iT邦幫忙

2022 iThome 鐵人賽

DAY 18
0
DevOps

Azure DevOps進階應用系列 第 18

【Azure DevOps進階應用】拆解BuildImage Job成為job範本

  • 分享至 

  • xImage
  •  

前一篇「BuildCode YAML拆解task為step範本」已經將BuildCode Job都拆解成為template,CI Pipeline還剩下BuildImage、DeployCloudRun兩個Job,這篇就延續前面的經驗,繼續拆解BuildImage Job吧!

下面是拆解前的Job內容:

  - job: BuildImage
    dependsOn: BuildCode
    steps:
      - task: DownloadBuildArtifacts@0
        displayName: 下載Pipeline Artifact
        inputs:
          buildType: 'current'
          cleanDestinationFolder: true
          downloadType: 'single'
          artifactName: '$(pipelineArtifact)'
          downloadPath: '$(System.ArtifactsDirectory)/'
      - task: ExtractFiles@1
        displayName: Unzip zip
        inputs:
          archiveFilePatterns: '$(System.ArtifactsDirectory)/$(pipelineArtifact)/$(buildResultZipName)'
          destinationFolder: '$(System.ArtifactsDirectory)/BuildImage'
          cleanDestinationFolder: true
          overwriteExistingFiles: true
      - task: Docker@2
        displayName: Build image
        inputs:
          repository: '$(imgRepository)'
          command: 'build'
          Dockerfile: $(buildDockerfile)
          buildContext: '$(System.ArtifactsDirectory)/BuildImage'
          arguments: '--no-cache'
          tags: |
            latest
      - task: Docker@2
        displayName: "Login to Container Registry"
        inputs:
          command: login
          containerRegistry: $(imgRegistryService)
      - task: Bash@3
        displayName: Push docker image
        inputs:
          targetType: 'inline'
          script: |
            docker push -a $(imgRepository)

在這個Job裡面主要只有兩件事:

  1. 從Pipeline Artifacts下載成品檔案(前面壓縮的zip檔)並且解壓縮
  2. Build Docker Image

這裡我們把下載檔案並且解壓縮的部份拆解為一個step template:

parameters:
  - name: artifactName
    type: string
    default: OutputFiles
  - name: unzip
    type: boolean
    default: false
  - name: zipFileName
    type: string
    default: ''
  - name: unzipToFolderPath
    type: string
    default: ''

steps:
  - task: DownloadBuildArtifacts@0
    displayName: 下載Pipeline Artifact
    inputs:
      buildType: 'current'
      cleanDestinationFolder: true
      downloadType: 'single'
      artifactName: '${{ parameters.artifactName }}'
      downloadPath: '$(System.ArtifactsDirectory)/'
  - ${{ if eq(parameters.unzip, true) }}:
    - task: ExtractFiles@1
      displayName: Unzip zip
      inputs:
        archiveFilePatterns: '$(System.ArtifactsDirectory)/${{ parameters.artifactName }}/${{ parameters.zipFileName }}'
        ${{ if ne(parameters.unzipToFolderPath, '') }}:
          destinationFolder: ${{ parameters.unzipToFolderPath }}
        ${{ else }}:
          destinationFolder: $(System.ArtifactsDirectory)/${{ parameters.artifactName }}/${{ replace(parameters.zipFileName, '.zip','') }}
        cleanDestinationFolder: true
        overwriteExistingFiles: true

在上面拆解為step template的內容中設計了一個unzip參數可以設定是否需要解壓縮,steps裡也可以看到透過條件判斷式來決定是否加入負責解壓縮的task與設定解壓縮路徑的屬性。如果解壓縮的路徑沒有設定(空字串),那就使用預設的路徑(zip檔案去掉副檔名)。

要留意的是task上面的條件判斷式前面有減號「-」,在屬性上面的則是沒有減號「-」的。

再把Docker Build的部份拆解成為另一個step template:

parameters:
  - name: imgRepository
    type: string
  - name: imgTags
    type: object
  - name: buildDockerfile
    type: string
    default: Build.Dockerfile
  - name: buildContext
    type: string
  - name: containerRegistry
    type: string

steps:
  - task: Docker@2
    displayName: Build image
    inputs:
      repository: '${{ parameters.imgRepository }}'
      command: 'build'
      Dockerfile: ${{ parameters.buildDockerfile }}
      buildContext: ${{ parameters.buildContext }}
      arguments: '--no-cache'
      tags: ${{ parameters.imgTags }}
  - task: Docker@2
    displayName: "Login to Container Registry"
    inputs:
      command: login
      containerRegistry: ${{ parameters.containerRegistry }}
  - task: Bash@3
    displayName: Push docker image
    inputs:
      targetType: 'inline'
      script: |
        docker push -a ${{ parameters.imgRepository }}

上一篇
【Azure DevOps進階應用】BuildCode YAML拆解task為step範本
下一篇
【Azure DevOps進階應用】CloudRun環境變數檔與多個Git Repo
系列文
Azure DevOps進階應用30
圖片
  直播研討會
圖片
{{ item.channelVendor }} {{ item.webinarstarted }} |
{{ formatDate(item.duration) }}
直播中

尚未有邦友留言

立即登入留言