Valid Jenkins syntax rejected as NglParseError

[CarlInglisBJSS]

First, thanks for a fantastic tool.

I've found a bit of an edge case, which I fully appreciate is outside of the scope of a groovy parser, but I was wondering if you have any thoughts on how to address it.

// ...
  lock(lockValue) {
    // do stuff
  }
// ...

This is valid Jenkins syntax (because the pre-parser turns it into something more groovy-ish) but NGL is reporting:

  31    error    Unexpected input: '{' @ line 31, column 19.  NglParseError    

I'm working my way through our files with NGL, and so for this file I can just not NGL it as it's a small file that doesn't change. However we use this construct in several places in much bigger and more complex pipelines.

Any thoughts would be appreciated.

Thanks.

[nvuillam]

Hi @CarlInglisBJSS , NGL parse error is caused by the default groovy parser, so maybe it's related to the groovy version?
NGL is still needs to upgrade because it's stuck on groovy 3 for the moment

Maybe u could share the full jenkinsfile?

[CarlInglisBJSS]

Hi @nvuillam,

Unfortunately I can't share the full file as it's client-confidential - however commenting out the lock and digging deeper into the resulting errors has shown that the actual problem was a missing comma on this block - which was inside the lock code:

      resultsMap = readJSON(
        text: awsResults['stdout']   // missing comma at the end here
        returnPOJO: true
      )

I guess it was reporting the issue on the lock line since that was the start of the block that it fell over on.

This sample file produces a similar problem (although not quite the same):

void doStuffInALock(String lockValue) {

  String dummyVariable

  lock(lockValue) {

    Object resultsMap = readJSON(
        text: dummyVariable
        returnPOJO: true
    )

  }

}

Thanks for the rapid response and for causing me to try and create a sample, which didn't do what I expected, and thus caused me to dig deeper. :)