Azure Pipelines Documentation

Build Pipelines

This document provides a comprehensive guide to creating, configuring, and managing build pipelines in Azure DevOps.

What is a Build Pipeline?

A build pipeline is a series of automated steps that compile source code, run tests, and produce build artifacts. These artifacts are then typically published for use in downstream deployment (release) pipelines.

Creating a Build Pipeline

You can create build pipelines using either the classic editor or YAML.

YAML Pipelines

YAML pipelines offer a code-based approach, allowing you to store your pipeline definitions alongside your application code. This promotes versioning, reusability, and easier collaboration.

A basic YAML build pipeline structure might look like this:

trigger:
- main

pool:
  vmImage: 'ubuntu-latest'

steps:
- script: echo Hello, world!
  displayName: 'Run a one-line script'

- script: |
    echo Add another greet
    echo Hello from YAML!
  displayName: 'Run a multi-line script'

Classic Editor

The classic editor provides a visual interface for defining your build process. It's suitable for simpler pipelines or for those new to Azure Pipelines.

Key Concepts in Build Pipelines

• Triggers: Events that initiate a pipeline run, such as code commits, pull requests, or scheduled times.
• Agent Pools: Groups of agents (virtual machines or physical servers) that execute your pipeline jobs.
• Tasks: Individual automation steps that perform specific actions, like compiling code, running tests, publishing artifacts, or deploying.
• Jobs: A collection of steps that run on a single agent.
• Stages: A logical grouping of jobs, often representing a distinct phase in the build process (e.g., Build, Test, Package).
• Artifacts: The output of a build, such as compiled binaries, deployment packages, or test results.

Common Build Tasks

• NuGet, npm, Maven, Gradle: For package management and building.
• DotNetCoreCLI: For .NET Core projects.
• MSBuild: For .NET Framework projects.
• Copy Files: To copy files between directories.
• Publish Build Artifacts: To make outputs available for later stages or other pipelines.
• Run PowerShell/Bash/Command Script: For custom scripting.

Best Practices for Build Pipelines

• Automate everything: From code checkout to artifact publishing.
• Keep builds fast: Optimize your build process to reduce wait times.
• Build only once: Produce your artifacts once and promote them through subsequent stages.
• Use YAML: For version control and better collaboration.
• Secure your pipeline: Manage secrets and service connections carefully.
• Integrate tests: Run unit, integration, and other tests as part of your build.

Further Reading

• Microsoft Docs: Build Pipelines
• Microsoft Docs: YAML Schema