Schema Validator

Overview

Schema validator checks if the provided doc node complies with the json schema defined - i.e. if a property that has to be an array is not provided as a string, etc. If the object is not as expected it is possible to throw an error (that combined with catch step can stop the further flow execution) as well as the result of the validation can be stored in te doc node.

This step is useful to validate the integrity of the received data.

For more details please refer to JSON Schema Documentation

You can generate JSON Schema for your input using JSON Schema Tool

Configuration

• Data path - path to the doc node on which to execute the schema validation
• JSON schema(overrides JSON inline schema) - schema definition for the doc node (Requires schema to be set up first in ACE's schemas section. Afterwards the needed schema has to be selected under this selector.)
• JSON Inline schema - schema definition for the doc node
• throw as error - if checked, failed schema validation will throw an error
• target path - doc node on which to store the validation result (empty array if successful, error object if failed)

note

Providing either of JSON Schema or JSON Inline Schema properties is required. If both are provided, JSON schema property has more preference.

Examples

The example flow bellow showcase how a doc node can be validated.

We assume that the flow input looks like this:

{
    "input": {
        "stringProp": "test prop"
    }
}

In the below example, the Schema Validator step is configured to verify that the 'stringProp' property on the doc input node is a required property that exists and its value is a string. The validation results will be stored on the validations node (we have opted out of throwing an error so the result will be appended to the validations node).

So if the input node does not contain 'stringProp' key or if its corresponding value is not string, then validations array will contain validation errors.

The below example uses JSON inline schema property, similarly we can use the JSON schema property defining the schema under schemas and selecting it.

Example flow with only json inline schema provided

    tags:
      - general
    steps:
      - stepType: ajv
        config:
          dataPath: input
          targetPath: validations
          schema:
            title: testSchema
            nodeKey: testSchema
            properties:
              stringProp:
                type: string
            type: object
            required:
              - stringProp
        name: Schema Validator
        description: ""
        condition: ""
    sampleData:
      input:
        stringPropUpdated: test

Result with valid input

Input

{
"input":
  {
  "stringProp":"test"
  }
}

Result

{
  "doc": {
    "input": {
      "stringProp": "test"
    },
    "validations": []
  },
  "errors": [],
  "performance": {
    "steps": [
      {
        "step": "ajv",
        "executionTime": 6
      }
    ],
    "executionTimeOfFlow": 7,
    "timeMetric": "ms"
  }
}

Result with invalid input

Input

{
  "input": {
    "stringPropUpdated": "test"
  }
}

Result

{
  "doc": {
    "input": {
      "stringPropUpdated": "test"
    },
    "validations": [
      {
        "keyword": "required",
        "dataPath": "",
        "schemaPath": "#/required",
        "params": {
          "missingProperty": "stringProp"
        },
        "message": "should have required property 'stringProp'"
      }
    ]
  },
  "errors": [],
  "performance": {
    "steps": [
      {
        "step": "ajv",
        "executionTime": 7
      }
    ],
    "executionTimeOfFlow": 7,
    "timeMetric": "ms"
  }
}

Result with above invalid input, when Throw as error checked

{
  "doc": {
    "input": {
      "stringPropUpdated": "test"
    }
  },
  "errors": [
    {
      "error": "Error in flow 'schemaValidatorTest' : Schema validation error",
      "validationErrors": [
        {
          "keyword": "required",
          "dataPath": "",
          "schemaPath": "#/required",
          "params": {
            "missingProperty": "stringProp"
          },
          "message": "should have required property 'stringProp'"
        }
      ]
    }
  ],
  "performance": {
    "steps": [
      {
        "step": "ajv",
        "executionTime": 9
      }
    ],
    "executionTimeOfFlow": 10,
    "timeMetric": "ms"
  }
}

In the above example, the Schema Validator step is next updated with testSchemaUpdated option which verify that the stringPropUpdated property on the doc input node is a required property that exists and its value is a string.

So if the input node does not contain 'stringPropUpdated' key or if its corresponding value is not string, it validations array will contain validation errors.

Here as both JSON Schema and JSON inline schema properties are provided, the JSON Schema property has more preference.

Example schema

{
  "title": "testSchemaUpdated",
  "nodeKey": "testSchemaUpdated",
  "properties": {
    "stringPropUpdated": {
      "type": "string"
    }
  },
  "type": "object",
  "required": ["stringPropUpdated"]
}

Example flow with both json schema and json inline schema provided(json schema overriding json inline schema)

    tags:
      - general
    steps:
      - stepType: ajv
        config:
          dataPath: input
          targetPath: validations
          schema:
            title: testSchema
            nodeKey: testSchema
            properties:
              stringProp:
                type: string
            type: object
            required:
              - stringProp
          schemaReference: testSchemaUpdated
        name: Schema Validator
        description: ""
        condition: ""
    sampleData:
      input:
        stringPropUpdated: test

Result with valid input

Input

{
  "input": {
    "stringPropUpdated": "test"
  }
}

Result

{
  "doc": {
    "input": {
      "stringPropUpdated": "test"
    },
    "validations": []
  },
  "errors": [],
  "performance": {
    "steps": [
      {
        "step": "ajv",
        "executionTime": 26
      }
    ],
    "executionTimeOfFlow": 28,
    "timeMetric": "ms"
  }
}

Result with invalid input

Input

{
  "input": {
    "stringProp": "test"
  }
}

Result

{
  "doc": {
    "input": {
      "stringProp": "test"
    },
    "validations": [
      {
        "keyword": "required",
        "dataPath": "",
        "schemaPath": "#/required",
        "params": {
          "missingProperty": "stringPropUpdated"
        },
        "message": "should have required property 'stringPropUpdated'"
      }
    ]
  },
  "errors": [],
  "performance": {
    "steps": [
      {
        "step": "ajv",
        "executionTime": 18
      }
    ],
    "executionTimeOfFlow": 18,
    "timeMetric": "ms"
  }
}