Infrastructure: Restore Windows Storybook focus-regression parity

Aberta
  • Algo errado?
#4,347 0 comentários 0 reações 0 responsáveis Ver no GitHub

Mantenedores costumam responder em até 1 dia

Ninguém assumiu esta issue ainda.

Avaliação

Dificuldade
5/5
Tempo estimado
Mais de uma semana
Facilidade para iniciantes
25/100
Tipo de issue
Funcionalidade
Clareza
Razoavelmente clara
Status de atividade
Ativa
Stack de tecnologia
react-native, storybook, typescript
Domínio
desktop-dev, testing-qa

Direção de pesquisa

Start with packages/agentic/components/src/components/button/button.stories.tsx, packages/agentic/components/src/common/desktopFocus.wdio.ts, and the executable-test contract in packages/agentic/storybook-desktop/README.md. Compare the retired coverage from PR #4208 with current generated story IDs and native testIDs, then verify the named focus-survival, visibility, status, CI, and artifact requirements on an unlocked Windows Fabric lane without restoring Appium or the retired Jest harness.

Escrita pelo modelo de indexação a partir do texto da issue.

Descrição

Goal

Restore the complete regression coverage from the retired Windows Jest Storybook harness through the supported native Storybook/WDIO path. Parent: #4275. Follow-up to the completed retirement milestone #4254.

Status - 2026-10-01

Open; retirement is complete, regression parity is not. PR #4294 removed the legacy harness. PRs #4306, #4313, #4314, and #4320 delivered native providers, executable story tests, modern focus behavior, and FocusZone regressions.

The current catalog has 29 named WDIO cases across 17 stories, including Button, Switch, Input, TabList, and FocusZone focus behavior. Those newer tests are valuable but do not recreate the original ten delayed focus-survival cases. The current story callbacks/helpers contain no original three-second post-activation survival cohort.

Required historical coverage

The load-bearing regression from PR #4208 was a delayed 0xc0000409 application termination after focus acquisition. Preserve a minimum three-second post-activation observation, native focus confirmation, and terminal app-crash detection for each original consumer:

Consumer Required regression
Tag Focus acquisition and delayed survival
Accordion Header focus acquisition and delayed survival
Tab Selected-tab focus acquisition and delayed survival
ListboxItem Focus acquisition and delayed survival
Checkbox Focus acquisition and delayed survival
MenuItem Selected-item focus acquisition and delayed survival
ListItem Selected-item focus acquisition and delayed survival
Radio Focus acquisition and delayed survival
Switch Focus acquisition and delayed survival
Card Interactive-card focus acquisition and delayed survival
Button Real Tab movement from one enabled button to the next

Also reconcile the original Button/Icon/Callout visibility and Callout native-window status checks with the replacement coverage. Use generated current story IDs and stable native testIDs, not the obsolete IDs/XPath selectors from the removed harness.

Acceptance criteria

  • Each of the ten delayed cases is authored as a named, typed, component-colocated WDIO case and retains the three-second regression window.
  • The Button real-Tab movement case and original visibility/status assertions have active replacement coverage.
  • The tests verify native focus and application survival, not only JavaScript state or immediate renderability. Crashes during the observation window terminate the run as failures.
  • Real pointer activation is qualified on an unlocked interactive Windows Fabric endpoint. Programmatic focus can supplement, but must not be mislabeled as a physical-click regression pass.
  • All required cases execute in an active Windows Fabric CI lane, unquarantined, with actual pass evidence. Hosted keyboard/pointer skips do not establish parity.
  • Accordion's currently excluded Fabric module has a safe, documented native test path; exclusions are not silently counted as coverage.
  • Changes to FocusVisual or its consumers trigger/review this cohort, and run artifacts identify each executed/skipped case and exact app/helper/manifest identity.
  • Ownership-safe teardown, process/input recovery, documentation, and relevant package checks remain intact. Do not restore Appium or the retired Jest harness.

Evidence and coordination

Linguagem predominante
TypeScript
Estrelas
1.4k
Forks
179
Merge médio
1d 10h
PRs com merge (30d)
50

Preparar o ambiente

Primeiros passos

  1. Leia a issue inteira e depois o guia de contribuição do projeto.
  2. Comente na issue dizendo que vai assumir — evita que duas pessoas façam o mesmo trabalho.
  3. Faça um fork do repositório e trabalhe em uma branch.
  4. Abra um pull request que referencie o número da issue.

Mais de microsoft/fluentui-react-native

Todas as issues de microsoft/fluentui-react-native

Issues semelhantes

Mais issues de TypeScript

Receba novas issues na sua caixa de entrada

Um resumo curto de issues do GitHub para quem está começando.