JsonUnit is a library that simplifies JSON comparison in tests.
There are several different APIs you can use. They all have more or less the same features, just the usage is slightly different.
The recommended API is AssertJ integration which combines the power of JsonUnit and AssertJ.
import static net.javacrumbs.jsonunit.assertj.JsonAssertions.assertThatJson;
import static net.javacrumbs.jsonunit.assertj.JsonAssertions.json;
...
// compares two JSON documents (note lenient parsing of expected value)
assertThatJson("{\"a\":1, \"b\":2}").isEqualTo("{b:2, a:1}");
// objects are automatically serialized before comparison
assertThatJson(jsonObject).isEqualTo("{\n\"test\": 1\n}");
// AssertJ map assertions (numbers are converted to BigDecimals)
assertThatJson("{\"a\":1}").isObject().containsEntry("a", BigDecimal.valueOf(1));
// Type placeholders
assertThatJson("{\"a\":1, \"b\": {\"c\" :3}}")
.isObject().containsValue(json("{\"c\" :\"${json-unit.any-number}\"}"));
// AssertJ string assertion
assertThatJson("{\"a\": \"value\"")
.node("a").isString().isEqualTo("value");
// AssertJ array assertion
assertThatJson("{\"a\":[{\"b\": 1}, {\"c\": 1}, {\"d\": 1}]}")
.node("a").isArray().contains(json("{\"c\": 1}"));
// Can ignore array order
assertThatJson("{\"a\":[{\"b\": 1}, {\"c\": 1}, {\"d\": 1}]}")
.when(Option.IGNORING_ARRAY_ORDER).node("a").isArray()
.isEqualTo(json("[{\"c\": 1}, {\"b\": 1} ,{\"d\": 1}]"));
// custom matcher
assertThatJson("{\"test\":-1}")
.withConfiguration(c -> c.withMatcher("positive", greaterThan(valueOf(0))))
.isEqualTo("{\"test\": \"${json-unit.matches:positive}\"}");
// and
assertThatJson("{\"test\":{\"a\":1, \"b\":2, \"c\":3}}").and(
a -> a.node("test.a").isEqualTo(1),
a -> a.node("test.b").isEqualTo(2)
);
// JsonPath support
assertThatJson(json)
.inPath("$.store.book")
.isArray()
.contains(json(
" {\n" +
" \"category\": \"reference\",\n" +
" \"author\": \"Nigel Rees\",\n" +
" \"title\": \"Sayings of the Century\",\n" +
" \"price\": 8.96\n" +
" }"
));
JsonUnit tries to be clever when parsing the expected value. If the value can be parsed as valid JSON, it's parsed so. If it can't be parsed, it's considered to be just a string to be compared. It usually works, but it can lead to unexpected situations, usually with primitive values like numbers and booleans.
// This test does NOT pass. "1" is parsed as JSON containing number 1, the actual value is a string.
assertThatJson("{\"id\":\"1\", \"children\":[{\"parentId\":\"1\"}]}")
.inPath("children[*].parentId")
.isArray()
.containsOnly("1");
// You have to wrap the expected value by `JsonAssertions.value()`
// to prevent parsing
assertThatJson("{\"id\":\"1\", \"children\":[{\"parentId\":\"1\"}]}")
.inPath("children[*].parentId")
.isArray()
.containsOnly(value("1"));
// "true" is valid JSON so it gets parsed to primitive `true`
// Have to wrap it to JsonAssertions.value() in order to make sure it's not parsed
assertThatJson("{\"root\":[\"true\"]}").node("root").isArray().containsExactly(value("true"));
On the other hand, if you want to make sure that the expected value is parsed as JSON, use JsonAssertions.json()
.
You can move between vanilla AssertJ and JsonUnit using asInstanceOf
as in the following example:
import static net.javacrumbs.jsonunit.assertj.JsonAssertions.JSON;
import static net.javacrumbs.jsonunit.assertj.JsonAssertions.assertThatJson;
...
record DummyResponse(String trackingId, String json) {}
DummyResponse resp = new DummyResponse("abcd-0001", "{ \"foo\": \"bar\" }");
assertThat(resp)
.hasFieldOrPropertyWithValue("trackingId", "abcd-0001") // <- Assertj API
.extracting("json")
.asInstanceOf(JSON)
.isObject() // <- JsonUnit API
.containsEntry("foo", "bar");
Following Kotlin API is supported (notice different import)
// Kotlin
import net.javacrumbs.jsonunit.assertj.assertThatJson
assertThatJson("""{"root":{"a":1, "b": 2}}""") {
isObject
node("root.a").isEqualTo(1)
node("root.b").isEqualTo(2)
}
To use AssertJ integration, import
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-assertj</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
For more examples see the tests.
You use Hamcrest matchers in the following way
import static net.javacrumbs.jsonunit.JsonMatchers.*;
import static org.junit.Assert.*;
import static net.javacrumbs.jsonunit.core.util.ResourceUtils.resource;
...
assertThat("{\"test\":1}", jsonEquals("{\"test\": 1}"));
assertThat("{\"test\":1}", jsonPartEquals("test", 1));
assertThat("{\"test\":[1, 2, 3]}", jsonPartEquals("test[0]", 1));
assertThat("{\"test\":{\"a\":1, \"b\":2, \"c\":3}}",
jsonEquals("{\"test\":{\"b\":2}}").when(IGNORING_EXTRA_FIELDS));
// Can use other Hamcrest matchers too
assertThat("{\"test\":1}", jsonPartMatches("test", is(valueOf(1))))
assertThat("{\"test\":1}", jsonEquals(resource("test.json")));
To use import
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
For more examples see the tests.
JsonUnit supports Spring MVC test assertions. For example
import static net.javacrumbs.jsonunit.spring.JsonUnitResultMatchers.json;
...
mockMvc.perform(get("/sample").andExpect(
json().isEqualTo("{\"result\":{\"string\":\"stringValue\", \"array\":[1, 2, 3],\"decimal\":1.00001}}")
);
mockMvc.perform(get("/sample").andExpect(
json().node("result.string2").isAbsent()
);
mockMvc.perform(get("/sample").andExpect(
json().node("result.array").when(Option.IGNORING_ARRAY_ORDER).isEqualTo(new int[]{3, 2, 1})
);
mockMvc.perform(get("/sample").andExpect(
json().node("result.array").matches(everyItem(lessThanOrEqualTo(valueOf(4))))
);
Following Kotlin DSL is supported:
mockMvc.get(path).andExpect {
jsonContent {
node("root").isEqualTo(CORRECT_JSON)
}
}
Inside jsonContent
you have access to all AssertJ API capabilities as described here.
To use import
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-spring</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
For more examples see the tests.
To integrate with Spring WebTest client do
import static net.javacrumbs.jsonunit.spring.WebTestClientJsonMatcher.json;
...
client.get().uri(path).exchange().expectBody().consumeWith(
json().isEqualTo("{\"result\":{\"string\":\"stringValue\", \"array\":[1, 2, 3],\"decimal\":1.00001}}")
);
client.get().uri(path).exchange().expectBody().consumeWith(
json().node("result.string2").isAbsent()
);
client.get().uri(path).exchange().expectBody().consumeWith(
json().node("result.array").when(Option.IGNORING_ARRAY_ORDER).isEqualTo(new int[]{3, 2, 1})
);
client.get().uri(path).exchange().expectBody().consumeWith(
json().node("result.array").matches(everyItem(lessThanOrEqualTo(valueOf(4))))
);
For Kotlin, you can use our bespoke DSL
import net.javacrumbs.jsonunit.spring.jsonContent
...
client.get().uri(path).exchange().expectBody()
.jsonContent {
isEqualTo(CORRECT_JSON)
}
Import
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-spring</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
For more examples see the tests.
Since version 4.0.0 JsonUnit supports Spring MockMvc AssertJ assertions. You can use it like this:
import static net.javacrumbs.jsonunit.assertj.JsonAssertions.jsonUnitJson;
...
assertThat(mvc.get().uri("/sample"))
.hasStatusOk()
.bodyJson()
.convertTo(jsonUnitJson()) // Switch to JsonUnit assert
.inPath("result.array") // This is JsonUnit
.isArray()
.containsExactly(1, 2, 3);
import static net.javacrumbs.jsonunit.spring.JsonUnitRequestMatchers.json;
...
mockServer.expect(requestTo(URI))
.andExpect(json().isEqualTo(json))
.andRespond(withSuccess(jsonResponse, MediaType.APPLICATION_JSON_UTF8));
To use import
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-spring</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
For more examples see the tests.
JsonUnit supports Kotest assertions.
Import:
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-kotest</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
And enjoy:
"""{"test":1}""" should equalJson("""{"test": 1}""")
// Provide configuration
"""{"test":1.01}""" should equalJson("""{"test":1}""", configuration { withTolerance(0.1) })
// Use inPath
"""{"test":1}""" inPath "test" should equalJson("1")
// Clues with nesting
"""{"test": {"nested": 1}}""".inPath("test").asClue {
it inPath "nested" should equalJson("2")
}
"""{"test":1}""".inPath("test").shouldBeJsonNumber()
// shouldBeJsonNumber returns BigDecimal, so we can use standard kotest assertions
// PLease note that numbers are converted to BigDecimals
.shouldBeEqualComparingTo(valueOf(1))
// The same for arrays generated by JsonPath
"""{"test": [{"a": "a"}, {"a": true}, {"a": null}, {"a": 4}]}""".inPath("$.test[*].a")
.shouldBeJsonArray()
.shouldContainExactly("a", true, null, valueOf(4))
// ... and objects
"""{"a":1, "b": true}""".shouldBeJsonObject().shouldMatchAll(
"a" to { it should beJsonNumber() },
"b" to { it should beJsonBoolean() }
)
See the tests for more examples.
JsonUnit support all this features regardless of API you use.
You can use JsonPath navigation together with JsonUnit. It has native support in AssertJ integration, so you can do something like this:
// AssertJ style
assertThatJson(json)
.inPath("$.store.book")
.isArray()
.contains(json(
" {\n" +
" \"category\": \"reference\",\n" +
" \"author\": \"Nigel Rees\",\n" +
" \"title\": \"Sayings of the Century\",\n" +
" \"price\": 8.96\n" +
" }"
));
For the other API styles you have to first import JsonPath support module
<dependency>
<groupId>net.javacrumbs.json-unit</groupId>
<artifactId>json-unit-json-path</artifactId>
<version>4.0.0</version>
<scope>test</scope>
</dependency>
and then use instead of actual value
import static net.javacrumbs.jsonunit.jsonpath.JsonPathAdapter.inPath;
...
// Fluent assertions
assertThatJson(inPath(json, "$.store.book[*].author"))
.when(Option.IGNORING_ARRAY_ORDER)
.isEqualTo("['J. R. R. Tolkien', 'Nigel Rees', 'Evelyn Waugh', 'Herman Melville']");
Sometimes you need to ignore certain values when comparing. It is possible to use ${json-unit.ignore}
or #{json-unit.ignore}
placeholder like this
// AssertJ API
assertThatJson("{\"a\":1}")
.isEqualTo(json("{\"a\":\"${json-unit.ignore}\"}"));
Please note that the assertion will fail if the test
element is missing in the actual value.
If the element needs to be ignored completely you can use ${json-unit.ignore-element}
placeholder.
// AssertJ API
assertThatJson("{\"root\":{\"test\":1, \"ignored\": null}}")
.isEqualTo("{\"root\":{\"test\":1, \"ignored\": \"${json-unit.ignore-element}\"}}");
The assertion will not fail if the element is missing in the actual value.
whenIgnoringPaths
configuration option makes JsonUnit ignore the specified paths in the actual value. If the path
matches, it's completely ignored. It may be missing, null or have any value. Also when(paths(...), thenIgnore()
can be used.
// AssertJ style
assertThatJson("{\"root\":{\"test\":1, \"ignored\": 1}}")
.whenIgnoringPaths("root.ignored"))
.isEqualTo("{\"root\":{\"test\":1}}");
// Hamcrest matcher
assertThat(
"{\"root\":{\"test\":1, \"ignored\": 2}}",
jsonEquals("{\"root\":{\"test\":1, \"ignored\": 1}}").whenIgnoringPaths("root.ignored")
);
Array index placeholder
assertThatJson("[{\"a\":1, \"b\":2},{\"a\":1, \"b\":3}]")
.whenIgnoringPaths("[*].b")
.isEqualTo("[{\"a\":1, \"b\":0},{\"a\":1, \"b\":0}]");
Please note that if you use JsonPath, you should start the path to be ignored by $
Also note that whenIgnoringPaths
method supports full JsonPath syntax only in AssertJ API, all the other flavors support only
exact path or array index placeholder as described above.
JsonPath with whenIgnoringPaths example:
// AssertJ API
assertThatJson("{\"fields\":[" +
"{\"key\":1, \"name\":\"AA\"}," +
"{\"key\":2, \"name\":\"AB\"}," +
"{\"key\":3, \"name\":\"AC\"}" +
"]}")
.whenIgnoringPaths("$.fields[?(@.name=='AA')].key")
.isEqualTo("{\"fields\":[" +
"{\"key\":2, \"name\":\"AA\"}," +
"{\"key\":2, \"name\":\"AB\"}," +
"{\"key\":3, \"name\":\"AC\"}" +
"]}");
It is also possible to use regular expressions to compare string values
assertThatJson("{\"test\": \"ABCD\"}")
.isEqualTo("{\"test\": \"${json-unit.regex}[A-Z]+\"}");
For matching just part of the string, you can use this (we have to escape twice, once for Java, once for JSON)
assertThatJson("{\"test\": \"This is some text followed by: ABCD, followed by this\"}")
.isEqualTo("{\"test\": \"${json-unit.regex}^\\\\QThis is some text followed by: \\\\E[A-Z]+\\\\Q, followed by this\\\\E$\"}");
Since this is quite hard to write, you can implement an expression builder like this.
If you want to assert just a type, but you do not care about the exact value, you can use any-* placeholder like this
assertThatJson("{\"test\":\"value\"}")
.isEqualTo("{test:'${json-unit.any-string}'}");
assertThatJson("{\"test\":true}")
.isEqualTo("{\"test\":\"${json-unit.any-boolean}\"}");
assertThatJson("{\"test\":1.1}")
.isEqualTo("{\"test\":\"${json-unit.any-number}\"}");
You can also use hash instead of string #{json-unit.any-string}
for example if you are using language with string interpolation
like Kotlin.
In some special cases you might want to use your own matcher in the expected document.
assertThatJson("{\"test\":-1}")
.withMatcher("positive", greaterThan(valueOf(0)))
.isEqualTo("{\"test\": \"${json-unit.matches:positive}\"}");
In even more special cases, you might want to parametrize your matcher.
Matcher<?> divisionMatcher = new DivisionMatcher();
assertThatJson("{\"test\":5}")
.withMatcher("isDivisibleBy", divisionMatcher)
.isEqualTo("{\"test\": \"${json-unit.matches:isDivisibleBy}3\"}");
private static class DivisionMatcher extends BaseMatcher<Object> implements ParametrizedMatcher {
private BigDecimal param;
public boolean matches(Object item) {
return ((BigDecimal)item).remainder(param).compareTo(ZERO) == 0;
}
public void describeTo(Description description) {
description.appendValue(param);
}
@Override
public void describeMismatch(Object item, Description description) {
description.appendText("It is not divisible by ").appendValue(param);
}
public void setParameter(String parameter) {
this.param = new BigDecimal(parameter);
}
}
If you need a matcher with more than one parameter, you can implement it like this.
There are multiple options how you can configure the comparison
TREATING_NULL_AS_ABSENT - fields with null values are equivalent to absent fields. For example, this test passes
assertThatJson("{\"test\":{\"a\":1, \"b\": null}}")
.when(TREATING_NULL_AS_ABSENT)
.isEqualTo("{\"test\":{\"a\":1}}");
IGNORING_ARRAY_ORDER - ignores order in arrays
assertThatJson("{\"test\":[1,2,3]}")
.when(IGNORING_ARRAY_ORDER)
.isEqualTo("{\"test\":[3,2,1]}");
IGNORING_EXTRA_ARRAY_ITEMS - ignores unexpected array items
assertThatJson("{\"test\":[1,2,3,4]}")
.when(IGNORING_EXTRA_ARRAY_ITEMS)
.isEqualTo("{\"test\":[1,2,3]}");
assertThatJson("{\"test\":[5,5,4,4,3,3,2,2,1,1]}")
.when(IGNORING_EXTRA_ARRAY_ITEMS, IGNORING_ARRAY_ORDER)
.isEqualTo("{\"test\":[1,2,3]}");
IGNORING_EXTRA_FIELDS - ignores extra fields in the compared value
assertThatJson("{\"test\":{\"a\":1, \"b\":2, \"c\":3}}")
.when(IGNORING_EXTRA_FIELDS)
.isEqualTo("{\"test\":{\"b\":2}}");
IGNORE_VALUES - ignores values and compares only types
assertThatJson("{\"a\":2,\"b\":\"string2\"}")
.when(paths("a", "b"), then(IGNORING_VALUES))
.isEqualTo("{\"a\":1,\"b\":\"string\"}");
FAIL_FAST - stops comparison at the first difference. May provide performance benefits for large JSONs and use-cases that do not require to know all the differences.
It is possible to combine options.
assertThatJson("{\"test\":[{\"key\":3},{\"key\":2, \"extraField\":2},{\"key\":1}]}")
.when(IGNORING_EXTRA_FIELDS, IGNORING_ARRAY_ORDER)
.isEqualTo("{\"test\":[{\"key\":1},{\"key\":2},{\"key\":3}]}");
REPORTING_DIFFERENCE_AS_NORMALIZED_STRING - reports errors as normalized strings. Applies some basic normalization to compared values and then reports full json strings, so you can see the difference in the IDE diff.
In Hamcrest assertion you can set the option like this
assertThat("{\"test\":{\"a\":1, \"b\":2, \"c\":3}}",
jsonEquals("{\"test\":{\"b\":2}}").when(IGNORING_EXTRA_FIELDS));
You can define options locally (for specific paths) by using when(path(...), then(...))
:
// ignore array order for [*].a
// AssertJ
assertThatJson("{\"test\":{\"a\":1,\"b\":2,\"c\":3}}").when(paths("test.c"), then(IGNORING_VALUES))
.isEqualTo("{\"test\":{\"a\":1,\"b\":2,\"c\":4}}");
// ignore array order everywhere but [*].b
assertThatJson("[{\"b\":[4,5,6]},{\"b\":[1,2,3]}]")
.when(IGNORING_ARRAY_ORDER)
.when(path("[*].b"), thenNot(IGNORING_ARRAY_ORDER))
.isEqualTo("[{\"b\":[1,2,3]},{\"b\":[4,5,6]}]");
// ignore extra fields in the object "a"
assertThatJson("{\"a\":{\"a1\":1,\"a2\":2},\"b\":{\"b1\":1,\"b2\":2}}")
.when(path("a"), then(IGNORING_EXTRA_FIELDS))
.isEqualTo("{\"a\":{\"a1\":1},\"b\":{\"b1\":1}}"))
// ignore extra array items in the array
assertThatJson("{\"a\":[1,2,3]}")
.when(path("a"), then(IGNORING_EXTRA_ARRAY_ITEMS))
.isEqualTo("{\"a\":[1,2]}");
// Hamcrest
assertThat("{\"test\":{\"a\":1,\"b\":2,\"c\":3}}",
jsonEquals("{\"test\":{\"a\":1,\"b\":2,\"c\":4}}").when(path("test.c"), then(IGNORING_VALUES)));
Note that TREATING_NULL_AS_ABSENT and IGNORING_VALUES require exact paths to ignored fields:
// ignoring number and str
assertThatJson("{\"a\":2,\"b\":\"string2\"}")
.when(paths("a", "b"), then(IGNORING_VALUES))
.isEqualTo("{\"a\":1,\"b\":\"string\"}");
// treat null B as absent B
assertThatJson("{\"A\":1,\"B\":null}")
.when(path("B"), then(TREATING_NULL_AS_ABSENT))
.isEqualTo("{\"A\":1}");
All other options require paths to objects or arrays where values or order should be ignored.
You can use negative numbers to index arrays form the end
assertThatJson("{\"root\":{\"test\":[1,2,3]}}")
.node("root.test[-1]").isEqualTo(3);
Numbers are by default compared in the following way:
- If the type differs, the number is different. So 1 and 1.0 are different (int vs. float). This does not apply when Moshi is used since it parses all numbers as Doubles.
- Floating number comparison is exact, down to the scale - 1.0 and 1.00 are considered to be different.
You can change this behavior by setting tolerance. If you set tolerance to 0
two numbers are considered equal if they are
equal mathematically even though they have different type or precision (a.compareTo(b) == 0
).
assertThatJson("{\"test\":1.00}").node("test").withTolerance(0).isEqualTo(1);
If you set tolerance to non-zero value, the values are considered equal if abs(a-b) <= tolerance
.
assertThatJson("{\"test\":1.00001}").node("test").withTolerance(0.001).isEqualTo(1);
Or you can use Hamcrest matcher
import static java.math.BigDecimal.valueOf;
...
assertThatJson("{\"test\":1.10001}").node("test")
.matches(closeTo(valueOf(1.1), valueOf(0.001)));
If you are interested why 1 and 1.0 are treated as different numbers please read this comment.
If you want to have special handling of numerical values, you can inject your own number comparator.
assertThatJson("{\"a\":1.0}")
.withConfiguration(c -> c.withNumberComparator(numberComparator))
.isEqualTo("{\"a\":1.00}");
Sometimes you have dots in JSON element names, and you need to address those elements. It is possible to escape dots like this
assertThatJson("{\"name.with.dot\": \"value\"}").node("name\\.with\\.dot").isStringEqualTo("value");
Writing JSON string in Java is huge pain. JsonUnit parses expected values leniently, so you do not have to quote keys, and you can use single quotes instead of double quotes. Please note that the actual value being compared is parsed in strict mode.
assertThatJson("{\"a\":\"1\", \"b\":2}").isEqualTo("{b:2, a:'1'}");
If you need to customize Jackson 2 Object Mapper, you can do using SPI.
Implement net.javacrumbs.jsonunit.providers.Jackson2ObjectMapperProvider
.
public class Java8ObjectMapperProvider implements Jackson2ObjectMapperProvider {
private final ObjectMapper mapper;
private final ObjectMapper lenientMapper;
public Java8ObjectMapperProvider() {
mapper = new ObjectMapper().registerModule(new JavaTimeModule());
mapper.configure(SerializationFeature.WRITE_DATES_AS_TIMESTAMPS, false);
lenientMapper = new ObjectMapper().registerModule(new JavaTimeModule());
lenientMapper.configure(JsonParser.Feature.ALLOW_UNQUOTED_FIELD_NAMES, true);
lenientMapper.configure(JsonParser.Feature.ALLOW_COMMENTS, true);
lenientMapper.configure(JsonParser.Feature.ALLOW_SINGLE_QUOTES, true);
}
@Override
public ObjectMapper getObjectMapper(boolean lenient) {
return lenient ? lenientMapper : mapper;
}
}
and register it in META-INF/services/net.javacrumbs.jsonunit.providers.Jackson2ObjectMapperProvider
.
See this example.
Although the differences are printed out by the assert statement, sometimes you use JsonUnit with other libraries like
Jadler that do not print the differences between documents. In such case, you can switch on the
logging. JsonUnit uses SLF4J. The only thing you need to do is to configure your logging
framework to log net.javacrumbs.jsonunit.difference
on DEBUG level.
If you need better difference reporting, you can implement DifferenceListener
and use it this way
assertThatJson("{\"test\":-1}")
.withDifferenceListener(listener)
.isEqualTo(json);
JsonUnit is trying to cleverly match which JSON library to use. In case you need to change the default behavior, you can
use json-unit.libraries
system property. For example -Djson-unit.libraries=jackson2,gson
or System.setProperty("json-unit.libraries", "jackson2");
. Supported values are gson, json.org, moshi, jackson2
JsonUnit is licensed under Apache 2.0 licence.
Released notes moved to a separate file