#18 progress: changed analyzers to be functions so that processing ca…

…n be done w/ or w/o Actors.

src/main/scala/chalk/slab/AnalysisEngine.scala

@@ -6,41 +6,68 @@ import akka.util.Timeout
 import scala.collection.mutable.ListBuffer
 import scala.concurrent.duration._
 
-trait AnalysisComponent[X,Y,-Z<:Y,+W<:Y] extends Actor with ActorLogging {
-
+/**
+  * An analysis function that takes a Slab with declared annotation types in it and outputs
+  * a new Slab with additional annotations of a new type.
+  *
+  * Documentation for the type variables:
+  *   C = Content type
+  *   B = Base annonation type
+  *   I = Input annotation type
+  *   O = Output annotation type
+  */ 
+trait AnalysisFunction[C,B,-I<:B,+O<:B] extends (Slab[C,B,I] => Slab[C,B,B with O])
+
+/**
+  * An actor that mixes-in an AnalysisFunction and hands Slabs contained in Process messages over
+  * to the function.
+  */
+trait AnalysisComponent[C,B,-I<:B,+O<:B] extends Actor with ActorLogging with AnalysisFunction[C,B,I,O] {
   import AnalysisComponent._
-
-  def process(slab: Slab[X,Y,Z]): Slab[X,Y,Y with W]
-
   def receive = {
-    case Process(slab) =>
-      sender ! process(slab.asInstanceOf[Slab[X,Y,Z]])
+    case Process(slab) => sender ! apply(slab.asInstanceOf[Slab[C,B,I]])
   }
-
 }
 
+/**
+  * Companion object, e.g. to hold messages that can be processed by an AnalysisComponent actor.
+  */
 object AnalysisComponent {
-  case class Process[X,Y,Z<:Y](slab: Slab[X,Y,Z])
+  case class Process[C,B,I<:B](slab: Slab[C,B,I])
 }
 
-class SentenceSegmenterActor[AnnotationTypes <: StringAnnotation]
-    extends AnalysisComponent[String, StringAnnotation, AnnotationTypes, Sentence] {
-
-  def process(slab: Slab[String, StringAnnotation, AnnotationTypes]) = 
+/**
+  * A simple regex sentence segmenter.
+  */
+trait SentenceSegmenter extends AnalysisFunction[String, StringAnnotation, StringAnnotation, Sentence] {
+  def apply(slab: Slab[String, StringAnnotation, StringAnnotation]) =
     slab ++ "[^\\s.!?]+[^.!?]+[.!?]".r.findAllMatchIn(slab.content).map(m => Sentence(m.start, m.end))
 }
 
-class TokenizerActor[AnnotationTypes <: Sentence]
-    extends AnalysisComponent[String, StringAnnotation, AnnotationTypes, Token] {
-
-  def process(slab: Slab[String, StringAnnotation, AnnotationTypes]) =
+/**
+  * An actor that uses SentenceSegmenter.
+  */
+class SentenceSegmenterActor extends SentenceSegmenter
+    with AnalysisComponent[String,StringAnnotation,StringAnnotation,Sentence]
+
+/**
+  * A simple regex tokenizer.
+  */
+trait Tokenizer extends AnalysisFunction[String, StringAnnotation, Sentence, Token] {
+  def apply(slab: Slab[String, StringAnnotation, Sentence]) =
     slab ++ slab.iterator[Sentence].flatMap(sentence =>
       "\\p{L}+|\\p{P}+|\\p{N}+".r.findAllMatchIn(sentence.in(slab).content).map(m =>
         Token(sentence.begin + m.start, sentence.begin + m.end)))
-
 }
 
+/**
+  * An actor that uses Tokenizer.
+  */
+class TokenizerActor extends AnalysisComponent[String, StringAnnotation, Sentence, Token] with Tokenizer
 
+/**
+  * Example application doing actor based Slab processing.
+  */
 object AnalysisEngine {
 
   import AnalysisComponent._
@@ -56,8 +83,8 @@ object AnalysisEngine {
     implicit val timeout = Timeout(10 seconds)    
 
 
-    val sentenceSegmenter = system.actorOf(Props[SentenceSegmenterActor[StringAnnotation]])
-    val tokenizer = system.actorOf(Props[TokenizerActor[Sentence]])
+    val sentenceSegmenter = system.actorOf(Props[SentenceSegmenterActor])
+    val tokenizer = system.actorOf(Props[TokenizerActor])
 
     for {
       slab1 <- (sentenceSegmenter ? Process(slab)).mapTo[Slab[String,StringAnnotation,Sentence]]