Jackson strips leading underscore - jackson

Our swagger specification defines a property _links.
places-result:
type: object
properties:
_links:
$ref: "#/components/schemas/links-underscore-type"
Using swagger-codegen, we see this name end up in the output class:
#JsonProperty("_links")
private LinksUnderscoreType _links = null;
#Schema(
description = ""
)
#Valid
public LinksUnderscoreType getLinks() {
return this._links;
}
However, Jackson adds a property with and without the leading underscore:
{
"result": {
"links": {
"self": "http://localhost:8083/places?page=181",
"previous": "http://localhost:8083/places?page=180",
"first": "http://localhost:8083/places?page=0"
},
"_links": {
"self": "http://localhost:8083/places?page=181",
"previous": "http://localhost:8083/places?page=180",
"first": "http://localhost:8083/places?page=0"
}
}
As it is a Spring Boot application, we should be able to influence the naming strategy using this property:
spring.jackson.property-naming-strategy: LOWER_CASE
None of the accepted values seems to preserve the specified property name.
How can we preserve the leading underscore?

It turns out the non-underscore property came in because it was matched by the getter method. The solution is to not match the getter methods. This can be achieved in the Spring Boot config:
spring:
jackson:
mapper:
AUTO_DETECT_GETTERS: false
Or in code:
objectMapper.configure(MapperFeature.AUTO_DETECT_GETTERS, false);

Related

Use a projection to modify only one field but keeping default for all others

First of all I'd like to say I love what i've seen so far from Spring Data JPA and Spring Data REST. Thanks a lot to all people involved.
Problem description
I have an entity model similar to the classes below. One parent and two different child entities referencing the parent als a ManyToOne Assoziation. For one of the childs i like to have the default rendering of all its properites and links as it is when no projection is applied to the parent.
The other child should be mapped to a simple string array containing only the id or some specific field.
Code and example JSONs
#Entity
public class Parent {
#Id
private Long id;
private String parentValue;
#OneToMany(mappedBy = "parent")
private List<Child1> child1;
#OneToMany(mappedBy = "parent")
private List<Child2> child2;
// ... getters and setters
}
#Entity
public class Child1 {
#Id
private Long id;
private String child1Value;
#ManyToOne
Parent parent;
// ... getters and setters
}
#Entity
public class Child2 {
#Id
private Long id;
#ManyToOne
Parent parent;
}
the response when getting the collection resource of parent is this
{
"_embedded": {
"parents": [
{
"parentValue": "Parent1",
"child1": [
{
"child1Value": "Child1",
"_links": {
"parent": {
"href": "http://localhost:8080/parents/1"
}
}
}
],
"child2": [
{
"_links": {
"parent": {
"href": "http://localhost:8080/parents/1"
}
}
}
],
// removed remaining json to shorten the example
}
But what i like to achieve is the following JSON
{
"_embedded": {
"parents": [
{
"parentValue": "Parent1",
"child1": [
{
"child1Value": "Child1",
"_links": {
"parent": {
"href": "http://localhost:8080/parents/1"
}
}
}
],
"child2": [1],
What i tried so far
Added an excerptProjection to the ParentRepository:
#RepositoryRestResource(excerptProjection = ParentRepository.ArrayProjection.class)
public interface ParentRepository extends PagingAndSortingRepository<Parent, Long>{
public interface ArrayProjection {
String getParentValue();
List<Child1> getChild1();
#Value("#{target.child2.![id]}")
List<Long> getChild2();
}
}
Edited: In the first version of the question, the Projection was incorrect regarding the return type of getChild1(), as it should return the complete collection not only one element. Thanks #kevvvvyp for still trying to help.
The result is similar to what i want, but the links on the Child1 property are missing now:
{
"_embedded": {
"parents": [
{
"child2": [
2
],
"child1": {
"child1Value": "Child1"
},
"parentValue": "Parent1",
// removed remaining json to shorten example
Also the approach with the excerptProjection means i'd have to change the projection everytime the entity changes. Which probalby won't happen to much but that means somebody will forget to change it in the future ;-)
I think a projection is the right way to go here, what about...
interface ArrayProjection {
#JsonIgnore
#Value("#{target}")
Parent getParent();
default String getParentValue(){
return this.getParent().getParentValue();
}
default Child1 getChild1(){
//TODO what are we actually trying to return here? Given we join the parent to
// the child on id & id is the primary key on both entities can we ever get more than one child?
return CollectionUtils.firstElement(this.getParent().getChild1());
}
default List<Long> getChild2() {
return this.getParent().getChild2().stream()
.map(Child2::getId)
.collect(Collectors.toList());
}
}
Response...
GET http://localhost:8080/api/parents
HTTP/1.1 200
Vary: Origin
Vary: Access-Control-Request-Method
Vary: Access-Control-Request-Headers
Content-Type: application/json
Transfer-Encoding: chunked
Date: Wed, 31 Mar 2021 21:08:54 GMT
Keep-Alive: timeout=60
Connection: keep-alive
{
"_embedded": {
"parents": [
{
"parentValue": "Parent1",
"child1": {
"child1Value": "Child1"
},
"child2": [
1
],
"_links": {
"self": {
"href": "http://localhost:8080/api/parents/1"
},
"parent": {
"href": "http://localhost:8080/api/parents/1{?projection}",
"templated": true
}
}
}
]
},
"_links": {
"self": {
"href": "http://localhost:8080/api/parents"
},
"profile": {
"href": "http://localhost:8080/api/profile/parents"
}
},
"page": {
"size": 20,
"totalElements": 1,
"totalPages": 1,
"number": 0
}
}
Response code: 200; Time: 713ms; Content length: 724 bytes
In reponse to your concern, if we code using default method's we will see a compile error when that entity changes. It might be possible to use a class projection instead.
Also I would consider if we really want to return a projection by default... it might confuse a client who then tries to create/update a parent (this is perhaps why you've kept the id's hidden by default?).
Sample - https://github.com/kevvvvyp/sdr-sample
Update
Another (more complex) way to acheive this could be to make use of Jackson's SimpleBeanPropertyFilter, you could customise the JSON response based on some value (the entity class, a value in a particular field, an annotation etc).

Type mismatch during Gson deserialize json response with kotlin

I am trying to serialize the json response below, but I am unsure how to do it.
This is the Json my backend returns:
[
{
"title": "Dummy section, should not be seen",
"type": "dummy_test",
"metadata": []
},
{
"title": "Title1",
"type": "categories_products",
"metadata": [
{
"id": "1272"
}
]
},
{
"title": "Title2",
"type": "categories_products",
"metadata": [
{
"id": "996"
}
]
}
]
This is my ExploreItem class:
data class ExploreItem(
#SerializedName("metadata") val metadata: List<Metadata> = listOf(),
#SerializedName("title") val title: String = "",
#SerializedName("type") val type: String = ""
) {
enum class ExploreItemType(val value: String) {
#SerializedName("unknown")
UNKNOWN("unknown"),
#SerializedName("other_companies")
OTHER_COMPANIES("other_companies"),
#SerializedName("categories_products")
CATEGORIES_PRODUCTS("categories_products"),
#SerializedName("popular_categories")
POPULAR_CATEGORIES("popular_categories")
}
}
data class Metadata(
#SerializedName("id") val id: String = ""
)
And now I am trying to serialize it in the repository like this:
Serializer.defaultJsonParser.fromJson(response.body!!.string(),ExploreItem::class.java )
but it doesn't work because it's expecting a list of ExploreItem. How can I rewrite the serializer expression to parse it into a list?
From your error
Type mismatch. Required:List Found:ExploreItem!
Post errors is very important, Gson is telling you that it wants a List and not an object of ExploreItem.
In other words, you are telling to Gson with the call Serializer.defaultJsonParser.fromJson(response.body!!.string(),ExploreItem::class.java )
"Hey Gson, from the string I want an object ExploreItem", and Gson is telling you "Hey my friend, you string start with [ ] for sure it is a list of something and not a single object."
You need to pass in the Serializer.defaultJsonParser.fromJson(response.body!!.string(),List<ExploreItem>::class.java)
P.s: I'm not sure about the Kotlin syntax

Return string from Web API .NET Core get operation

I have a get operation that I want to return a string from. An example would be
"000875"
When I return this string from a controller in my Web API Controller in full .NET, it formats it like this:
{
"Property": "000875"
}
When I return the string in my converted .NET Core Controller it returns this:
{
"$id": "1",
"$type": "System.Net.Http.HttpResponseMessage, System.Net.Http",
"Version": "1.1",
"Content": {
"$id": "2",
"$type": "System.Net.Http.StringContent, System.Net.Http",
"Headers": [
{
"Key": "Content-Type",
"Value": [
"application/json; charset=utf-8"
]
}
]
},
"StatusCode": "OK",
"ReasonPhrase": "OK",
"Headers": [],
"TrailingHeaders": [],
"RequestMessage": null,
"IsSuccessStatusCode": true
}
It is interesting to note that the value is not even in there!
I am running some interesting JSON Serialization to make BreezeJs work with .NET Core. It is possible that it is the cause of this weirdness:
.AddNewtonsoftJson(opt =>
{
// Let Breeze say how we serialize. This adds in the Type and Id options the way breeze expects
var jsonSerializerSettings = JsonSerializationFns.UpdateWithDefaults(opt.SerializerSettings);
......
I am hoping for a way to get strings through without all this mess. Can that be done?
I get the impression that the subject action definition returns HttpResponseMessage.
public HttpResponseMessage MyAction(....
HttpRequestMessage is no longer a first class citizen in asp.net-core framework and will be treated as a normal model and serialized.
That explains the JSON you are seeing with your controller
The syntax needs to be updated to return IActionResult derived responses
public IActionResult MyAction() {
//...
return Ok("000875");
}
ActionResult<T>
public ActionResult<string> MyAction() {
//...
if(somecondition)
return NotFound();
return "000875";
}
or the model itself.
public string MyAction() {
//...
return "000875";
}
Reference Controller action return types in ASP.NET Core Web API

MarkLogic - Xpath on JSON document

MarkLogic Version: 9.0-6.2
I am trying to apply Xpath in extract-document-data (using Query Options) on a JSON document shown below. I need to filter out "Channel" property if the underneath property "OptIn" has a value of "True".
{
"Category":
{
"Name": "Severe Weather",
"Channels":[
{
"Channel":
{
"Name":"Email",
"OptIn": "True"
}
},
{
"Channel":
{
"Name":"Text",
"OptIn": "False"
}
}
]
}
}
I tried below code,
'<extract-document-data selected="include">' +
'<extract-path>//*[OptIn="True"]/../..</extract-path>' +
'</extract-document-data>' +
which is only pulling from "Channel" property as shown below.
[
{
"Channel": {
"Name": "Email",
"OptIn": "True"
}
}
]
But my need is to pull from parent "Category" property, but filter out the Channels that have OptIn value as False.
Any pointers?
If I understand correctly, you'd like to extract 'Category', but only with those 'Channel's that have 'OptIn' equalling 'true', right?
Extract-document-data is not advanced enough for that. You best extract entire Categories which have at least one OptIn equalling true (//Category[//OptIn = 'true']), and use a REST transform on the search response to trim down the unwanted Channels..
HTH!

How to specify which fields to return with #query anotation in Spring Data?

When I run the following query using the #query anotation in Spring Boot, it returns the correct result:
SELECT p FROM Collection p WHERE LOWER(p.description) LIKE LOWER(CONCAT('%',:searchTerm, '%'))
{
"_embedded": {
"collections": [
{
"place": "Blessington",
"description": "Collection of old shoes for recycling",
"_links": {
"self": {
"href": "http://localhost:8080/collections/1"
},
"collection": {
"href": "http://localhost:8080/collections/1"
}
}
}
]
},
"_links": {
"self": {
"href": "http://localhost:8080/collections/search/findByDescription?searchTerm=shoe"
}
}
}
When I try to specify the fields to return:
SELECT p.description FROM Collection p WHERE LOWER(p.description) LIKE LOWER(CONCAT('%',:searchTerm, '%'))
I get the following error:
{
"cause": null,
"message": "PersistentEntity must not be null!"
}
How do I specify which fields to return with #query annotation in Spring Data?
Yes it seems the question that Manish posted the link to has the answer.
Answer: You can't.
Spring data will return the whole entity, not individual fields. You can't make it do that. If you want to do that you have to use projections. See linked post.
Thanks #Manish
The suggested link does not fully cover all possibilities.
From the Hopper release of Spring Data it is possible to directly return Projections directly from Query methods:
https://spring.io/blog/2016/05/03/what-s-new-in-spring-data-hopper#projections-on-repository-query-methods
So you can therefore do:
public interface ThingRepository extends JpaRepository<Thing, Long>{
#Query("select t from Thing t .....")
public List<ThingProjection> findBySomeCriteria(...);
}