An Avro Schema

The Field of a Record

Field Sort ordering

Get the TypeName for an Avro Schema

For primitive types, this is an unqualified name, but for complex types it may be qualified.

Add documentation to a Schema.

If the Schema does not support documentation (i.e, it's not a Record or Enum) this function has no effect.

Add aliases to a Schema.

If the Schema does not support aliases (i.e, it's not a named type) this function has no effect.

Add a logical type to a Schema.

If the Schema does not support a logical type, (e.g., it's a named type) this function has no effect.

Errors which can occur when trying to read Avro with an incompatible Schema.

Display a Schema mismatch error.

Errors which can occur when a Schema is poorly described.

Display a Schema invalid error.

Validates an Avro schema.

Schema's produced using the Codec module or parsed from JSON are typically valid, but there are some ways to create invalid Schemas which may not be well received by other implementations.

This function checks for the following infelicities:

• Unions must not directly contain other unions,
• Unions must not ambiguous (contain more than one schemas with the same name or simple type),
• Enums must not have duplicate values,
• Enums with a default values must include the value in their options,
• Type names and record field names are valid, and
• Named types are not defined more than once.

Other things which are currently not covered by this function:

• Default values are of the correct type.

The JSON parser will not parse schemas with invalid names or values of the wrong type.

If working with multiple Codecs, it can be a good idea to include a test in your test suite applying this function to your Codec schemas.

Turn an Avro schema into its canonical form.

Schema canonical form can be used to determine if two schemas are functionally equivalent for writing Avro values, as it strips documentation and aliases as well as normalises names and rendering.

While canonical form technically refers to a formatted JSON string with no whitespace between terms, this function only transforms a Schema value, and one should compose the following functions if the canonical form string is required.

import Avro
import Avro.Schema as Schema exposing (Schema)
import Json.Encode as Encode


renderCanonical : Schema -> String
renderCanonical s =
    Schema.canonical s
        |> Avro.schemaEncoder
        |> Encode.encode 0