Convert multi dimensional array in Logstash

Question

How can i convert multi dimensional SQL json string to valid JSON object so that it can be mapped to Elasticsearch index template field types.

Example case:

Let's assume that this is my Elasticsearch index template.

{
  "settings": {
    "number_of_shards": "5"
  },
  "mappings": {
    "_doc": {
      "properties": {
        "manager": {
          "properties": {
            "age": {
              "type": "integer"
            },
            "name": {
              "type": "text"
            }
          }
        },
        "employees": {
          "type": "nested",
          "properties": {
            "age": {
              "type": "integer"
            },
            "name": {
              "type": "text"
            },
            "addresses": {
              "type": "nested",
              "properties": {
                "street": {
                  "type": "text"
                },
                "city": {
                  "type": "text"
                }
              }
            }
          }
        }
      }
    }
  }
}

Problematic case is "employees" array and it's inner "addresses" array JSON conversion.

Example SQL data for "employees":

[{"age":30,"name":"User 1","addresses":"[{\"street\":"Home street 1",\"city\":"New York"},{\"street\":"Home street 2",\"city\":"Washington"}]"}]

Logstash filter configuration: Conversion works if "employees" does not contain nested array. But problem comes when i array has nested array.

filter {
  ruby {
    code => "
      require 'json'

      employees_json_value = JSON.parse(event.get('employees').to_s)
      event.set('employees',employees_json_value )
    "
  }
}

Question is how can i convert "employees" column nested "addresses" array to JSON object so that i can map it to Elasticsearch index template.