Skip to content

Custom assertion to check whether a JSON string fuzzily matches a pattern for JVM languages.

License

Notifications You must be signed in to change notification settings

orangain/json-fuzzy-match

Repository files navigation

json-fuzzy-match Maven Central Version Java CI

json-fuzzy-match provides assertion to check whether a JSON string fuzzily matches a pattern for JVM languages. This is useful when you test JSON response including dynamic or generated value.

For example, think about testing the following JSON response.

{
    "id": "2c0a9fd7-be2c-4bc2-b134-acc3fa13d400", // Generated UUID
    "title": "Example Book",
    "price": "9.99",
    "currency": "USD",
    "amount": 10,
    "timestamp": "2019-09-25T13:34:17Z" // Dynamic timestamp
}

Instead of writing:

val tree = ObjectMapper().readTree(response.content)
assertThat(tree.get("id").textValue()).matches("[0-9a-z-]+")
assertThat(tree.get("title").textValue()).isEqualTo("Example Book")
assertThat(tree.get("price").textValue()).isEqualTo("9.99")
assertThat(tree.get("currency").textValue()).isEqualTo("USD")
assertThat(tree.get("amount").numberValue()).isEqualTo(10)
assertThat(tree.get("timestamp").isTextual).isTrue()

you can simply write using json-fuzzy-match:

// language=JSON
JsonStringAssert.assertThat(response.content).jsonMatches("""
    {
      "id": "#uuid",
      "title": "Example Book",
      "price": "9.99",
      "currency": "USD",
      "amount": 10,
      "timestamp": "#datetime"
    }
""".trimIndent())

It is recommended to use json-fuzzy-match with languages which have multi-line string literal such as Kotlin, Scala, Java 13+ and Groovy. Sample codes in this README are written in Kotlin.

Install

json-fuzzy-match is available on Maven Central.

Gradle Kotlin DSL

dependencies {
  testImplementation("io.github.orangain:json-fuzzy-match:0.7.0")
}

Maven

<dependencies>
  <dependency>
    <groupId>io.github.orangain</groupId>
    <artifactId>json-fuzzy-match</artifactId>
    <version>0.7.0</version>
  </dependency>
</dependencies>

Usage

json-fuzzy-match provides both AssertJ-style and JUnit-style assertions.

// AssertJ-style
import io.github.orangain.jsonmatch.JsonStringAssert
...
JsonStringAssert.assertThat("""{ "foo": "bar" }""").jsonMatches("""{ "foo": "#notnull" }""")
// JUnit-style
import io.github.orangain.jsonmatch.JsonMatch.assertJsonMatches
...
assertJsonMatches("""{ "foo": "bar" }""", """{ "foo": "#notnull" }""")

In the above examples, the second argument patternJson contains #notnull marker. The assertion means that value of foo field must exist and not null. There are several markers as followings:

Marker Description
#ignore Skip comparison for this field even if the data element or JSON key is present
#null Expects actual value to be null, and the data element or JSON key must be present
#notnull Expects actual value to be not-null
#present Actual value can be any type or even null, but the key must be present
#notpresent Expects the key to be not present at all
#array Expects actual value to be a JSON array
#object Expects actual value to be a JSON object
#boolean Expects actual value to be a boolean true or false
#number Expects actual value to be a number
#string Expects actual value to be a string
#uuid Expects actual (string) value to conform to the UUID format
#date Expects actual (string) value to conform to the local date in the ISO 8601 extended format
#datetime Expects actual (string) value to conform to the datetime with timezone in the ISO 8601 extended format
#regex STR Expects actual (string) value to match the regular-expression 'STR' (see examples below)
#[NUM] EXPR Advanced array marker. When NUM is provided, array must has length just NUM. When EXPR is provided, array's element must match the pattern 'EXPR' (see examples below)

Optional marker

You can use double hash ## to mark a field as optional. For example, ##string means that the field can be a string, null or not present.

Examples

Pattern {} { "a": null } { "a": "abc" }
{ "a": "#ignore" } ✅ match ✅ match ✅ match
{ "a": "#null" } ❌ not match ✅ match ❌ not match
{ "a": "#notnull" } ❌ not match ❌ not match ✅ match
{ "a": "#present" } ❌ not match ✅ match ✅ match
{ "a": "#notpresent" } ✅ match ❌ not match ❌ not match

Date marker

  • { "createdOn": "2020-07-23" } matches the pattern { "createdOn": "#date" }
  • { "createdOn": "2020/07/23" } does not match the pattern { "createdOn": "#date" }

Datetime marker

  • { "createdAt": "2020-07-23T14:56:11+09:00" } matches the pattern { "createdAt": "#datetime" }
  • { "createdAt": "2020-07-23T05:56:11Z" } matches the pattern { "createdAt": "#datetime" }
  • { "createdAt": "2020-07-23T05:56:11" } does not match the pattern { "createdAt": "#datetime" }

Regex marker

  • { "id": "abc" } matches the pattern { "id": "#regex [a-z]+" }
  • { "id": "123" } does not match the pattern { "id": "#regex [a-z]+" }

Advanced array marker

  • { "tags": ["awesome", "shop"] } matches the following patterns:
    • { "tags": "#[]" }
    • { "tags": "#[2]" }
    • { "tags": "#[] #string" }
    • { "tags": "#[2] #string" }

License

MIT License. See LICENSE.

Acknowledgement

I'm very grateful for the Karate and its authors. The idea of the marker is heavily inspired by the Karate's wonderful fuzzy matching feature. Though json-fuzzy-match does not depend on Karate now, the first version of this library only provided a thin wrapper of Karate's feature. Without it, I was not able to develop this library so quickly.

About

Custom assertion to check whether a JSON string fuzzily matches a pattern for JVM languages.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published