#165 — column-header tooltips: authored descriptions + column-level data tests

[cmbays]

Summary

Column headers on the unit-test given/expected DataTables now carry a focusable ⓘ tooltip showing the column's authored description plus its column-level data tests (unique, not_null, accepted_values with its values, relationships with to/field, package-qualified tests like dbt_expectations.*) — so you can see a column is expected-unique while reading fixture rows.

Closes #165

What changed

• Domain — TestMetadata POD (generic-test name / namespace / untyped kwargs) + four tolerant Node fields: column_descriptions (non-empty prose only, kept SEPARATE from the .contract column-set so a description edit never flags state:modified.contract), column_name, attached_node, test_metadata. Additive with_* builders (the with_incremental_mode precedent).
• Manifest adapter — ingests columns[*].description (fusion serializes an unset description as "" — dropped at translation) and the test-node attribution fields; singular tests (no test_metadata) and model-level tests (column_name: null) tolerated per ADR-5.
• Render — ColumnMetaPayload {description, tests} computed fully in Rust (JS only renders): per-table maps resolved against the model that owns the table's columns — ref(...) given → the input model, this given → the target model, expect → the target model — filtered to the rendered table's columns, key omitted when empty (no empty bubbles).
• Template — the #145 — surface incremental-model unit-test semantics #146 expect-tooltip pattern mirrored exactly: focusable <button> + CSS bubble on :hover AND :focus/:focus-visible, aria-label for AT, bubble aria-hidden, never a native title. One deliberate divergence: the bubble is position:fixed with JS-set coordinates, because the th lives inside the .table-fit overflow-x scroller that would clip an absolutely-positioned bubble (#145 — surface incremental-model unit-test semantics #146's bubble hangs off the un-clipped panel header; this one cannot). Visibility itself stays pure CSS. A click on the button is swallowed so it never re-sorts the DataTable. The Diff grid's unified column axis shares the same map.

Scope decisions (settled at issue triage — not re-litigated here)

• v1 resolves ONLY true column-scoped data tests: manifest test nodes with column_name set and attached (attached_node) to the owning model. Model-level tests with column args are deferred.
• Description source of truth = the manifest model node's columns map (works without --project-root). No raw-YAML path.
• Affordance only on columns with a description AND/OR ≥1 column test — no empty bubbles.

Additional v1 boundaries surfaced while building: source(...) givens resolve to nothing (manifest sources are not ingested in v0.x), seed-backed ref(...) givens likewise (resolve_target_model is model-only), and column-name matching is exact-case.

dbt-fusion verification (research-first)

Pinned at dbt-labs/dbt-fusion@9977b6cbb1b761065536300037560d8e3c037011:

• DbtColumn { name, data_type, description } — crates/dbt-schemas/src/schemas/dbt_column.rs; serialize_dbt_column_desc + #[serialize_always] means an unset description hits the wire as "", never an absent key — the adapter drops empties.
• ManifestDataTest { column_name, attached_node, test_metadata } — crates/dbt-schemas/src/schemas/manifest/manifest_nodes.rs (#[skip_serializing_none] ⇒ singular tests OMIT test_metadata).
• TestMetadata { name, kwargs (defaulted), namespace } — crates/dbt-schemas/src/schemas/nodes.rs.

Cross-checked against the REAL committed fixtures (not just synthetic JSON): jaffle-shop carries unique/not_null/accepted_values/relationships with namespace: null and an EMPTY description on stg_customers.customer_id; playground carries 10 singular tests without test_metadata, 33 model-level tests with null column_name, and dbt_expectations-namespaced tests — all pinned by a new real-fixture ingestion test.

Tests

• Unit: domain POD round-trips; adapter tolerance (empty descriptions, explicit nulls, singular/model-level tests); summarize_column_test arg rendering; per-owner given/this/expect resolution; empty-map wire omission.
• Headless (Chrome): focus reveals the bubble; bare column has NO button; given table resolves its own input model's metadata; aria parity.
• BDD: extends features/report_generation.feature (no new .feature file — the feature-count gate stays 12) with payload-fact scenarios over the real subprocess + committed fixture.

Dogfood + goldens

• Live (dbt-project, YAML only): unique/not_null on order_events_incremental.order_id + staging descriptions on the columns the marts unit tests mock.
• Synthetic showcase: playground's fct_encounters_incremental gains columns/descriptions + unique/not_null on encounter_id in BOTH the source YAML and the manifest (hand-edited via a verified byte-stable JSON round-trip; synthetic-only, no recompile, no root_path). The playground fixture already carried rich column metadata on dim_payers/mart_dq_summary, so all three goldens now show descriptions, unique/not_null, accepted_values (values: …), dbt_expectations.*, and the this-given resolution.
• All three goldens regenerated with the exact CI invocations and verified byte-identical to renderer output; per-golden payload diffs verified to be column_meta-only (Python structural diff, not just eyeballs).

Finding (out of lane, not fixed here)

StateComparator::models_in_scope arm 2 inserts EVERY modified node with zero unit tests — without filtering resource_type == "model". A baseline-mode PR that adds a generic test node surfaces it as a junk "model" card named by its hash suffix (observed live when the new synthetic test nodes briefly rendered as a165c01d01/b165c01d02 cards). Worked around in the fixture by mirroring the test nodes into the baseline; worth its own issue.

Gates (all run directly — lefthook skips in fresh worktrees)

• cargo fmt --all --check ✓
• cargo clippy --all-targets --locked -- -D warnings — exit 0 ✓ (one justified #[allow(clippy::struct_field_names)]: attached_node mirrors the dbt wire key)
• cargo nextest run — 746 passed ✓
• cargo test --test bdd — 86 scenarios / 550 steps ✓
• cargo test --test headless_zero_egress --locked -- --ignored ✓; cargo test --test headless_toggle --locked -- --ignored — 29 passed ✓
• RUSTDOCFLAGS="-D warnings" cargo doc --no-deps --document-private-items --locked ✓
• cargo deny check ✓
• Example byte-identity simulated locally for all three goldens ✓

Zero-egress invariants hold: no new resource refs, pure-CSS visibility + text-node DOM building, no JS parsing engines — all metadata computed in Rust.

🤖 Generated with Claude Code

---

[coderabbitai]

Warning

Review limit reached

@cmbays, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 29 minutes and 14 seconds. Learn how PR review limits work.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans include higher PR review limits than trial, open-source, and free plans. In all cases, reviews become available again over time. During sustained high-volume PR review activity, CodeRabbit may temporarily slow when the next review becomes available.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: c4830a0e-69f7-4a01-ada5-de14c79a18e0

📥 Commits

Reviewing files that changed from the base of the PR and between f2d8632 and 43b45ca.

⛔ Files ignored due to path filters (1)

• tests/snapshots/render_integration__rendered_chrome_jaffle_shop.snap is excluded by !**/*.snap

📒 Files selected for processing (18)

• dbt-project/models/marts/_incremental__models.yml
• dbt-project/models/staging/schema.yml
• examples/diff-showcase-report.html
• examples/jaffle-shop-report.html
• examples/playground-report.html
• features/report_generation.feature
• src/adapters/manifest.rs
• src/adapters/render.rs
• src/domain/manifest.rs
• src/domain/mod.rs
• templates/report.html
• tests/fixtures/MANIFEST.toml
• tests/fixtures/playground-baseline.json
• tests/fixtures/playground-current.json
• tests/fixtures/playground-source/models/marts/core/_core__models.yml
• tests/headless_toggle.rs
• tests/manifest_ingestion.rs
• tests/steps/report_generation.rs

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feature-165-column-tooltips

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[github-actions]

📄 Rendered report preview

All golden examples regenerated cleanly.

🟡 Golden examples

Committed to examples/ and byte-identity gated — the canonical reports contributors and consumers browse. Stable across PRs.

Report	View	Download
playground-report.html	▶ Open ↗	⬇ Download
jaffle-shop-report.html	▶ Open ↗	⬇ Download
diff-showcase-report.html	▶ Open ↗	⬇ Download

🐶 Live dogfood preview

This PR's own dbt-project/ diff, freshly compiled by fusion into an ephemeral manifest (never committed) and rendered with --pr-diff. Regenerated every PR — the live self-dogfood, not a committed example.

Report	View	Download
dbt-project-report.html	▶ Open ↗	⬇ Download

▶ Open ↗ opens the report in your browser in one click —
published to this repo's GitHub Pages under /pr-166/.
⬇ Download fetches the same self-contained HTML as a workflow
artifact (auth-gated; works fully offline). Either way the report
makes zero external resource requests.

The Pages preview may take ~1 min to update after this comment
posts. On PRs from forks the Open link is unavailable (read-only
token) — use Download.

Alternative: GitHub CLI

# gh CLI >= 2.63 extracts into ./report-preview-playground/.
gh run download 27248614343 -R breezy-bays-labs/cute-dbt -n report-preview-playground
open report-preview-playground/playground-report.html

---

_{Posted by report-preview.yml for 43b45ca24936c437f9c11fcc86749e8af5fc40d7. Affordance only — never blocks merge.}

[gemini-code-assist]

Code Review

This pull request implements column-header tooltips (cute-dbt#165) to display authored column descriptions and column-level data tests in the generated reports. It updates manifest ingestion to parse test attributions and column descriptions, builds the metadata payload in the rendering adapter, and adds interactive tooltips in the HTML template. Feedback on these changes highlights a potential performance bottleneck in column_meta_for_model due to redundant full-manifest scans, suggesting caching or pre-grouping. Additionally, a UI issue was identified where the position: fixed tooltips can become detached from column headers during scrolling, and it is recommended to reposition or hide them on scroll events.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.