How do I easily get a bracketed constituency tree

[drjiangyang]

I'm using Stanza's CoreNLP client to do constituency parsing on many texts. I want just the bracketed trees in this form:

(NP-SBJ (NNP Mr.) (NNP Vinken) )

However, following the client basic usage:

# get the first sentence
sentence = ann.sentence[0]

# get the constituency parse of the first sentence
constituency_parse = sentence.parseTree
print(constituency_parse)

I get the result like this:

child {
  child {
    child {
      child {
        value: "Chris"
      }
      value: "NNP"
      score: -9.281864166259766
    }
    ...
  }
  ...
  value: "S"
  score: -50.052059173583984
}
value: "ROOT"
score: -50.20326614379883

How do I easily convert it to a bracketed tree in Penn Treebank format?
Many thanks!

[AngledLuffa]

Can you use the Stanza constituency parser (which is substantially more accurate), or do you specifically need the CoreNLP constituency parser?

[drjiangyang]

Thank you Maintainer!
The Stanza constituency parser runs well. Given that it outperforms the CoreNLP constituency parser in accuracy, I'll choose it in my task.