec_timeouts)
• ⚠️ cargo test --all-features (not run; blocked by failing codex-core tests)

View task →

[aibrahim-oai]

@codex review this

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 995ccfffa4

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".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Avoid flattening per-mock ResponseMock logs

In this test you flatten ResponseMock::requests() from four mount_sse_once_match mocks, but base_mock inserts the ResponseMock matcher before the custom matcher and wiremock evaluates matchers in order (MockBuilder::and appends; MountedMock::matches uses all). That means each mock records every POST to /responses even when its custom matcher fails, so with multiple mocks you’ll see duplicates (and ordering grouped by mock), making the request_bodies.len()/index assertions wrong or flaky. Consider logging via a single mock (e.g., mount_sse_sequence), or ensure the logging matcher runs last so only actual matches are captured.

Useful? React with 👍 / 👎.

[aibrahim-oai]

@codex fix this

[chatgpt-codex-connector]

Summary

• Switched the auto-compaction test to log requests via a single SSE sequence and simplified request body collection to avoid duplicate entries from per-mock logs. codex-rs/core/tests/suite/compact.rsL1012-L1082

Testing

• ✅ just fmt
• ✅ just fix -p codex-core
• ❌ cargo test -p codex-core (failed: default_client::tests::test_create_client_sets_default_headers; exec::tests::kill_child_process_group_kills_grandchildren_on_timeout; models_manager::manager::tests::{refresh_available_models_drops_removed_remote_models,refresh_available_models_refetches_when_cache_stale,refresh_available_models_sorts_and_marks_default,refresh_available_models_uses_cache_when_fresh}; shell_snapshot::tests::{linux_sh_snapshot_includes_sections,timed_out_snapshot_shell_is_terminated}; unified_exec::tests::{multi_unified_exec_sessions,reusing_completed_process_returns_unknown_process})

View task →