The simple Apache Solr client for Scala. This is based on the SolrJ and provides optimal interface for Scala.
Add the following dependency into your build.sbt
to use solr-scala-client.
libraryDependencies += "com.github.takezoe" %% "solr-scala-client" % "0.0.27"
If you want to test SNAPSHOT version, add the following dependency instead of above:
resolvers += "sonatype-oss-snapshot" at "https://oss.sonatype.org/content/repositories/snapshots"
libraryDependencies += "com.github.takezoe" %% "solr-scala-client" % "x.x.x-SNAPSHOT"
This is a simplest example to show usage of solr-scala-client.
import com.github.takezoe.solr.scala._
val client = new SolrClient("https://localhost:8983/solr")
// register
client
.add(Map("id"->"001", "manu" -> "Lenovo", "name" -> "ThinkPad X201s"))
.add(Map("id"->"002", "manu" -> "Lenovo", "name" -> "ThinkPad X220"))
.add(Map("id"->"003", "manu" -> "Lenovo", "name" -> "ThinkPad X121e"))
.commit
// query
val result = client.query("name: %name%")
.fields("id", "manu", "name")
.sortBy("id", Order.asc)
.getResultAsMap(Map("name" -> "ThinkPad"))
result.documents.foreach { doc: Map[String, Any] =>
println("id: " + doc("id"))
println(" manu: " + doc("manu"))
println(" name: " + doc("name"))
}
It's also possible to use the case class as the search result and parameters instead of Map.
// query
val result = client.query("name: %name%")
.fields("id", "manu", "name")
.sortBy("id", Order.asc)
.getResultAs[Product](Param(name = "ThinkPad"))
result.documents.foreach { product: Product =>
println("id: " + product.id)
println(" manu: " + product.manu)
println(" name: " + product.name)
}
Following notations are available to embed variables to the query:
- %VAR_NAME% : place holder to set a single word (parameter would be escaped)
- ?VAR_NAME? : place holder to set an expression (&, | and ! are available in an expression)
-
$VAR_NAME$ : string replacement (parameter would be not escaped)
See examples of parameterized queries and assembled Solr queries.
// %VAR_NAME% (Single keyword)
client.query("name: %name%").getResultAsMap(Map("name" -> "ThinkPad X201s"))
// => name:"ThinkPad X201s"
// $VAR_NAME$ (String replacement)
client.query("name: $name$").getResultAsMap(Map("name" -> "ThinkPad AND X201s"))
// => name:ThinkPad AND X201s
// ?VAR_NAME? (Expression)
client.query("name: ?name?").getResultAsMap(Map("name" -> "ThinkPad & X201s"))
// => name:("ThinkPad" AND "X201s")
Configure the query to return the highlighted content by QueryBuilder#highlight()
.
The highlighted content is set as the "highlight" property to the Map or the case class.
val result = client.query("content: Scala")
// NOTE: unique key field is required.
.fields("id")
// Specify the highlighted field, prefix and postfix (prefix and postfix are optional).
.highlight("content", "<strong>", "</strong>")
.getResultAsMap()
result.documents.foreach { doc: Map[String, Any] =>
println("id: " + doc("id"))
println(doc("highlight")) // highlighted content is set as the "highlight" property
}
solr-scala-client expects that the unique key is "id". If your schema has the different field as the unique key, you can specify the unique key name as following:
val result = client.query("content: Scala")
.id("documentId") // Specify the unique key name
.fields("documentId")
.highlight("content", "<strong>", "</strong>")
.getResultAsMap()
solr-scala-client has also asynchronous API based on AsyncHttpCleint.
val client = new AsyncSolrClient("https://localhost:8983/solr")
// Register
client
.register(Map("id" -> "005", "name" -> "ThinkPad X1 Carbon", "manu" -> "Lenovo"))
.onComplete{
case Success(x) => println("registered!")
case Failure(t) => t.printStackTrace()
}
// Query
client.query("name:%name%")
.fields("id", "manu", "name")
.facetFields("manu")
.sortBy("id", Order.asc)
.getResultAsMap(Map("name" -> "ThinkPad X201s"))
.onComplete {
case Success(x) => println(x)
case Failure(t) => t.printStackTrace()
}
See more example at AsyncSolrClientSample.scala.
- Add sortBy overload for multiple sorting fields
- Fix scala-parser-combinators dependency
- Support Scala 3
- Include facetpivot in mapquery result
- Facet Pivot Fields support
- Update dependent libraries
- Spatial parameters support
- Scala 2.13.0 support
- Scala 2.13.0-M5 support
- Support specifying collection with transaction
- Add support for grouping and qTime in the response
- Allow batch processing with a specific collection
- Fix response leaking bug
- Upgrade to SolrJ-7.1.0
- Switch backend to OkHttp from async-http-client
- Allow specifying collection name when building query
- Add implementation of CloudSolrClient with and without authentication
- Upgrade Scala and async-http-client
- Scala 2.12 support and library updating
- Small refactoring
- Upgrade to SolrJ-6.1.0
- Change group id and package name to
com.github.takezoe
- Publish to the Maven central repository
- Add
QueryBuilderBase#fq()
- Add
QueryBuilderBase#setRequestHandler()
- Add date facet
- Support for streaming results
- Add
SolrClient#shutdown()
QueryBuilder
became immutable- Upgrade solrj version to 4.5.1
- Fix escaping in string literal.
- Bug fix
- Added recommendation search support.
- Added
rollback
andwithTransaction
toSolrScalaClient
. - Added Asynchronous API.
- Add build for Scala 2.10
- Upgrade to SolrJ 4.2.0
- Support highlighting
- Fixed some
ExpressionParser
bugs.
ExpressionParser
became pluggable and addedGoogleExpressionParser
as an optional implementation ofExpressionParser
.- Converts the full-width space to the half-width space in the expression before calling ExpressionParser.
- Introduced the SolrServer factory.
Auth.basic
moved toSolrServerFactory.basicAuth
andSolrServerFactory.dummy
for unit testing.
- Expanding expression to the Solr query by
?VAR_NAME?
inSolrClient#query()
. - Bug fix
- Added case class support in update operations.
- Added
commit()
method toSolrClient
.
- Added initializer which configures
SolrClient
. - Added basic authentication support as initializer.
- Added facet search support.
- Added case class support as query results and query parameters.
- Initial public release.