Create mock messages from Avro schemas.
Call the generator with the schema:
import generateMsg from '@ovotech/avro-mock-generator'
const schema = {
type: 'record',
fields: [{ name: 'nbChickens', type: 'int' }],
}
console.log(generateData(schema));
// { nbChickens: 25672672 }
All fields will contain randomly generated data that respects their type.
An options
object can optionnaly be provide as the second argument.
Supported Options:
generators
: Ankey
/value
object of generator functions.key
: thetype
(orlogicalType
)value
: should be a generator function(type, context) => value
where -type
: the content of thetype
field in the schema, either astring
for simple type, or the type configuration for complex types -context
: an object with contextual data, including thegenerators
It is possible to override the default generators, and add support for extra types/logicalTypes by providing
pickUnion
: Array of strings to drive which member of union type to choose. Can be the short name of fully namespaced names. When this option is not provided, the first element in the union will be chosen
Use the Seeded
factory to create a version of the generator that will use deterministic randomness.
The factory will return a function with the same signature as the default generator.
Based on the Avro 1.9.0 specification.
- All primitive types
- All logical types
- including custom logicalTypes using the
options
parameter. If alogicalType
is missing a generator, data will be generated matching the underlyingtype
.
- including custom logicalTypes using the
- All complex types
- Note that for
enum
types, the first element of the array will always be chosen.
- Note that for
- Type Alias
Partial support for namespaces. Only union types are namespaced, unconditionally.
All contributions are welcome, just fork the repository then create a PR. Once merged we will release a new version.
Simply create a new release on master, with the tag being the new version (eg: 1.0.0).