fixes to #409

dhowe · Apr 18, 2017 · 9bd2428 · 9bd2428
1 parent dab42b7
commit 9bd2428
Show file tree

Hide file tree

Showing 3 changed files with 53 additions and 25 deletions.
diff --git a/java/rita/RiLexicon.java b/java/rita/RiLexicon.java
@@ -163,11 +163,13 @@ public String randomWord() {
 
   /**
    * Returns a random word from the lexicon with the specified part-of-speech
-   * 
    * @see PosTagger
    */
   public String randomWord(String pos) {
-    return randomWordByLength(pos, -1);
+    String word = randomWordByLength(pos, -1);
+    if (word == null)
+      throw new RiTaException("No words with pos="+pos+" found");
+    return word;
   }
 
   /**
@@ -186,15 +188,23 @@ public String randomWord(int syllableCount) {
    */
   public String randomWord(String pos, int syllableCount) {
 
+    boolean pluralize = pos.equals("nns");
+
+    if (pos.equals("v")) pos = "vb";
+    if (pos.equals("r")) pos = "rb";
+    if (pos.equals("a")) pos = "jj";
+    if (pos.equals("n") || pos.equals("nns")) 
+      pos = "nn";
+
     Iterator<String> it = getIterator(pos);
 
     if (it != null) {
 
-      if (syllableCount < 0)
-	return it.next(); // ignore syllable-count
+      if (syllableCount < 0) { // ignore syllable-count
+	return pluralize ? RiTa.pluralize(it.next()) : it.next(); 
+      }
 
       Map<String, String> lookup = lexImpl.getLexicalData();
-
       while (it.hasNext()) {
 
 	String word = it.next();
@@ -205,17 +215,19 @@ public String randomWord(String pos, int syllableCount) {
 	}
 
 	String sylStr = data.split("\\|")[0].trim();
-	if (sylStr.split(SP).length == syllableCount)
-	  return word;
+
+	if (sylStr.split(SP).length == syllableCount) {  
+	  return pluralize ? RiTa.pluralize(word) : word; 
+	}
       }
     }
-
+    
     return E;
   }
 
   private Iterator<String> getIterator(String pos) {
-    Iterator<String> it = (pos == null) ? lexImpl.randomIterator() : 
-      	lexImpl.randomPosIterator(pos);
+    Iterator<String> it = (pos == null) ? lexImpl.randomIterator() : lexImpl
+	.randomPosIterator(pos);
     return it;
   }
 
@@ -229,13 +241,20 @@ public String randomWordByLength(String pos, int targetLength) { // NIAPI
 
     Iterator<String> it = getIterator(pos);
 
-    if (targetLength < 0)
-      return it.next();
+    try {
+
+      if (targetLength < 0) {
+        return it.next();
+      }
 
-    while (it.hasNext()) {
-      String s = it.next();
-      if (s.length() == targetLength)
-	return s;
+      while (it.hasNext()) {
+        String s = it.next();
+        if (s.length() == targetLength)
+          return s;
+      }
+
+    } catch (Exception e) {
+      return null;
     }
 
     return E;

diff --git a/java/rita/RiTa.java b/java/rita/RiTa.java
@@ -145,6 +145,7 @@ public static String trim(String s) {
   /* 
     RiLexicon delegates:
       rhymes
+      randomWord
       similarBySound
       similarByLetter
       similarBySoundAndLetter
@@ -155,10 +156,6 @@ public static String trim(String s) {
       isAdjective
       isVerb
   */
-  public static boolean isAlliteration(String wordA, String wordB) {
-    return getLexicon().isAlliteration(wordA, wordB);
-  }
-
   public static String[] getRhymes(String input) {
     return getLexicon().rhymes(input);
   }
@@ -186,6 +183,10 @@ public String randomWord(int syllableCount) {
   public String randomWord(String pos, int syllableCount) {
     return getLexicon().randomWord(pos, syllableCount);
   }
+
+  public static boolean isAlliteration(String wordA, String wordB) {
+    return getLexicon().isAlliteration(wordA, wordB);
+  }
 
   public boolean isAdverb(String s) {
     return getLexicon().isAdverb(s);
@@ -255,9 +256,8 @@ public String[] similarByLetter(String s, int minEditDistance) {
     return getLexicon().similarByLetter(s, minEditDistance);
   }
 
-  public String[] similarByLetter(String input, int minEditDistance,
-      boolean preserveLength) {
-    return getLexicon().similarByLetter(input, minEditDistance, preserveLength);
+  public String[] similarByLetter(String input, int med, boolean preserveLength) {
+    return getLexicon().similarByLetter(input, med, preserveLength);
   }
 
   public String[] similarBySound(String input, int minDist) {
@@ -268,12 +268,11 @@ public String[] alliterations(String input) {
     return getLexicon().alliterations(input);
   }
 
-  static RiLexicon lexicon;
   static RiLexicon getLexicon() {
     if (lexicon == null)
       lexicon = new RiLexicon();
     return lexicon; 
-  }
+  } static RiLexicon lexicon;
 
   public static int getWordCount(String s) {
     return RiTa.tokenize(s).length;

diff --git a/java/rita/support/RiPos.java b/java/rita/support/RiPos.java
@@ -1,5 +1,7 @@
 package rita.support;
 
+import rita.*;
+
 
 /**
  * Enumerated types for various Part-of-Speech representations.
@@ -198,4 +200,12 @@ public boolean equals(Object obj)
     return this.tag.equals(((RiPos)obj).getTag());
   }
 
+  public static void main(String[] args) {
+    RiLexicon rl = new RiLexicon();
+    for (int i = 0; i < PENN_TAGS.length; i++) {
+      System.out.print(PENN_TAGS[i].toString()+": ");
+      System.out.println(rl.randomWord(PENN_TAGS[i].toString()));
+    }
+  }
+
 }// end