How can I iterate through JSON objects even there is error using Scala Jackson

Currently, jackson is rejecting the whole JSON when there is blank property value.

I want to use com.fasterxml.jackson.* to parse JSON code.

As you see below input JSON, name attribute is blank some of the elements.

Hense iterate through JSON objects will ignore by Jackson.

hence there will be 2 elements formed as part of the output.

I am using below code but no luck

 def readJsonString[T](content: String)(implicit m: Manifest[T]): T = {
val objectMapper = new ObjectMapper(new JsonFactory().enable(Feature.ALLOW_COMMENTS)) with ScalaObjectMapper
objectMapper.registerModule(DefaultScalaModule)
objectMapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false)
objectMapper.configure(DeserializationFeature.WRAP_EXCEPTIONS, false)
objectMapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false)
objectMapper.configure(SerializationFeature.FAIL_ON_EMPTY_BEANS, false)
objectMapper.readValue(content)
}

//Existing Json that I want to use as input where some attribute has blank value

[
{
"name": "Invalid",
"ruleType": "validation_1",
"inputs": [ { "Name": "", "country": ["USA"] } ]
},
{
"name": "",
"ruleType": "validation_2",
"inputs": [ { "Name": "Test", "place": ["USA"] } ]
},
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": []
}
{
"name": "Valid",
"ruleType": "validation_2",
"inputs": [ { "Name": "Test", "place": ["USA"] } ]
},
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": [ { "Name": "Test", "place": ["France"] } ]
}
]

//New Json that will created from above which has proper name attribute value

[
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": [ { "Name": "Test", "country": ["USA"] } ]
}
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": [ { "Name": "Test", "place": ["France"] } ]
}
]

Here is one solution that would work using jackson:

Case classes to map your json to:

  case class NamePlace(Name: String, country: Seq[String])
case class NameRuleTypeInputs(name: String, ruleType: String, inputs: Seq[NamePlace])

Custom serializer to include field validations before serializing:

  class NameRuleTypeInputsSerializer(defaultSerializer: JsonSerializer[Object]) extends JsonSerializer[NameRuleTypeInputs] {
override def serialize(value: NameRuleTypeInputs, gen: JsonGenerator, serializers: SerializerProvider): Unit = {
if (isValid(value)) {
defaultSerializer.serialize(value, gen, serializers)
}
}
private def isValid(value: NameRuleTypeInputs) = {
!Option(value.name).getOrElse("").isEmpty &&
Option(value.inputs).getOrElse(Seq.empty).nonEmpty &&
!value.inputs.exists(i => Option(i.Name).getOrElse("").isEmpty)
}
}

Updated your objectMapper to include the custom serializer:

  val objectMapper = new ObjectMapper(new JsonFactory().enable(Feature.ALLOW_COMMENTS)) with ScalaObjectMapper
objectMapper.registerModule(DefaultScalaModule)
objectMapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false)
objectMapper.configure(DeserializationFeature.WRAP_EXCEPTIONS, false)
objectMapper.configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false)
objectMapper.configure(SerializationFeature.FAIL_ON_EMPTY_BEANS, false)
objectMapper.registerModule(new SimpleModule(){
override def setupModule(context: Module.SetupContext): Unit = {
super.setupModule(context)
context.addBeanSerializerModifier(new BeanSerializerModifier() {
override def modifySerializer(config: SerializationConfig, beanDesc: BeanDescription, serializer: JsonSerializer[_]): JsonSerializer[_] = {
if(classOf[NameRuleTypeInputs] isAssignableFrom beanDesc.getBeanClass) {
new NameRuleTypeInputsSerializer(serializer.asInstanceOf[JsonSerializer[Object]])
} else {
serializer
}
}
})
}
})

Methods to read/write json:

  def readJsonString[T](content: String)(implicit m: Manifest[T]): T = {
objectMapper.readValue(content)
}
def writeJsonString(nameRuleTypeInputsList: Seq[NameRuleTypeInputs]): String = {
// Using pretty printer for readability
objectMapper.writerWithDefaultPrettyPrinter().writeValueAsString(nameRuleTypeInputsList)
}

A small test:

val testJson =
"""
[
{
"name": "Invalid",
"ruleType": "validation_1",
"inputs": [ { "Name": "", "country": ["USA"] } ]
},
{
"name": "",
"ruleType": "validation_2",
"inputs": [ { "Name": "Test", "country": ["USA"] } ]
},
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": []
},
{
"name": "Valid",
"ruleType": "validation_2",
"inputs": [ { "Name": "Test", "country": ["USA"] } ]
},
{
"name": "Valid",
"ruleType": "validation_1",
"inputs": [ { "Name": "Test", "country": ["France"] } ]
}
]
""".stripMargin

  val namedRuleTypeInputs: Seq[NameRuleTypeInputs] = readJsonString[Seq[NameRuleTypeInputs]](testJson)
println(writeJsonString(namedRuleTypeInputs))

Output:

[ {
"name" : "Valid",
"ruleType" : "validation_2",
"inputs" : [ {
"Name" : "Test",
"country" : [ "USA" ]
} ]
}, {
"name" : "Valid",
"ruleType" : "validation_1",
"inputs" : [ {
"Name" : "Test",
"country" : [ "France" ]
} ]
} ]

Useful Reference: https://www.baeldung.com/jackson-serialize-field-custom-criteria