Does JSON Schema have a switch like structure? - jsonschema

Consider this following example,
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "Animal",
"type": "object",
"properties": {
"type": {
"type": "string",
"description": "Type of animal."
},
"data": {
"$ref": "#/$definations/cat"
}
},
"$definations":{
"cat" : {
"properties" : {
"meow" : {
"type": "string"
}
}
}
},
"required": ["type"]
}
and the correct JSON is ,
{
"type" : "cat",
"data" : {
"meow" : "OK"
}
}
Now I am having enum of Animals, and the data ref will vary based on type of Animal.
I have have tried if else but it seems not efficient as the condition will keep on growing.
Also used anyOf but how will I make sure that meow will always belong to animal type cat and not dog.
Can we have something like,
cat : { "$ref" : "#/$definations/cat" },
dog : { "$ref" : "#/$definations/dog" }
EDIT : Or dynamic value in ref like #/$definations/{type-value} ?
Thanks in advance.

I have have tried if else but it seems not efficient as the condition
will keep on growing.
Can we have something like...
No. JSON Schema (2019-09 and previous) doesn't have a "switch".
You'll need to use allOf to create multiple if then conditions.

After #Relequestual's response and some more digging I found there is no such way. At least in this version, fingers crossed for the future releases.
Here is my solution, feel free to suggest improvements.
{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "Animal",
"type": "object",
"properties": {
"type": {
"type": "string",
"description": "Type of animal."
}
},
"$definations": {
"cat": {
"properties": {
"meow": {
"type": "string"
}
},
"required": [
"meow"
]
},
"dog": {
"properties": {
"bhow": {
"type": "string"
}
},
"required": [
"bhow"
]
}
},
"oneOf": [
{
"properties": {
"type": {
"const": "cat"
},
"data": {
"$ref": "#/$definations/cat"
}
}
},
{
"properties": {
"type": {
"const": "dog"
},
"data": {
"$ref": "#/$definations/dog"
}
}
}
],
"required": [
"type"
]
}

Related

How to improve JSON Schema validation results when using "anyOf"

We want to validate an array of elements with a JSON Schema Validator (https://github.com/networknt/json-schema-validator if that makes any difference).
Unfortunately we get quite ambiguous error messages if the validation fails and I am wondering if there is a way to improve this.
To make it clearer, I have created a small sample:
Each element in the array has a property "pet_type" which we can be sure to be there all the time (yeah I know that is discussable, but for the sake of argument let's ignore that for now). Each element has then some other properties.
This is the json schema right now.
Envelope (Basically with an array of elements)
{
"title": "Envelope",
"type": "object",
"properties": {
"pets": {
"type": "array",
"items": {
"anyOf": [
{
"$ref": "./Cat.json"
},
{
"$ref": "./Dog.json"
}
]
}
}
}
}
Cat (one of the elements in the array)
{
"title": "Cat",
"allOf": [
{
"$ref": "Pet.json"
},
{
"type": "object",
"properties": {
"hunts": {
"type": "boolean"
},
"age": {
"type": "integer"
},
"pet_type": {
"type": "string",
"enum": [
"Cat"
],
"pattern": "Cat"
}
},
"required": [
"pet_type",
"age"
]
}
]
}
Dog:
{
"title": "Dog",
"allOf": [
{
"$ref": "Pet.json"
},
{
"type": "object",
"properties": {
"bark": {
"type": "boolean"
},
"breed": {
"type": "string"
},
"pet_type": {
"type": "string",
"enum": [
"Dog"
],
"pattern": "Dog"
}
},
"required": [
"bark"
]
}
]
}
And the Pet.json, which specifies that "pet_type" is required on each of the elements.
{
"title": "Pet",
"type": "object",
"discriminator": {
"propertyName": "pet_type"
},
"properties": {
"pet_type": {
"type": "string",
"enum": [
"Cat",
"Dog"
]
}
},
"required": [
"pet_type"
]
}
The problem I have is that the error messages get very long, because the element is validated against all of the constraints, without narrowing it down a bit.
Let's say the JSON to validate looks like this:
{
"pets": [
{
"pet_type": "Cat",
"hunts": true
}
]
}
"pet_type" is set to "Cat", so for a human it is "clear" that it should only validate against the Cat JSON Schema and then show that "age" is missing.
What really happens is that there are 4 violations:
age is missing (from the Cat JSON Schema)
bark is missing (from the Dog JSON Schema)
pet_type is violating the "Dog" enum contstraint (from the Dog JSON Schema)
pet_type is violating the "Dog" regex pattern (from the Dog JSON Schema)
Note: I have added the enum/regex pattern in an act of trying to change the behavior, but it did not work.
I understand from a technical standpoint why it behaves like it does, I just want to know if it is possible to tell the validator somehow to first narrow down the validation based on pet_type and then keep validating?
What I tried to get it working:
I have tried to set the pet_type to string with an enum "Cat" and "Dog" and added one of those values to the Cat/Dog JSON Schema respectively in an attempt to make it clear which event has which pet_type.
I have tried to remove the Pet.json Schema completely, but that did not change anything.
I have tried to use if/then, but somehow this gets rid of all validation errors:
{
"title": "Envelope",
"type": "object",
"properties": {
"pets": {
"type": "array",
"items": {
"anyOf": [
{
"if": {
"properties": {
"pet_type": {
"const": "Cat"
}
}
},
"then": {
"$ref": "./Cat.json"
}
},
{
"if": {
"properties": {
"pet_type": {
"const": "Dog"
}
}
},
"then": {
"$ref": "./Dog.json"
}
}
]
}
}
}
}
My sample I tried to validate gives no errors anymore:
{
"pets": [
{
"pet_type": "Cat",
"hunts": true
}
]
}
Turns out the solution with if/then was correct, and the json-schema-validator actually detects the problems, but has some code that should improve the error messages, but gets rid of the validation errors instead.
I will bring the problem to their attention.
For now I have found a workaround to use nested if/then/else statements:
{
"title": "Envelope",
"type": "object",
"properties": {
"pets": {
"type": "array",
"items": {
"if": {
"properties": {
"pet_type": {
"const": "Cat"
}
}
},
"then": {
"$ref": "./Cat.json"
},
"else": {
"if": {
"properties": {
"pet_type": {
"const": "Dog"
}
}
},
"then": {
"$ref": "./Dog.json"
},
"else": {
"properties": {
"pet_type": {
"type": "string",
"enum": [
"Cat",
"Dog"
]
}
}
}
}
}
}
},
"additionalProperties": false
}

JSON Schema object types not working for $ref definitions

I am using JSON Schema Draft-07. I am having a JSON Schema like below that works as expected
{
"tests": {
"type": "object",
"required": [
"name",
"desc"
],
"properties": {
"name": {
"type": "string"
},
"desc": {
"type": "string"
}
}
}
}
But if I refer the same type using definition, siblings properties\rules doesn't take affect , for e.g.
JSON Schema Store
{
"my-schema.json": {
"$id": "my-schema.json",
"$schema": "http://json-schema.org/draft-07/schema",
"definitions": {
"objectParameter": {
"type": "object"
}
},
"properties": {
"tests": {
"$ref": "#/definitions/objectParameter",
"properties": {
"desc": {
"type": "string"
},
"name": {
"type": "string"
}
},
"required": [
"name",
"desc"
]
}
},
"required": [
"tests"
],
"type": "object"
}
}
Not sure what's the difference $ref is bringing here.
Validation Data : "tests": {"namer": "12", "desc": "12"} or "tests": {} should fail but getting passed .
"doesn't work" isn't very specific, but is it possible you are using an implementation that follows draft specification version 7 or earlier, in which the presence of a $ref keyword nullifies the effect of any other sibling keywords? If so, you can resolve the situation by wrapping an "allOf" around the subschemas:
"allOf": [
{ "$ref": ... },
{
"required": ...,
"properties": { ... }
}
]

JSON Schema switch object properties based on enum

I've been struggling with "switch" in JSON Schema. Went through couple of GitHub and SO discussions on this topic but haven't find solution.
My intention is to vary "payload" object properties based on "id" enum that will have 30 different mappings ("payload" definitions per enum "id").
For example first message json object will have amount and other properties but for the demo purpose let's go only with one property (amout):
{
"message": {
"id": 1,
"correlationId": "a0011e83-280e-4085-b0f1-691059aaae61",
"payload": {
"amount": 100
}
}
}
And second json:
{
"message": {
"id": 2,
"correlationId": "a0011e83-280e-4085-b0f1-691059aaae61",
"payload": {
"code": "xyz"
}
}
}
Is there a way to build JSON Schema (draft 7 or any other) in this manner?
What you're asking for is a fairly common requirement. Using oneOf/anyOf should get you where you want.
In those cases where the alternatives are mutually exclusive (due to the different "id" values), I'm in favour of anyOf to allow Schema Validator to stop checking when encountering the first matching subschema – whereas oneOf implies that all other alternatives must not match, e.g. in case of "id": 1 a validator would only have to check against the first subschema in an anyOf to indicate that it is valid while for oneOf it'd have to check against the other 29 to ensure that those aren't also valid. But you may find oneOf more expressive for human consumers of your schema.
For your particular scenario, I'd imagine something along the lines of the following schema:
{
"type": "object",
"required": ["message"],
"properties": {
"message": {
"type": "object",
"required": ["id", "correlationId", "payload"],
"properties": {
"id": { "enum": [1, 2, 3] },
"correlationId": { "type": "string" },
"payload": { "type": "object" }
},
"anyOf": [
{
"properties": {
"id": { "const": 1 },
"payload": { "$ref": "#/definitions/payload1" }
}
},
{
"properties": {
"id": { "const": 2 },
"payload": { "$ref": "#/definitions/payload2" }
}
},
{
"properties": {
"id": { "const": 3 },
"payload": { "$ref": "#/definitions/payload3" }
}
},
]
}
},
"definitions": {
"payload1": {
"type": "object",
"required": ["amount"],
"properties": {
"amount": { "type": "integer" }
}
},
"payload2": {
"type": "object",
"required": ["code"],
"properties": {
"code": { "type": "string" }
}
},
"payload3": {
"type": "object",
"required": ["foo"],
"properties": {
"foo": { "type": "string" }
}
}
}
}

How to extent an object with 1 parameter in json-schema

I wrote one JSON schema before, but now, as I am trying to make it a bit more advanced I get stuck.
(I am open to 'good practice' tips in the comments)
(Is the $id optional? should I remove it for simplicity in the example code?)
Goal:
I am trying to make a schema with an object definition (example_obj) that is recursively used. This object may only have 1 argument (or or and or value). But in the root of the json, I want to add 1 additional property.
json-schema
{
"definitions": {
"example_obj": {
"$id": "#/definitions/example_obj",
"type": "object",
"maxProperties": 1,
"properties": {
"or": {
"$id": "#/definitions/example_obj/properties/or",
"type": "array",
"items": {
"$id": "#/definitions/example_obj/properties/or/items",
"$ref": "#/definitions/example_obj"
}
},
"and": {
"$id": "#/definitions/example_obj/properties/and",
"type": "array",
"items": {
"$id": "#/definitions/example_obj/properties/and/items",
"$ref": "#/definitions/example_obj"
}
},
"value": {
"$id": "#/definitions/example_obj/properties/value",
"type": "string"
}
}
}
},
"type": "object",
"title": "The Root Schema",
"required": ["filter_version"],
"allOf": [
{
"$ref": "#/definitions/example_obj"
},
{
"properties": {
"filter_version": {
"$id": "#/properties/filter_version",
"type": "string",
"pattern": "^([0-9]+\\.[0-9]+)$"
}
}
}
]
}
json which I want to pass validation:
{
"filter_version": "1.0",
"or": [
{
"and": [
{
"value": "subject"
}
]
},
{
"or": [
{
"value": "another subject"
}
]
}
]
}
Issue:
When I try to extend example_obj for the root definition it seems to fail because the example_obj object does not allow more then 1 property by design.
In other words, it appears that every check for the number of argument that I add to example_obj is also performed on the additional property (i.e. filter_version).
Does anyone know where to place this check for 'exactly 1 argument' so that it is not evaluated on the root object?
Attempts:
I tried working with different ways of determining the requirements of example_obj, but with no success. Like with replacing "maxProperties": 1 with:
"oneOf": [
{
"required": [
"or"
]
},
{
"required": [
"and"
]
},
{
"required": [
"where"
]
},
{
"required": [
"where not"
]
}
],
Thanks in advance for any help!!
Checking my schema with the online schema validator.
(In the end I need to validate it in Python, in case it matters)
You can use oneOf instead of maxProperties to get around this.
{
"type": "object",
"properties": {
"filter_version": {
"type": "string",
"pattern": "^([0-9]+\\.[0-9]+)$"
}
},
"required": ["filter_version"],
"allOf": [{ "$ref": "#/definitions/example_obj" }],
"definitions": {
"example_obj": {
"type": "object",
"properties": {
"or": { "$ref": "#/definitions/example-obj-array" },
"and": { "$ref": "#/definitions/example-obj-array" },
"value": { "type": "string" }
},
"oneOf": [
{ "required": ["or"] },
{ "required": ["and"] },
{ "required": ["value"] }
]
},
"example-obj-array": {
"type": "array",
"items": { "$ref": "#/definitions/example_obj" }
}
}
}
P.S. You are using $id wrong. I know there is a tool out there that generates schemas like this and causes this confusion. The way $id is used here is a no-op. It doesn't hurt, but it doesn't do anything other than bloating your schema.

JsonSchema is not validated with oneOf

Need help to find the bug with this schema. It has oneOf operator.
Schema is here :
`{
"type": "object",
"required": [
"type",
"body"
],
"properties": {
"type": {
"description": "type of the document to post",
"type": "string",
"enum": [
"123",
"456"
]
},
"body": {
"type": "object",
"description": "body",
"oneOf": [{
"$ref": "#/definitions/abc",
"$ref": "#/definitions/def"
}]
}
},
"definitions": {
"abc": {
"type": "array",
"description": "abc",
"properties" : {
"name" : { "type" : "string" }
}
},
"def": {
"type": "array",
"description": "users","properties" : {
"name" : { "type" : "string" }
}
}
}
}`
My Json is this :
`{
"type": "123",
"body": {
"abc": [{
"name": "test"
}]
}
}`
It does not validate with tv4 and I also tried this online tool. It works without oneOf operator. Otherwise it does not validate it any tool.
Edit :
After reading the answers I modified the schema. New schema is :
{
"type": "object",
"properties": {
"type": {
"description": "type of the document to post",
"type": "string",
},
"body": {
"type": "object",
"description": "body",
"properties": {
"customers": {
"type": "array"
}
},
"anyOf": [
{
"title": "customers prop",
"properties": {
"customers": {
"type": "array",
"description": "customers",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string"
}
}
}
}
}
}
]
}
}
}
And json is here
{
"type": "customer",
"body": {
"none": [
{
"name": "test"
}
]
}
}
But it validates. I want to enforce one of "customers" or "users" in the body. To test I have removed users from the body.
Pl help.
The issue is that the data is passing both of your sub-schemas. oneOf means "match exactly one" - if you want "match at least one", then use anyOf.
In fact, both of your sub-schemas will pass all data. The reason is that properties is ignored when dealing with arrays.
What you presumably wanted to do instead is specify properties for the items in the array. For this, you need the items keyword:
"definitions": {
"abc": {
"type": "array",
"items": {
"type": "object",
"properties" : {
"name" : { "type" : "string" }
}
}
}
}
(You'll also need to add some distinct constraints - at the moment, both the "abc" and "def" definitions are identical apart from description, which makes the oneOf impossible because it will always match both or neither.)
Since you have the type at root level, you probably want the oneOf statement to check that an object with type "customer" has customers in the body (even though I would suggest skipping the body and placing customers and users directly in root object).
This works with your example, will require that an object with type "customer" has a body with "customers", and to clarify the matching, I let customer have the property "name" while the user has "username":
{
"type": "object",
"properties": {
"type": { "type": "string" },
"body": {
"type": "object",
"properties": {
"customers": {
"type": "array",
"items": { "$ref": "#/definitions/customer" }
},
"users": {
"type": "array",
"items": { "$ref": "#/definitions/user" }
}
}
}
},
"definitions": {
"customer": {
"type": "object",
"properties": { "name": { "type": "string" } },
"required": [ "name" ]
},
"user": {
"type": "object",
"properties": { "username": { "type": "string" } },
"required": [ "username" ]
}
},
"oneOf": [
{
"properties": {
"type": {
"pattern": "customer"
},
"body": {
"required": [ "customers" ]
}
}
},
{
"properties": {
"type": {
"pattern": "user"
},
"body": {
"required": [ "users" ]
}
}
}
]
}
When using "type": "array" then the item type is defined in the "items" property not "properties" property... Also both types in oneOf are same, but only one must match.
Try
...
"definitions": {
"abc": {
"type": "array",
"description": "abc",
"items" : {
"name" : { "type" : "string" }
}
},
"def": {
"type": "array",
"description": "users",
"items" : {
"username" : { "type" : "string" }
}
}
}