LUCENE-4981: Deprecate PositionFilter.

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1483403 13f79535-47bb-0310-9956-ffa450edef68

lucene/CHANGES.txt

@@ -76,6 +76,12 @@ Changes in backwards compatibility policy
 * LUCENE-3907: EdgeNGramTokenFilter does not support backward grams and does
   not update offsets anymore. (Adrien Grand)
 
+* LUCENE-4981: PositionFilter is now deprecated as it can corrupt token stream
+  graphs. Since it main use-case was to make query parsers generate boolean
+  queries instead of phrase queries, it is now advised to use
+  QueryParser.setAutoGeneratePhraseQueries(false) (for simple cases) or to
+  override QueryParser.newFieldQuery. (Adrien Grand, Steve Rowe)
+
 Bug Fixes
 
 * LUCENE-4997: Internal test framework's tests are sensitive to previous 

lucene/analysis/common/src/java/org/apache/lucene/analysis/position/PositionFilter.java

@@ -22,11 +22,19 @@ import java.io.IOException;
 import org.apache.lucene.analysis.TokenFilter;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
+import org.apache.lucene.queryparser.classic.QueryParser;
 
 /** Set the positionIncrement of all tokens to the "positionIncrement",
  * except the first return token which retains its original positionIncrement value.
  * The default positionIncrement value is zero.
+ * @deprecated (4.4) PositionFilter makes {@link TokenStream} graphs inconsistent
+ *             which can cause highlighting bugs. Its main use-case being to make
+ *             {@link QueryParser} generate boolean queries instead of phrase
+ *             queries, it is now advised to use
+ *             {@link QueryParser#setAutoGeneratePhraseQueries(boolean) QueryParser.setAutoGeneratePhraseQueries(false)}
+ *             (for simple cases) or to override {@link QueryParser#newFieldQuery}.
  */
+@Deprecated
 public final class PositionFilter extends TokenFilter {
 
   /** Position increment to assign to all but the first token - default = 0 */

lucene/analysis/common/src/java/org/apache/lucene/analysis/position/PositionFilterFactory.java

@@ -20,6 +20,7 @@ package org.apache.lucene.analysis.position;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.position.PositionFilter;
 import org.apache.lucene.analysis.util.TokenFilterFactory;
+import org.apache.lucene.util.Version;
 
 import java.util.Map;
 
@@ -37,7 +38,9 @@ import java.util.Map;
  *
  * @see org.apache.lucene.analysis.position.PositionFilter
  * @since solr 1.4
+ * @deprecated (4.4)
  */
+@Deprecated
 public class PositionFilterFactory extends TokenFilterFactory {
   private final int positionIncrement;
 
@@ -48,6 +51,9 @@ public class PositionFilterFactory extends TokenFilterFactory {
     if (!args.isEmpty()) {
       throw new IllegalArgumentException("Unknown parameters: " + args);
     }
+    if (luceneMatchVersion != null && luceneMatchVersion.onOrAfter(Version.LUCENE_44)) {
+      throw new IllegalArgumentException("PositionFilter is deprecated as of Lucene 4.4. You should either fix your code to not use it or use Lucene 4.3 version compatibility");
+    }
   }
 
   @Override

lucene/analysis/common/src/test/org/apache/lucene/analysis/core/TestRandomChains.java

@@ -174,8 +174,6 @@ public class TestRandomChains extends BaseTokenStreamTestCase {
           PathHierarchyTokenizer.class,
           HyphenationCompoundWordTokenFilter.class,
           DictionaryCompoundWordTokenFilter.class,
-          // TODO: corrumpts graphs (offset consistency check):
-          PositionFilter.class,
           // TODO: it seems to mess up offsets!?
           WikipediaTokenizer.class,
           // TODO: doesn't handle graph inputs

lucene/analysis/common/src/test/org/apache/lucene/analysis/ngram/EdgeNGramTokenFilterTest.java

@@ -19,13 +19,16 @@ package org.apache.lucene.analysis.ngram;
 
 import org.apache.lucene.analysis.Analyzer;
 import org.apache.lucene.analysis.MockTokenizer;
+import org.apache.lucene.analysis.TokenFilter;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.BaseTokenStreamTestCase;
 import org.apache.lucene.analysis.Tokenizer;
 import org.apache.lucene.analysis.core.KeywordTokenizer;
 import org.apache.lucene.analysis.core.WhitespaceTokenizer;
 import org.apache.lucene.analysis.position.PositionFilter;
+import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
 
+import java.io.IOException;
 import java.io.Reader;
 import java.io.StringReader;
 import java.util.Random;
@@ -101,9 +104,39 @@ public class EdgeNGramTokenFilterTest extends BaseTokenStreamTestCase {
                               false);
   }
 
+  private static class PositionFilter extends TokenFilter {
+    
+    private final PositionIncrementAttribute posIncrAtt = addAttribute(PositionIncrementAttribute.class);
+    private boolean started;
+    
+    PositionFilter(final TokenStream input) {
+      super(input);
+    }
+    
+    @Override
+    public final boolean incrementToken() throws IOException {
+      if (input.incrementToken()) {
+        if (started) {
+          posIncrAtt.setPositionIncrement(0);
+        } else {
+          started = true;
+        }
+        return true;
+      } else {
+        return false;
+      }
+    }
+    
+    @Override
+    public void reset() throws IOException {
+      super.reset();
+      started = false;
+    }
+  }
+
   public void testFirstTokenPositionIncrement() throws Exception {
     TokenStream ts = new MockTokenizer(new StringReader("a abc"), MockTokenizer.WHITESPACE, false);
-    ts = new PositionFilter(ts, 0); // All but first token will get 0 position increment
+    ts = new PositionFilter(ts); // All but first token will get 0 position increment
     EdgeNGramTokenFilter filter = new EdgeNGramTokenFilter(TEST_VERSION_CURRENT, ts, 2, 3);
     // The first token "a" will not be output, since it's smaller than the mingram size of 2.
     // The second token on input to EdgeNGramTokenFilter will have position increment of 0,