playwright-java-release

作者: microsoft

在滚动PR合并后准备Playwright Java发布——切出发布分支,标记Maven版本,起草GitHub发布,并勾选Java…

npx skills add https://github.com/microsoft/playwright-java --skill playwright-java-release

Use this skill once the chore: roll driver to 1.X.0 PR has merged into main and the upstream JS v1.X.0 is published. The rolling work itself is covered by the [[playwright-roll]] skill.

Throughout this doc, replace X with the minor version (e.g. 60 for 1.60.0) and <user> with the fork owner (gh api user --jq .login).

The full release checklist lives in the private microsoft/playwright-internal repo as the v1.X checklist issue. Find its number once:

unset GITHUB_TOKEN
ISSUE=$(gh search issues --repo microsoft/playwright-internal "v1.X checklist" --json number --jq '.[0].number')

Tick each Java box incrementally (one PATCH per item) so the issue reflects accurate state if the flow is interrupted:

gh api repos/microsoft/playwright-internal/issues/$ISSUE --jq '.body' > /tmp/body.md
# edit /tmp/body.md to flip "- [ ]" → "- [x]" on the relevant Java item
gh api repos/microsoft/playwright-internal/issues/$ISSUE -X PATCH --field body=@/tmp/body.md

1. Cut the release branch

Push release-1.X from current upstream/main (which now contains the merged roll commit):

git fetch upstream main
git push upstream upstream/main:refs/heads/release-1.X

2. Draft the GitHub release

Generate the release notes from the upstream docs:

cd ~/playwright
node utils/render_release_notes.mjs java 1.X > /tmp/v1.X.0-release-notes.md

The renderer leaves JS-isms that need fixing for Java. Apply these substitutions — the list is not exhaustive, eyeball the diff before publishing:

  • toMatchAriaSnapshot() → matchesAriaSnapshot()
  • toHaveCSS() → hasCSS() (and other toHaveX matchers → hasX)
  • browser.on('context') → browser.onContext()
  • browserContext.on('download' | 'frameattached' | ...) → browserContext.onDownload() / onFrameAttached() / …

Create the draft directly against release-1.X — drafting against main and retargeting later is fragile because every gh release edit rotates the untagged-<hash> ID:

gh release create v1.X.0 --repo microsoft/playwright-java --draft \
  --title "v1.X.0" --notes-file /tmp/v1.X.0-release-notes.md --target release-1.X

3. Bump the Maven version on the release branch

Cut mark-v-1.X.0 off upstream/release-1.X, run set_maven_version.sh, and PR back to the release branch:

git checkout -b mark-v-1.X.0 upstream/release-1.X
./scripts/set_maven_version.sh 1.X.0
git add -u
git commit -m "chore: mark 1.X.0"
git push -u origin mark-v-1.X.0
gh pr create --repo microsoft/playwright-java --head <user>:mark-v-1.X.0 --base release-1.X \
  --title "chore: mark 1.X.0" \
  --body "Updates Maven version in all modules to \`1.X.0\` for the v1.X release."

set_maven_version.sh only invokes mvn versions:set on pom.xml, tools/*/pom.xml, and examples/pom.xml, but the root invocation cascades through the reactor, so the expected diff is 11 poms: root + driver/ + driver-bundle/ + playwright/ (from the reactor cascade) + 6 under tools/ + examples/, all flipping 1.<prev>.0-SNAPSHOT → 1.X.0. Any other file in the diff is a red flag.

4. Publish

The user publishes the draft release manually once the mark-v-1.X.0 PR is merged. After publishing, the v1.X.0 tag triggers the Azure DevOps pipelines that push the artifacts to Maven Central (.azure-pipelines/publish.yml) and the Docker images to ACR (.azure-pipelines/publish-docker.yml).

来自 microsoft 的更多技能

oss-growth
microsoft
OSS增长黑客角色
agent-framework-azure-ai-py
microsoft
使用Microsoft Agent Framework Python SDK(agent-framework-azure-ai)构建Azure AI Foundry代理。在创建使用AzureAIAgentsProvider的持久化代理、使用托管工具(代码解释器、文件搜索、网络搜索)、集成MCP服务器、管理对话线程或实现流式响应时使用。涵盖函数工具、结构化输出和多工具代理。
development
airunway-aks-setup
microsoft
在AKS上设置AI Runway——从裸集群到运行模型。涵盖集群验证、控制器安装、GPU评估、提供商设置和首次部署。适用场景:“设置AI Runway”、“接入AKS集群”、“安装AI Runway”、“airunway设置”、“将模型部署到AKS”、“在AKS上进行GPU推理”、“在AKS上配置KAITO”、“在AKS上运行LLM”、“在AKS上使用vLLM”、“在AKS上设置模型服务”、“AI Runway控制器”。
devops
appinsights-instrumentation
microsoft
使用Azure Application Insights对Web应用进行插桩的指南。提供遥测模式、SDK设置和配置参考。适用场景:如何对应用进行插桩、App Insights SDK、遥测模式、什么是App Insights、Application Insights指南、插桩示例、APM最佳实践。
devops
applicationinsights-web-ts
microsoft
使用Application Insights JavaScript SDK(@microsoft/applicationinsights-web)为浏览器/Web应用添加检测。用于真实用户监控(RUM)——页面视图、点击、AJAX/fetch依赖项、异常、自定义事件,以及与后端OpenTelemetry追踪关联的浏览器端GenAI代理追踪。涵盖SDK加载器脚本和npm设置、框架扩展(React、React Native、Angular)、点击分析、遥测初始化器,以及从浏览器发出的代理/工具/模型跨度所遵循的OTel GenAI语义约定。
devops
azure-ai-anomalydetector-java
microsoft
使用适用于 Java 的 Azure AI 异常检测器 SDK 构建异常检测应用程序。在实现单变量/多变量异常检测、时间序列分析或 AI 驱动的监控时使用。
development
azure-ai-language-conversations-py
microsoft
使用azure-ai-language-conversations Python SDK实现对话语言理解(CLU)。当使用ConversationAnalysisClient分析对话意图和实体、构建NLP功能或将语言理解集成到应用程序中时使用。
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python。用于机器学习工作区、作业、模型、数据集、计算资源和管道。 触发词:“azure-ai-ml”、“MLClient”、“工作区”、“模型注册表”、“训练作业”、“数据集”。
development