Declarative Syntax Support
Pipeline Templates can be written using Declarative Syntax.
An Example
Pipeline Template
pipeline {
agent any
stages {
stage("example") {
steps {
echo pipelineConfig.message (1)
static_code_analysis() (2)
echo dev.arbitrary_field
}
}
}
}| 1 | Declarative templates are able to resolve the injected pipelineConfig variable |
| 2 | Steps loaded by libraries, such as static_code_analysis in this example, can be resolved in the steps block |
Pipeline Configuration
message = "example"
libraries{
sonarqube (1)
}
application_environments{
dev{
arbitrary_field = "whatever" (2)
}
}| 1 | the sonarqube library provides the static_code_analysis step in this example |
| 2 | the dev application environment is resolvable form the declarative pipeline template |
Step Resolution
Occassionally, library developers will create steps in JTE libraries that override actual Jenkins DSL steps (ex: build).
When writing a template using the scripted pipeline syntax, these JTE library steps will take precedence.
However, when writing a pipeline template using the scripted pipeline syntax, the Jenkins DSL step will take precedence.
To invoke the JTE library step, use the script block.
Using the example of a JTE step called build, which overrides the Jenkins DSL step build:
Pipeline Template
pipeline {
agent any
stages {
stage("example") {
steps {
build() (1)
script{
build() (2)
steps.build() (3)
}
}
}
}
}| 1 | will invoke the Jenkins DSL build step |
| 2 | will invoke the JTE library build step |
| 3 | will invoke the Jenkins DSL build step using the steps variable, which provides direct access to invoke DSL methods |