If you want to maintain multiple versions of an object for JSON conversion, Google GSON library has nice @Since annotation for it. This annotation can be applied to fields and classes.
For example, suppose you are maintaining the multiple versions of the your REST API and you are using JSON as end response payload. In next iteration of your API you are added some of fields for particular JSON entity and you don’t want to send the newly added fields to previous API version then @Since annotation comes into picture. Lets see in following listing, how we use this feature.
To make use of this annotation, you must configure the Gson instance to specific version using GsonBuilder, of course.
Output of above as
If you are not specifying any version it will include the all the fields regardless of its version.
Disable HTML Escaping
By default, during conversion any html characters contained, will be converted to their corresponding Unicode i.e. < to \u003c, > to \u003e , & to \u0026and so on
For passing the html characters as is, you need to configure the gson instance to with use of GsonBuilder#disableHtmlEscaping() method.
Following listing shows the usage -
Output as follows
Exclude the fields from Json Output
For this, Google GSON has 4 ways to deal,
fields with transient and static modifier
exclude with fields with specific modifiers
Using Expose annotation
User defined exclusion strategies
Lets see each of this in detail-
1. fields with transient and static modifier
This is default behavior, where you modify the field with transient modifier it will not included as we saw in first article in series and of course, one with static modifier are also excluded as they are part of class and not the instance.
2. exclude fields with specific modifiers
You can configure gson instance, such that it will exclude the fields with modifiers you specified, e.g. you can exclude/ignore the fields which have protected or private modifier.
To leverage this you need to use GsonBuilder#excludeFieldsWithModifiers() as shown in following series of listing.
For this example, we are not including the private fields. Although you can exclude fields with any modifiers (any modifiers that applies to fields)
In following output, you can see private fields are excluded as per our program, Although you can ignore fields which can have protected, synchronized etc.
3. Using Expose annotation
Gson also provides an annotation which you can mark for the fields, thus those fields will be excluded from serialized output..
For using this we need to following two
configure gson instance as follows
Gson gson= GsonBuilder().excludeFieldsWithoutExposeAnnotation().create()
and mark the fields with @Expose annotations, which are to be included in final json output. So, one which are not marked will be excluded
Following listing shows the detailed usage
Ouput as follows-
4. User defined exclusion strategy
Google Gson library provides very granular level of exclusions of fields based on type of field and field attributes.
For this you need to implement the ExclusionStrategy interface by implementing the two of its methods such as shouldSkipClass() and shouldSkipField(). Former can implemented such that it will skip fields based on type of it and later can implemented based on field attribute such its modifier, its annotation etc
and now after implementation ExclusionStrategy interface, you need to pass it to GsonBuilder#setExclusionStrategies() method for configuring the gson instance.
Following is minimal implementation of use of Exclusion strategy, in which we exclude the fields are having type as String, List and also field which are having annotation as Deprecated
Note the following implementation of ExclusionStrategy
Our custom Developer class for showing example
Output of above as follows
Custom field naming using field naming policy
As we saw in last article that Gson defaulted to include the output json field name same as that of class field name and also we can override this with use annotation @serilizedName.
In Gson there is more we can define the json field name to change to Uppercase, lowercase etc without use of any annotation.
For this you need to use GsonBuilder().setFieldNamingPolicy() method and passing the appropriate FieldNamePolicy
Following listing shows usage