声明性管道在语法方面有一些限制。之所以出现此错误，是因为在environment块中，只能分配两种类型的表达式：

• 字符串（单引号或双引号）
• 从函数调用返回的值

但是，您需要注意，环境变量仅存储字符串值，因此，如果您从函数调用返回数组（或任何其他类型），它将自动转换为toString()表示形式。

pipeline {
    agent any 

    environment {
        MYPODS = getPods()
    }

    stages {
        stage("Test") {
            steps {
                script {
                    println "My pods = ${env.MYPODS}"
                }
            }
        }
    }
}

def getPods() {
    return ['pod1', 'pod2']
}

控制台输出：

[Pipeline] node
[Pipeline] {
[Pipeline] withEnv
[Pipeline] {
[Pipeline] stage
[Pipeline] { (Test)
[Pipeline] script (hide)
[Pipeline] {
[Pipeline] echo
<java.lang.String@ed6c7b35 value=[pod1, pod2] hash=-311657675>
[Pipeline] echo
MYPODS = [pod1, pod2]
[Pipeline] echo
Item: [
[Pipeline] echo
Item: p
[Pipeline] echo
Item: o
[Pipeline] echo
Item: d
[Pipeline] echo
Item: 1
[Pipeline] echo
Item: ,
[Pipeline] echo
Item:  
[Pipeline] echo
Item: p
[Pipeline] echo
Item: o
[Pipeline] echo
Item: d
[Pipeline] echo
Item: 2
[Pipeline] echo
Item: ]
[Pipeline] }
[Pipeline] // script
[Pipeline] }
[Pipeline] // stage
[Pipeline] }
[Pipeline] // withEnv
[Pipeline] }
[Pipeline] // node
[Pipeline] End of Pipeline
Finished: SUCCESS

溶液

如果要存储字符串值列表，则可以将其定义为以,字符分隔的单个值字符串。在这种情况下，只需将其标记为值列表即可。请考虑以下示例：

pipeline {
    agent any 

    environment {
        MYPODS = 'pod1,pod2,pod3'
    }

    stages {
        stage("Test") {
            steps {
                script {
                    MYPODS.tokenize(',').each {
                        println "Item: ${it}"
                    }
                }
            }
        }
    }
}

输出量：

[Pipeline] node
[Pipeline] {
[Pipeline] withEnv
[Pipeline] {
[Pipeline] stage
[Pipeline] { (Test)
[Pipeline] script
[Pipeline] {
[Pipeline] echo
Item: pod1
[Pipeline] echo
Item: pod2
[Pipeline] echo
Item: pod3
[Pipeline] }
[Pipeline] // script
[Pipeline] }
[Pipeline] // stage
[Pipeline] }
[Pipeline] // withEnv
[Pipeline] }
[Pipeline] // node
[Pipeline] End of Pipeline
Finished: SUCCESS

您的环境应该在代理之后

pipeline {

agent any
environment {
    def mypods = []
}
stages {
    stage('Getting pods') {
        steps {
            script {
               withKubeConfig(caCertificate: '.....', credentialsId: '.....', serverUrl: '.....') {
                    env.mypods = sh "kubectl get pod | grep Running | awk '{print \$1}'"
                }
            }
        }
    }
    stage('Print pods') {
        steps {
            script {
                mypods.each {
                    println "Item: $it"
                }
            }
        }
    }
}

}

您也可以使用.split（“，”）函数和for-in循环进行迭代。

pipeline {
agent any 

environment {
    MYPODS = 'pod1,pod2,pod3'
}

stages {
    stage("Test") {
        steps {
            script {
              for (pod in MYPODS.split(",")) {
                  println("${pod}")
               }
            }
        }
    }
}

}