Files
kandev/docs/plans/resizable-markdown-tables/task-02-prove-resize-interaction.md
T

3.2 KiB

id, title, status, wave, depends_on, plan, spec
id title status wave depends_on plan spec
02-prove-resize-interaction Prove Markdown table resizing and mobile parity done 2
01-build-resize-renderer
plan.md ../../specs/ui/resizable-markdown-tables.md

Task 02: Prove Resize Interaction and Mobile Parity

Acceptance

  • A desktop browser test drags a separator from its body-row hit area and proves equal-and-opposite adjacent width changes, an unchanged third column, a stable table width, and 64-pixel clamping.
  • Desktop tests prove double-click reset, arrow-key adjustment, and Enter reset.
  • A locally scrolling wide table keeps separators aligned after scrolling and creates no chat- or document-level horizontal overflow.
  • Mobile Chrome exposes no separator controls and preserves ordinary table wrapping plus wide-table local scrolling.
  • Existing Markdown wrapping regressions remain green.

TDD sequence

  1. RED: extend the desktop wrapping fixture with a three-column interaction case and prove current tables have no separator.
  2. GREEN: drive pointer, double-click, and keyboard behavior against the Task 01 implementation; refine only externally observable defects.
  3. RED/GREEN: add mobile absence and parity assertions to the existing phone fixture.
  4. Run both complete focused files to catch wrapping or scroll-owner regressions.

Verification

make build-web
(cd apps/web && pnpm e2e:run tests/chat/markdown-wrap.spec.ts)
(cd apps/web && pnpm e2e:run --project mobile-chrome tests/chat/mobile-markdown-wrap.spec.ts)
(cd apps/web && pnpm run typecheck)
(cd apps && pnpm --filter @kandev/web lint)
git diff --check

Files likely touched

  • apps/web/e2e/tests/chat/markdown-wrap.spec.ts
  • apps/web/e2e/tests/chat/mobile-markdown-wrap.spec.ts
  • Task 01 production files only if browser evidence reveals an interaction bug

Dependencies

  • 01-build-resize-renderer

Parallelism

sequential. Browser assertions depend on the completed renderer contract and may legitimately feed a defect back into Task 01 files.

Inputs

  • Existing exact pnpm-table and wide-table fixtures in both wrapping suites
  • docs/specs/ui/resizable-markdown-tables.md, especially desktop interaction, local-scroll ownership, and mobile capability scenarios

Output contract

Report desktop and mobile command results, before/after column and table widths, the tested hit-point location, local/document overflow measurements, files changed, and any unsupported browser behavior.

Result

  • RED: the desktop user-flow test failed because the shared renderer exposed no resize separator.
  • GREEN: the rebuilt production bundle passed all 8 desktop Markdown wrapping tests and both Mobile Chrome Markdown tests.
  • The desktop test drags from the first body row, verifies equal-and-opposite 60-pixel adjacent changes, an unchanged third column and table width, the 64-pixel clamp, double-click reset, 8-pixel keyboard adjustment, and Enter reset.
  • A wide desktop table keeps its separator aligned after 160 pixels of local scroll. Chat and document overflow checks remain clean.
  • Mobile Chrome exposes no separators while ordinary tables wrap and wide tables retain local scrolling. No unsupported browser behavior remains in scope.