Merge pull request #1 from imotov/fix-positions-increments-and-keyword-handling

Improve incrementToken implementation in MorphologyFilter

dictionary-reader/src/test/java/org/apache/lucene/morphology/AnalyzersTest.java

@@ -16,8 +16,15 @@
 package org.apache.lucene.morphology;
 
 import org.apache.lucene.analysis.Analyzer;
+import org.apache.lucene.analysis.BaseTokenStreamTestCase;
+import org.apache.lucene.analysis.TokenFilter;
+import org.apache.lucene.analysis.core.LowerCaseFilter;
+import org.apache.lucene.analysis.miscellaneous.SetKeywordMarkerFilter;
+import org.apache.lucene.analysis.standard.StandardFilter;
+import org.apache.lucene.analysis.standard.StandardTokenizer;
 import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
 import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
+import org.apache.lucene.analysis.util.CharArraySet;
 import org.apache.lucene.morphology.analyzer.MorphologyAnalyzer;
 import org.apache.lucene.morphology.analyzer.MorphologyFilter;
 import org.apache.lucene.morphology.english.EnglishAnalyzer;
@@ -31,10 +38,9 @@ import java.io.*;
 import java.util.*;
 
 import static org.hamcrest.Matchers.equalTo;
-import static org.junit.Assert.assertThat;
 
 
-public class AnalyzersTest {
+public class AnalyzersTest extends BaseTokenStreamTestCase {
 
     @Test
     public void shouldGiveCorrectWordsForEnglish() throws IOException {
@@ -116,4 +122,44 @@ public class AnalyzersTest {
 
         assertThat(result, equalTo(answer));
     }
+
+    @Test
+    public void testPositionIncrement() throws IOException {
+        EnglishAnalyzer englishAnalyzer = new EnglishAnalyzer();
+        assertTokenStreamContents(
+                englishAnalyzer.tokenStream("test", "There are tests!"),
+                new String[]{"there", "are", "be", "test"},
+                new int[]{0, 6, 6, 10},
+                new int[]{5, 9, 9, 15},
+                new String[]{"<ALPHANUM>", "<ALPHANUM>", "<ALPHANUM>", "<ALPHANUM>"},
+                new int[]{1, 1, 0, 1}
+        );
+    }
+
+    @Test
+    public void testKeywordHandling() throws IOException {
+        Analyzer analyzer = new EnglishKeywordTestAnalyzer();
+        assertTokenStreamContents(
+                analyzer.tokenStream("test", "Tests shouldn't be stemmed, but tests should!"),
+                new String[]{"tests", "shouldn't", "be", "stem", "but", "test", "shall"}
+        );
+    }
+
+    private static class EnglishKeywordTestAnalyzer extends Analyzer {
+        @Override
+        protected TokenStreamComponents createComponents(String s) {
+            StandardTokenizer src = new StandardTokenizer();
+            TokenFilter filter = new StandardFilter(src);
+            CharArraySet dontStem = new CharArraySet(1, false);
+            dontStem.add("Tests");
+            filter = new SetKeywordMarkerFilter(filter, dontStem);
+            filter = new LowerCaseFilter(filter);
+            try {
+                filter = new MorphologyFilter(filter, new EnglishLuceneMorphology());
+            } catch (IOException ex) {
+                throw new RuntimeException("cannot create EnglishLuceneMorphology", ex);
+            }
+            return new TokenStreamComponents(src, filter);
+        }
+    }
 }

english/pom.xml

@@ -19,11 +19,5 @@
             <version>1.2-SNAPSHOT</version>
         </dependency>
 
-        <dependency>
-            <groupId>junit</groupId>
-            <artifactId>junit</artifactId>
-            <version>4.8.2</version>
-            <scope>test</scope>
-        </dependency>
     </dependencies>
 </project>

morph/src/main/java/org/apache/lucene/morphology/analyzer/MorphologyFilter.java

@@ -19,18 +19,22 @@ package org.apache.lucene.morphology.analyzer;
 import org.apache.lucene.analysis.TokenFilter;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
+import org.apache.lucene.analysis.tokenattributes.KeywordAttribute;
 import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
 import org.apache.lucene.morphology.LuceneMorphology;
 
 import java.io.IOException;
 import java.util.Iterator;
+import java.util.List;
 
 
 public class MorphologyFilter extends TokenFilter {
     private LuceneMorphology luceneMorph;
     private Iterator<String> iterator;
     private final CharTermAttribute termAtt = addAttribute(CharTermAttribute.class);
+    private final KeywordAttribute keywordAttr = addAttribute(KeywordAttribute.class);
     private final PositionIncrementAttribute position = addAttribute(PositionIncrementAttribute.class);
+    private State state = null;
 
     public MorphologyFilter(TokenStream tokenStream, LuceneMorphology luceneMorph) {
         super(tokenStream);
@@ -39,27 +43,39 @@ public class MorphologyFilter extends TokenFilter {
 
 
     final public boolean incrementToken() throws IOException {
-        boolean oldToken = true;
-        while (iterator == null || !iterator.hasNext()) {
+        if (iterator != null) {
+            if (iterator.hasNext()) {
+                restoreState(state);
+                position.setPositionIncrement(0);
+                termAtt.setEmpty().append(iterator.next());
+                return true;
+            } else {
+                state = null;
+                iterator = null;
+            }
+        }
+        while (true) {
             boolean b = input.incrementToken();
             if (!b) {
                 return false;
             }
-            String s = new String(termAtt.buffer(), 0, termAtt.length());
-            if (luceneMorph.checkString(s)) {
-                oldToken = false;
-                iterator = luceneMorph.getNormalForms(s).iterator();
-            } else {
-                return true;
+            if (!keywordAttr.isKeyword() && termAtt.length() > 0) {
+                String s = new String(termAtt.buffer(), 0, termAtt.length());
+                if (luceneMorph.checkString(s)) {
+                    List<String> forms = luceneMorph.getNormalForms(s);
+                    if (forms.isEmpty()) {
+                        continue;
+                    } else if (forms.size() == 1) {
+                        termAtt.setEmpty().append(forms.get(0));
+                    } else {
+                        state = captureState();
+                        iterator = forms.iterator();
+                        termAtt.setEmpty().append(iterator.next());
+                    }
+                }
             }
+            return true;
         }
-        String s = iterator.next();
-        termAtt.setEmpty();
-        termAtt.append(s);
-        if (oldToken) {
-            position.setPositionIncrement(0);
-        }
-        return true;
     }
 
 }

pom.xml

@@ -15,6 +15,10 @@
         <tag>HEAD</tag>
     </scm>
 
+    <properties>
+        <lucene.version>5.1.0</lucene.version>
+    </properties>
+
     <distributionManagement>
         <repository>
             <id>bintray</id>
@@ -24,9 +28,9 @@
 
     <dependencies>
         <dependency>
-            <groupId>junit</groupId>
-            <artifactId>junit</artifactId>
-            <version>4.8.2</version>
+            <groupId>org.apache.lucene</groupId>
+            <artifactId>lucene-test-framework</artifactId>
+            <version>${lucene.version}</version>
             <scope>test</scope>
         </dependency>
         <dependency>
@@ -38,12 +42,12 @@
         <dependency>
             <groupId>org.apache.lucene</groupId>
             <artifactId>lucene-core</artifactId>
-            <version>5.1.0</version>
+            <version>${lucene.version}</version>
         </dependency>
         <dependency>
             <groupId>org.apache.lucene</groupId>
             <artifactId>lucene-analyzers-common</artifactId>
-            <version>5.1.0</version>
+            <version>${lucene.version}</version>
         </dependency>
     </dependencies>