bizreach / elastic-scala-httpclient

Scala client and code generator for Elasticsearch


elastic-scala-httpclient Build Status

Elasticsearch HTTP client for Scala with code generator.

Client version Elasticsearch Scala version
3.0.0 - 5.2.x - 2.11 / 2.12
2.0.4 - 2.0.6 2.3.5 2.12
2.0.0 - 2.0.3 2.3.5 2.11
1.0.6 1.7.3 2.11
1.0.0 - 1.0.5 1.1.0 2.11

How to use

Add a following dependency into your build.sbt at first.

libraryDependencies += "" %% "elastic-scala-httpclient" % "3.1.0"

You can access Elasticsearch via HTTP Rest API as following:

case class Tweet(name: String, message: String)
case class TweetMessage(message: String)


ESClient.using("http://localhost:9200"){ client =>
  val config = "twitter" / "tweet"

  // Insert
  client.insert(config, Tweet("takezoe", "Hello World!!"))
  client.insertJson(config, """{name: "takezoe", message: "Hello World!!"}""")

  // Update
  client.update(config, "1", Tweet("takezoe", "Hello Scala!!"))
  client.updateJson(config, "1", """{name: "takezoe", message: "Hello World!!"}""")

  // Update partially
  client.updatePartially(config, "1", TweetMessage("Hello Japan!!"))
  client.updatePartiallyJson(config, "1", """{name: "takechan" }""")

  // Delete
  client.delete(config, "1")

  // Find one document
  val tweet: Option[(String, Tweet)] = client.find[Tweet](config){ builder =>
    builder.query(termQuery("_id", "1"))

  // Search documents
  val list: List[ESSearchResult] = client.list[Tweet](config){ builder =>
    builder.query(termQuery("name", "takezoe"))

If you have to recycle ESClient instance, you can manage lyfecycle of ESClient manually.

// Call this method once before using ESClient

val client = ESClient(""http://localhost:9200"")
val config = "twitter" / "tweet"

client.insert(config, Tweet("takezoe", "Hello World!!"))

// Call this method before shutting down application

AsyncESClient that is an asynchrnous version of ESClient is also available. All methods of AsyncESClient returns Future.

elastic-scala-httpclient is a wrapper of Elasticsearch Java API. Therefore see its document to know details, especially how to build query.

Addtional requirements

Some methods of ESClient and AsyncESClient need Elasticsearch plug-ins. You have to install following plug-ins into Elasticsearch to use these methods:

Method Elasticsearch plug-in
searchByTemplate elasticsearch-sstmpl plug-in
listByTemplate elasticsearch-sstmpl plug-in
countByTemplate elasticsearch-sstmpl plug-in
countByTemplateAsInt elasticsearch-sstmpl plug-in

Furthermore you have to indicate to enable these methods as follows (In default, these methods throws IllegalStateException):

// Enable xxxxByTemplate methods
  scriptTemplateIsAvailable = true){ client =>

Code Generator

elastic-scala-codegen can generate source code from Elasticsearch schema json file.

At first, add following setting into project/plugins.sbt:

addSbtPlugin("" % "elastic-scala-codegen" % "3.1.0")

Then put Elasticsearch schema json file as PROJECT_ROOT/schema.json and execute sbt es-codegen. Source code will be generated into src/main/scala/models.

You can configure generation settings in PROJECT_ROOT/es-codegen.json. Here is a configuration example:

  "outputDir": "sec/main/scala",
  "dateType": "java8",
  "mappings": [
	  "path": "schemas/book.json",
	  "packageName": "",
	  "className": "Book",
	  "arrayProperties": [
	  "ignoreProperties": [
  "typeMappings": {
    "minhash": "String"

See ESCodegenConfig.scala to know configuration details.