Tinkerpop Blueprints Scala

Scala wrapper for tinkerpop blueprints, this library provide more scalastic code when working with graph database supported by blueprints.

Usage

For installation see install section.

• For Scala 2.10 see https://github.com/anvie/blueprints-scala/tree/scala-2.10
• For Scala 2.9 see https://github.com/anvie/blueprints-scala/tree/scala-2.9

More working and complete examples can be found in specs test.

Creating new vertex:

case class Person(name:String, kind:String)

val hercules = Person("hercules", "demigod")

db.save(hercules)

If your class subclassing from DbObject you will have more sweet code like using save() directly:

case class Person(name:String, kind:String) extends DbObject

val hercules = Person("hercules", "demigod").save()

The save() method return a Vertex object, if you want to get class back just:

val herculesObject = hercules.toCC[Person].get

Creating edges:

hercules --> "father" --> jupiter
hercules --> "mother" --> alcmene

Or you can also add multiple edges in one line:

hercules --> "father" --> jupiter --> "father" --> saturn

Creating mutual (both) edges:

jupiter <--> "brother" <--> neptune

Or more complex mutual edges:

jupiter <--> "brother" <--> neptune <--> "brother" <--> pluto

Shorthand property getter and setter:

jupiter.set("kind", "god")

val kind = jupiter.get[String]("kind").get

Getter get return Option instance, so you can handle unexpected return data efficiently using map:

jupiter.get[String]("kind") map { kind =>
	// do with kind here
}

Or getting value by using default value if empty:

jupiter.getOrElse[String]("status", "live")

Easy getting mutual connection, for example jupiter and pluto is brother both has IN and OUT edges, is very easy to get mutual connection using mutual method:

val jupitersBrothers = jupiter.mutual("brother")

Inspect helpers to print vertex list:

jupitersBrothers.printDump("jupiter brothers:", "name")

// will produce output:

jupiter brothers:
 + neptune
 + pluto

Using Gremlin Pipeline like a boss:

val battled = hercules.pipe.out("battled")
battled.printDump("hercules battled:", "name")

// output:

hercules battled:
 + nemean
 + hydra
 + cerberus

Syntactic sugar filtering on Gremlin Pipeline:

// get monster battled with hercules more than 5 times
val monsters = hercules.pipe.outE("battled").wrap.filter { edge =>
   edge.getOrElse[Int]("time", 0).toString.toInt > 5
}

Returning edges from chain by adding < on the last line:

var edge = hercules --> "battled" --> hydra <

edge.set("time", 2)

Using transaction:

transact {
	hercules --> "father" --> jupiter
	hercules --> "mother" --> alcmene
	val edge = hercules --> "battled" --> hydra <

	edge.set("time", 15)
}

Want more attributes that not fit for case class constructor parameter?

Use @Persistent annotation:

case class City(name:String){

    @Persistent var province = "Jakarta" // this attribute will be saved
    @Persistent var country = "Indonesia" // this attribute will be saved

    val code = 123 // this not

}

Support inheritance as well:

abstract class City(name:String) {
    @Persistent var province = ""
}

trait Street {
    @Persistent var street = ""
}

trait Code {
    @Persistent var postalCode = 0
}


case class Jakarta extends City("Jakarta") with Street with Code with DbObject

// usage

val address = Jakarta()
address.province = "Jakarta Barat"  // this will be saved
address.street = "KS. Tubun"        // this will be saved
address.postalCode = 12345          // this will be saved
address.save()

// getting back

db.getVertex(address.getVertex.getId).toCC[Jakarta] map { adr =>
    println(adr.province) // will printed `Jakarta Barat`
    println(adr.postalCode) // will printed `12345`
}

// instantiate to parent class
// this is valid

db.getVertex(address.getVertex.getId).toCC[City] map { adr =>
    println(adr.province) // will printed `Jakarta Barat`
}

// this also valid

db.getVertex(address.getVertex.getId).toCC[Street] map { adr =>
    println(adr.street) // will printed `KS. Tubun`
}

How it work? All you needs is to adding dependency (see install), import it, and summon Graph db instance with implicit:

import com.ansvia.graph.BlueprintsWrapper._

// you can use any graph database that support tinkerpop blueprints
// here is i'm using simple in-memory Tinkergraph db for example.
implicit val db = TinkerGraphFactory.createTinkerGraph()

//... work here ....

Done, now you can using Scalastic sweet syntactic sugar code.

Install

For Scala 2.11 currently only SNAPSHOT releases available:

"Sonatype repo" at "https://oss.sonatype.org/content/repositories/snapshots"

"com.ansvia.graph" %% "blueprints-scala" % "0.1.61-20150416-SNAPSHOT"

License

Apache 2

[] Robin Sy.

Support me via Gittip