[JENKINS-48308] Translate StringGroovyMethods.eachLine

When I tried to parse Maven output (which is running using sh -step), noticed that groovy code only sees first line of output. Example pipeline:

node() { 

def outputLog 

outputLog = sh(script: "ls -la /", returnStdout: true) 

println outputLog

//prints full log

outputLog.eachLine { line, count -> println "$count: $line" }

//prints only 1 line 

outputlog = "ls -la /".execute().text outputlog.eachLine { println it }

//prints all lines 

}

So workaround is to use that Groovy's execute.

---
Originally reported by hifi, imported from: Translate StringGroovyMethods.eachLine

• status: Open
  
• priority: Minor
  
• component(s): workflow-cps-plugin
  
• resolution: Unresolved
  
• votes: 0
  
• watchers: 3
  
• imported: 2025-12-07
  

Raw content of original issue

When I tried to parse Maven output (which is running using sh -step), noticed that groovy code only sees first line of output. Example pipeline:



node() { 

def outputLog 

outputLog = sh(script: "ls -la /", returnStdout: true) 

println outputLog

//prints full log

outputLog.eachLine { line, count -> println "$count: $line" }

//prints only 1 line 

outputlog = "ls -la /".execute().text outputlog.eachLine { println it }

//prints all lines 

}



So workaround is to use that Groovy's execute.

Reproduce the reported pipeline using sh with returnStdout and StringGroovyMethods.eachLine, then inspect the workflow-cps-plugin handling of that call. Compare it with the working execute().text example and verify that multi-line shell output is processed line by line when the issue is fixed.

Written by the indexing model from the issue text.