Extend tracing-test coverage for footnote branches and public API (#306)

[lodyai]

Summary

Closes #306.

Post-merge review of PR #303 (which added tracing instrumentation to the
inline-wrapping token- and fragment-classification pipeline) identified two
testing gaps that were out of scope for that PR. This change closes both.

Gap 1 — find_footnote_end branch coverage

find_footnote_end emits three distinct trace! events; the existing
tracing_tests module in src/wrap/tokenize/parsing.rs only covered the
prefix_mismatch path directly, and the unterminated_bracket path was
already covered. This adds a direct #[traced_test] for the remaining
footnote_label_span_recognized branch, calling find_footnote_end with a
well-formed [^4] input and asserting the emitted event message plus the
structured start=/end=/token_length=4 fields, while upholding the
content-free logging convention (!logs_contain("[^4]")).

The assertion uses the exact emitted spelling recognized (Oxford -ize),
matching production code, rather than the -ised variant.

Gap 2 — end-to-end tracing through the public API

All prior tracing tests exercised internal helpers directly. This adds the
first integration-level tracing test, in tests/wrap_unit/footnotes.rs, that
drives the public wrap_text API with "Some text.[^1]" and asserts the
fragment classified / kind=FootnoteRef DEBUG event fires. This confirms
the instrumentation is reachable through the public API and not accidentally
gated away.

Note on tracing-test configuration

Enabling tracing-test's no-env-filter feature was required for Gap 2: by
default #[traced_test] installs a per-crate env filter that only records the
test crate's own events, which would silently hide events emitted from the
mdtablefix library crate in an integration test. This was not anticipated in
the coding plan but is necessary for the end-to-end test to observe the
library's instrumentation.

Validation

• make check-fmt, make lint, make typecheck, make test — all green
  (lib 707/707, wrap_unit 84/84 including both new tests).
• coderabbit review --agent — 0 findings.

References

• Issue Extend tracing tests: cover all find_footnote_end branches and add end-to-end traced wrapping tests #306
• PR Add observability for inline wrapping token classification (#296) #303 — Add observability for inline wrapping token classification
• Issue Add observability infrastructure for inline-wrapping token classification #296 — Add observability infrastructure for inline-wrapping token classification
• Lody session: https://lody.ai/leynos/sessions/3bd173c3-eebb-44ec-9902-26f5e84b42d1

🤖 Generated with Claude Code

[sourcery-ai]

Sorry @LodyAI[bot], you have reached your weekly rate limit of 500000 diff characters.

Please try again later or upgrade to continue using Sourcery

[coderabbitai]

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

• @coderabbitai resume to resume automatic reviews.
• @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

• ▶️ Resume reviews
• 🔍 Trigger review

Walkthrough

The change enables crate-wide tracing capture in tests and adds assertions for footnote label recognition and FootnoteRef classification through both internal parsing and the public wrap_text API.

Changes

Tracing coverage

Layer / File(s)	Summary
Tracing test capture setup Cargo.toml	Enable the no-env-filter feature for tracing-test and document its purpose.
Footnote tracing assertions src/wrap/tokenize/parsing_tests.rs, tests/wrap_unit/footnotes.rs	Assert structured footnote label-span events and FootnoteRef classification events from find_footnote_end and wrap_text.

Possibly related issues

• #305 — Both changes add tracing assertions for footnote and fragment parsing paths.

Possibly related PRs

• leynos/mdtablefix#303 — Introduced the tracing instrumentation validated by these tests.
• leynos/mdtablefix#395 — Uses structured footnote tracing fields and content-free event assertions.
• leynos/mdtablefix#411 — Adds related footnote tracing snapshot infrastructure and tests.

Suggested labels: Issue

Suggested reviewers: leynos

Poem

Footnotes whisper, traces glow,
Through parser paths the signals flow.
[^4] leaves fields in line,
FootnoteRef proves the sign.
Tests catch each event in time.

---

Caution

Pre-merge checks failed

Please resolve all errors before merging. Addressing warnings is optional.

• Ignore

❌ Failed checks (1 error, 2 warnings)

Check name	Status	Explanation	Resolution
Testing (Overall)	❌ Error	find_footnote_end’s new test still only checks start=/end= presence, so a wrong span would pass.	Assert the exact span values, e.g. start=0 and end=4, and keep the return value/oracle tied to the parsed footnote.
Developer Documentation	⚠️ Warning	Cargo.toml adds tracing-test's no-env-filter feature, but no developer-guide/ADR update records this new test/build requirement; the guide only names tracing-test generically.	Add a note in docs/developers-guide.md (or an ADR) explaining the no-env-filter requirement for integration tests, and tick off any linked roadmap item if one exists.
Testing (Unit And Behavioural)	⚠️ Warning	The new direct find_footnote_end test only checks start=/end= exist, so it misses the span invariant and can pass on wrong offsets.	Assert the exact fields (start=0, end=4) and keep the !logs_contain("[^4]") check so the test fails on incorrect span metadata.

✅ Passed checks (17 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title matches the PR scope and includes the linked issue reference (#306).
Description check	✅ Passed	The description stays on-topic and describes the tracing test coverage changes clearly.
Linked Issues check	✅ Passed	The PR adds the requested branch tests and public API tracing test for #306.
Out of Scope Changes check	✅ Passed	The Cargo.toml feature change supports the new integration test and adds no unrelated scope.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
User-Facing Documentation	✅ Passed	No user-facing behaviour changed; the PR only adds tracing-test coverage and a dev-dependency feature, so docs/users-guide.md needs no update.
Module-Level Documentation	✅ Passed	PASS: Both changed Rust modules have clear file-level docstrings, and the nested tracing_tests module is documented too.
Testing (Property / Proof)	✅ Passed	The patch only adds branch-specific tracing tests and a tracing-test config; it introduces no new range-wide invariant or proof obligation.
Testing (Compile-Time / Ui)	✅ Passed	PASS: No compile-time/UI behaviour is introduced; the change adds runtime tracing assertions, and the repo already uses focused snapshots for wrap output where needed.
Unit Architecture	✅ Passed	Only dev-dependency and tests changed; production paths stay pure, with tracing exercised explicitly at the test boundary.
Domain Architecture	✅ Passed	Only dev-dependencies and test files changed; no domain, transport, persistence, or adapter code was altered.
Observability	✅ Passed	PASS: The PR only adds tracing tests and a dev-only tracing-test flag; no production observability, metrics, or alerting behaviour changes.
Security And Privacy	✅ Passed	Only tracing tests and a dev-dependency feature changed; no secrets, auth, permissions, or unsafe input handling were introduced, and test inputs are fake.
Performance And Resource Use	✅ Passed	PASS: keep changes test-only; no production hot paths or unbounded work were added, and the new tracing-test feature is dev-only.
Concurrency And State	✅ Passed	PASS: The diff is test-only plus a dev-dependency flag; it adds no shared mutable state, locks, async tasks, or ordering-sensitive runtime behaviour.
Architectural Complexity And Maintainability	✅ Passed	Only test coverage and a dev-dependency feature flag were added; no new runtime abstractions, layers, registries, or dependency cycles appear in the diff.
Rust Compiler Lint Integrity	✅ Passed	No new #[allow]/#[expect] suppressions or fake usage anchors were added, and the touched tests add no suspicious clones or unused imports.

📋 Issue Planner

Built with CodeRabbit's Coding Plans for faster development and fewer bugs.

View plan used: #306

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch issue-306-extend-tracing-tests-cover-all-find-footnote-end-branches-and-add-end-to-end-traced-wrapping-tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[pandalump]

@coderabbitai review

[coderabbitai]

✅ Action performed

Review finished.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@src/wrap/tokenize/parsing_tests.rs`:
- Around line 281-282: In the test covering find_footnote_end("[^4] tail", 0),
replace the broad logs_contain checks for "start=" and "end=" with exact
assertions for "start=0" and "end=4", preserving the existing log validation.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: d0d1332f-aaf3-448f-8043-05a595b66bba

📥 Commits

Reviewing files that changed from the base of the PR and between cbf7cb0 and fa76670.

📒 Files selected for processing (3)

• Cargo.toml
• src/wrap/tokenize/parsing_tests.rs
• tests/wrap_unit/footnotes.rs

[coderabbitai]

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Assert the exact span boundaries.

logs_contain("start=") and logs_contain("end=") only verify that the fields exist. For find_footnote_end("[^4] tail", 0), assert start=0 and end=4 so the test catches incorrect span metadata.

As per coding guidelines, tests must be substantive and fail for plausible incorrect implementations.

Suggested assertion tightening

-        assert!(logs_contain("start="));
-        assert!(logs_contain("end="));
+        assert!(logs_contain("start=0"));
+        assert!(logs_contain("end=4"));

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@src/wrap/tokenize/parsing_tests.rs` around lines 281 - 282, In the test
covering find_footnote_end("[^4] tail", 0), replace the broad logs_contain
checks for "start=" and "end=" with exact assertions for "start=0" and "end=4",
preserving the existing log validation.

Source: Coding guidelines