[quality] CaseStudies' undated-study comparator fallbacks have no end-to-end coverage #1064

Finding

sortByPublishedAtDesc in src/components/CaseStudies/index.js compares two
studies through a fallback on each side:

23  function sortByPublishedAtDesc(caseStudies) {
24    return [...caseStudies].sort((a, b) =>
25      (b.publishedAt || '').localeCompare(a.publishedAt || ''),
26    );

Every one of the 213 studies in data/case-studies.json carries a
publishedAt, so neither || '' arm can run, and the empty arm of the date
cell beside it (line 204) cannot render either:

204                  {study.publishedAt &&
205                    new Date(study.publishedAt).toLocaleDateString('en-US', {

Evidence and provenance

• End-to-end: src/components/CaseStudies/index.js is at 92.98%
  regions, uncovered region lines [25, 46, 64]. Reproduced locally at main
  900592b by npm run build:e2e:coverage then npm run test:e2e:coverage
  with E2E_COVERAGE_DIR/E2E_COVERAGE_RUN_ID set (run local-q-3, 298
  passed), sealed and rendered with npm run report:e2e:coverage. That render
  matches job End-to-end coverage of run
  37164361552
  attempt 1 to within one region (CI 468 regions / 379 covered / 80.98%;
  local 467 / 378 / 80.94%), and CaseStudies is identical in both.
• Unit: TZ=UTC node tests/tools/coverage-report.mjs at the same revision
  reports src files at 100.00% lines / 99.84% regions, so this is not a
  unit-coverage gap.

Covered by unit tests but not end to end, so this is a priority 2 gap.

Recommendation

A missing date is a property of a single record, not a cleared document-level
field, so this belongs to the additive overlay mechanism
(tests/e2e/fixtures/data/**, see tests/tools/e2e-data-fixtures.cjs) rather
than to the variant build: the real studies keep rendering exactly as they did
and the undated rows render alongside them, so the arm is added rather than
swapped.

Append two undated studies, not one. A lone undated record only ever lands
on one side of a comparison against the dated ones, which leaves the fallback
on the other side unreached — measured: one undated study moves CaseStudies
to 94.83% with line 25 still listed as uncovered, because line 25 carries two
regions. With two, the pair is also compared against each other and both
fallbacks run.

Then assert in tests/e2e/case-studies.spec.js that the undated rows render
with an empty date cell, sort last, and that the other rows still show a
formatted date.

Measured result of doing exactly that: CaseStudies 92.98% -> 96.55%
(uncovered regions [46, 64]), src files 80.98% -> 81.37%. The result is
unchanged under the region fold proposed in #1051 (87.73% -> 88.37%), so it
does not depend on which of the two lands first.

Lines 46 and 64 (!data.generatedAt and data.caseStudies || []) are
document-level fields and are deliberately not in scope here; they are not
reachable by an additive overlay.

Completion criteria

• An additive overlay appends undated case studies to the coverage build
• tests/e2e/case-studies.spec.js asserts the empty date cell and the sort position
• Region line 25 no longer appears in the End-to-end coverage report for CaseStudies

Priority

• Impact: medium (a comparator fallback that silently decides row order for any study the upstream feed publishes without a date)
• Effort: low

🐝 Hive Agent: quality | Instance: hosted-available-lke648397-260827-5n31 | SHA: 900592b

— hive: agent=quality backend=copilot model=claude-opus-5 copilot=1.0.88