OPENNLP-1236: Add arabic and greek stemmers support (#345)

Author: maximestein

opennlp-tools/src/main/java/opennlp/tools/stemmer/snowball/SnowballProgram.java

@@ -190,6 +190,11 @@ protected boolean eq_s(int s_size, String s)
 	return true;
     }
 
+		protected boolean eq_s(CharSequence s)
+	{
+		return eq_s(s.length(), s.toString());
+	}
+
     protected boolean eq_s_b(int s_size, String s)
     {
 	if (cursor - limit_backward < s_size) return false;
@@ -201,6 +206,11 @@ protected boolean eq_s_b(int s_size, String s)
 	return true;
     }
 
+    protected boolean eq_s_b(CharSequence s)
+		{
+			return eq_s_b(s.length(), s.toString());
+		}
+
     protected boolean eq_v(CharSequence s)
     {
 	return eq_s(s.length(), s.toString());
@@ -282,6 +292,10 @@ protected int find_among(Among v[], int v_size)
 	}
     }
 
+    protected int find_among(Among v[]){
+    	return find_among(v, v.length);
+		}
+
     // find_among_b is for backwards processing. Same comments apply
     protected int find_among_b(Among v[], int v_size)
     {
@@ -351,6 +365,10 @@ protected int find_among_b(Among v[], int v_size)
 	}
     }
 
+    protected int find_among_b(Among v[]){
+    	return find_among_b(v, v.length);
+		}
+
     /* to replace chars between c_bra and c_ket in current by the
      * chars in s.
      */

opennlp-tools/src/main/java/opennlp/tools/stemmer/snowball/SnowballStemmer.java

@@ -22,12 +22,14 @@
 public class SnowballStemmer implements Stemmer {
 
   public enum ALGORITHM {
+    ARABIC,
     DANISH,
     DUTCH,
     ENGLISH,
     FINNISH,
     FRENCH,
     GERMAN,
+    GREEK,
     HUNGARIAN,
     IRISH,
     ITALIAN,
@@ -47,7 +49,10 @@ public enum ALGORITHM {
   public SnowballStemmer(ALGORITHM algorithm, int repeat) {
     this.repeat = repeat;
 
-    if (ALGORITHM.DANISH.equals(algorithm)) {
+    if (ALGORITHM.ARABIC.equals(algorithm)) {
+      stemmer = new arabicStemmer();
+    }
+    else if (ALGORITHM.DANISH.equals(algorithm)) {
       stemmer = new danishStemmer();
     }
     else if (ALGORITHM.DUTCH.equals(algorithm)) {
@@ -65,6 +70,9 @@ else if (ALGORITHM.FRENCH.equals(algorithm)) {
     else if (ALGORITHM.GERMAN.equals(algorithm)) {
       stemmer = new germanStemmer();
     }
+    else if (ALGORITHM.GREEK.equals(algorithm)) {
+      stemmer = new greekStemmer();
+    }
     else if (ALGORITHM.HUNGARIAN.equals(algorithm)) {
       stemmer = new hungarianStemmer();
     }