This library is designed specifically for converting Clojure data structures to Avro-compatible Java classes (not Avro binary format!) back and forth in order to be able to use Schema Registry's serializers with Kafka and Clojure.
Note, that this library is not an Avro-wrapper for Clojure: for this purpose you might want to use damballa/abracad or asmyczek/simple-avro
Add this to your favorite build file (project.clj
or build.boot
)
[io.thdr/kfk.avro-bridge "0.1.0-SNAPSHOT"]
- All keys in Clojure maps are keywordized and kebab-cased by default.
- Enums are keywordized and kebab-cased
- When converting to Java objects, all keys and enums are by default snake_cased.
(ns test-avro
(:require [thdr.kfk.avro-bridge.core :as avro]
[camel-snake-kebab.core :as cases]
[cheshire.core :as json]))
(def schema (json/generate-string {:type "record",
:name "user",
:fields [{:name "id" :type "int"}
{:name "full_name" :type "string"}]}))
(def obj (avro/->java (avro/parse-schema schema)
{:id 1 :full-name "John Doe"}))
;; => <GenericRecord ...>, which can be passed to KafkaAvroSerializer, for example
(avro/->clj obj)
;; => {:id 1 :full-name "John Doe"}
;; You can override the default casing when converting to Java:
(def schema (json/generate-string {:type "record",
:name "user",
:fields [{:name "id" :type "int"}
{:name "fullName" :type "string"}]}))
(def obj (avro/->java (avro/parse-schema schema)
{:id 1 :full-name "John Doe"}
{:java-field-fn (comp name cases/->camelCase)}))
;; => #object[org.apache.avro.generic.GenericData$Record 0x13e22477 "{\"id\": 1, \"fullName\": \"John Doe\"}"]
;; And also when transforming to Clojure:
(avro/->clj obj {:clj-field-fn identity})
;; => {"id" 1, "fullName" "John Doe"}
Feel free to open an issue or PR
Copyright © 2016 Theodore Konukhov me@thdr.io
Distributed under the Eclipse Public License either version 1.0 or (at your option) any later version.