extracted docs of types include escaped linebreaks in signature

[woeps]

This rescript code:

type variant = FirstButVeryLongAndMeaninglessVariantConstructor

type t = {thisIsSomeReallyLongFieldWhichIsJustHereForDemonstration: bool}

gets extracted as:

{
  "name": "MinimalExample",
  "docstrings": [],
  "items": [
  {
    "id": "MinimalExample.variant",
    "kind": "type",
    "name": "variant",
    "signature": "type variant =\\n  | FirstButVeryLongAndMeaninglessVariantConstructor", <-- HERE
    "docstrings": [],
    "detail": 
    {
      "kind": "variant",
      "items": [
      {
        "name": "FirstButVeryLongAndMeaninglessVariantConstructor",
        "docstrings": [],
        "signature": "FirstButVeryLongAndMeaninglessVariantConstructor"
      }]
    }
  }, 
  {
    "id": "MinimalExample.t",
    "kind": "type",
    "name": "t",
    "signature": "type t = {\\n  thisIsSomeReallyLongFieldWhichIsJustHereForDemonstration: bool,\\n}", <-- HERE
    "docstrings": [],
    "detail": 
    {
      "kind": "record",
      "items": [{
        "name": "thisIsSomeReallyLongFieldWhichIsJustHereForDemonstration",
        "optional": false,
        "docstrings": [],
        "signature": "bool"
      }]
    }
  }]
}

Note the escaped linebreaks (\\n) in the signature.
This seems to happen, if a signature reaches some minimal length.

I would expect the line breaks to not be escaped.

edit:
I believe, this is due to calling Json.escape.

Not sure, what's best. But I guess, it's ether not escaping those special characters or "unescaping" them, when actually parsing the json in Tools_DocGen.res?