Pipeline syntax

Microtica has built-in support for defining and running pipelines. You write the pipeline declaratively in YAML, directly in the microtica.yaml file at the root of your source code.

Here is the skeleton of a Microtica pipeline:

runtime:
  computeType: SMALL|MEDIUM|LARGE

stages:
 - [stage-name]
 - [another-stage-name]

steps:
  step_name:
    [step-body]
  other_step_name:
    [step-body]

• runtime (optional) – the runtime configuration for the pipeline
  • computeType – runtime environment compute type
• stages (optional) – a list of stage names. Each stage can be assigned to one or multiple steps
• steps – a dictionary of steps

Basic pipeline definition

Here is a simple pipeline definition:

steps:
  Clone:
    title: Clone my source code from Git
    type: git-clone
  BuildAndTest:
    title: Build my microservice
    image: node:12-alpine
    commands:
     - npm install
     - npm run build
  PushDockerImage:
    type: docker-push
    image_name: my_app_image
    tag: {{version}}
    registry: dockerhub

This pipeline has three steps:

• Clone – clones the source code from the Git repository
• BuildAndTest – installs node modules and compiles the Node.js source code
• PushDockerImage – builds and pushes a Docker image to the specified registry

Runtime environment

Each pipeline step runs in a separate runtime environment to keep step execution secure and isolated.

You can set the compute type of the runtime environment at the pipeline level and at the individual step level.

To configure the compute type at the pipeline level, see the example below:

runtime:
  computeType: SMALL

To configure the compute type at the step level and override the pipeline configuration, see the example below:

runtime:
  computeType: SMALL

steps:
  Clone:
    title: Clone my source code from Git
    type: git-clone
  Build:
    image: node:12-alpine
    runtime:
      computeType: MEDIUM
    commands:
      - npm install
  Test:
    image: node:12-alpine
    commands:
      - npm test

With this configuration, the Test step uses the default compute environment of type SMALL, but the Build step uses the compute environment of type MEDIUM.

Execution flow

Steps run in serial, in the order they are defined in the pipeline.

Here is an example of a pipeline with multiple steps:

steps:
  Clone:
    title: Clone my source code from Git
    type: git-clone
  BuildMyApp:
    image: node:12-alpine
    commands:
      - npm install
      - npm run build
  TestMyApp:
    image: node:12-alpine
    commands:
      - npm test
  GenerateTestReports:
    image: node:12-alpine
    commands:
      - npm run generate-reports

The pipeline runs Clone first, then BuildMyApp. When that step finishes successfully, TestMyApp runs. Following the same logic, GenerateTestReports runs once TestMyApp completes successfully.

Next steps

• Add steps to a pipeline
• Run steps in parallel with stages
• Parametrize pipelines with variables