v1.1.0

[miyaflowercat]

New Features

1. Property Type Validation:

• Problem: Kraken does not validate the data types of properties received in Sonata requests against the types defined in the mapping template, leading to potential data inconsistencies and downstream errors.
• Solution: Implement strict type validation.
  • Mapping Template: The mapping template must clearly define the expected data type for each property (e.g., string, number).
  • Validation Logic: Kraken should perform type checking based on the mapping template.
  • Error Handling: If a type mismatch occurs, Kraken will:
    • Return a pre-defined, standardized error code.
    • Include a clear error message specifying:
      • The property name.
      • The expected data type.
      • The received data type.
    • Example Error Message: "Invalid type for property 'age'. Expected integer, but received string."
  • Example: If the mapping template specifies age: { type: "integer" }, and the Sonata request sends age: "twenty", Kraken should reject the request with the appropriate error.

2. Allowed Values Validation (Enumeration/Whitelist Validation):

• Problem: Kraken doesn't allow users to define a limited set of allowed values for specific properties, leading to potential data integrity issues.
• Solution: Implement value validation using an enumeration or whitelist approach.
  • Mapping Template: Allow users to define an array of allowed values within the property definition in the mapping template.
    • Example for numeric values: bandwidth: { type: "integer", enum: [1000, 10000] }
  • Validation Logic: Kraken should check if the received value exists within the defined enum array.
  • Error Handling: If the received value is not in the allowed set, Kraken will:
    • Return a pre-defined error code.
    • Include a clear error message specifying:
      • The property name.
      • The received value.
      • The allowed values.
  • Example: If the mapping template specifies bandwidth: { type: "integer", enum: [1000, 10000] }, and the Sonata request sends bandwidth: "10", or bandwidth: 5000, Kraken should reject the request.

3. Mandatory Property Validation (Null/Missing Checks):

• Problem: Mandatory properties defined in the mapping template can be null or missing in Sonata requests, causing issues downstream.
• Solution: Implement checks for mandatory properties.
  • Mapping Template: Use a required field (or similar mechanism in your schema) in the mapping template to define mandatory properties.
    • Example: name: { type: "string", required: true }
  • Validation Logic: Kraken should check:
    • If the property is present in the Sonata request.
    • If the property's value is not null (or equivalent for your data structure).
  • Error Handling: If a mandatory property is missing or null, Kraken will:
    • Return a pre-defined error code.
    • Include a clear error message specifying the missing/invalid property.
    • Example Error Message (Missing): "Missing required property 'name'."
    • Example Error Message (Null): "Property 'name' cannot be null."
  • Example: If the mapping template specifies name: { type: "string", required: true }, and the Sonata request either doesn't include name or includes name: null, Kraken should reject the request.

By implementing these improvements, Kraken will be much more effective at validating Sonata requests, ensuring data quality and preventing downstream errors.

Full Changelog: v1.0.0...v1.1.0

---

This discussion was created from the release v1.1.0.