You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@lucene.apache.org by mi...@apache.org on 2011/05/12 19:08:26 UTC
svn commit: r1102381 - in /lucene/dev/branches/branch_3x: ./ lucene/
lucene/backwards/ lucene/contrib/
lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/
lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/ solr/
Author: mikemccand
Date: Thu May 12 17:08:26 2011
New Revision: 1102381
URL: http://svn.apache.org/viewvc?rev=1102381&view=rev
Log:
LUCENE-3087: fix highlighter case that prevented highlighting exact phrase when tokens overlap
Modified:
lucene/dev/branches/branch_3x/ (props changed)
lucene/dev/branches/branch_3x/lucene/ (props changed)
lucene/dev/branches/branch_3x/lucene/backwards/ (props changed)
lucene/dev/branches/branch_3x/lucene/contrib/CHANGES.txt
lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/TokenSources.java
lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/TokenSourcesTest.java
lucene/dev/branches/branch_3x/solr/ (props changed)
Modified: lucene/dev/branches/branch_3x/lucene/contrib/CHANGES.txt
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_3x/lucene/contrib/CHANGES.txt?rev=1102381&r1=1102380&r2=1102381&view=diff
==============================================================================
--- lucene/dev/branches/branch_3x/lucene/contrib/CHANGES.txt (original)
+++ lucene/dev/branches/branch_3x/lucene/contrib/CHANGES.txt Thu May 12 17:08:26 2011
@@ -140,6 +140,10 @@ Bug fixes
* LUCENE-2943: Fix thread-safety issues with ICUCollationKeyFilter.
(Robert Muir)
+ * LUCENE-3087: Highlighter: fix case that was preventing highlighting
+ of exact phrase when tokens overlap. (Pierre Gossé via Mike
+ McCandless)
+
API Changes
* LUCENE-2867: Some contrib queryparser methods that receives CharSequence as
Modified: lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/TokenSources.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/TokenSources.java?rev=1102381&r1=1102380&r2=1102381&view=diff
==============================================================================
--- lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/TokenSources.java (original)
+++ lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/java/org/apache/lucene/search/highlight/TokenSources.java Thu May 12 17:08:26 2011
@@ -30,6 +30,7 @@ import org.apache.lucene.analysis.Token;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
import org.apache.lucene.analysis.tokenattributes.OffsetAttribute;
+import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.TermFreqVector;
@@ -157,10 +158,13 @@ public class TokenSources {
OffsetAttribute offsetAtt;
+ PositionIncrementAttribute posincAtt;
+
StoredTokenStream(Token tokens[]) {
this.tokens = tokens;
termAtt = addAttribute(CharTermAttribute.class);
offsetAtt = addAttribute(OffsetAttribute.class);
+ posincAtt = (PositionIncrementAttribute) addAttribute(PositionIncrementAttribute.class);
}
@Override
@@ -172,6 +176,10 @@ public class TokenSources {
clearAttributes();
termAtt.setEmpty().append(token);
offsetAtt.setOffset(token.startOffset(), token.endOffset());
+ posincAtt
+ .setPositionIncrement(currentToken <= 1
+ || tokens[currentToken - 1].startOffset() > tokens[currentToken - 2]
+ .startOffset() ? 1 : 0);
return true;
}
}
@@ -179,7 +187,6 @@ public class TokenSources {
String[] terms = tpv.getTerms();
int[] freq = tpv.getTermFrequencies();
int totalTokens = 0;
-
for (int t = 0; t < freq.length; t++) {
totalTokens += freq[t];
}
@@ -188,7 +195,8 @@ public class TokenSources {
for (int t = 0; t < freq.length; t++) {
TermVectorOffsetInfo[] offsets = tpv.getOffsets(t);
if (offsets == null) {
- throw new IllegalArgumentException("Required TermVector Offset information was not found");
+ throw new IllegalArgumentException(
+ "Required TermVector Offset information was not found");
}
int[] pos = null;
@@ -232,10 +240,9 @@ public class TokenSources {
.size()]);
ArrayUtil.mergeSort(tokensInOriginalOrder, new Comparator<Token>() {
public int compare(Token t1, Token t2) {
- if (t1.startOffset() == t2.startOffset())
- return t1.endOffset() - t2.endOffset();
- else
- return t1.startOffset() - t2.startOffset();
+ if (t1.startOffset() == t2.startOffset()) return t1.endOffset()
+ - t2.endOffset();
+ else return t1.startOffset() - t2.startOffset();
}
});
}
Modified: lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/TokenSourcesTest.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/TokenSourcesTest.java?rev=1102381&r1=1102380&r2=1102381&view=diff
==============================================================================
--- lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/TokenSourcesTest.java (original)
+++ lucene/dev/branches/branch_3x/lucene/contrib/highlighter/src/test/org/apache/lucene/search/highlight/TokenSourcesTest.java Thu May 12 17:08:26 2011
@@ -36,7 +36,10 @@ import org.apache.lucene.index.Term;
import org.apache.lucene.index.TermPositionVector;
import org.apache.lucene.search.DisjunctionMaxQuery;
import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.search.Query;
import org.apache.lucene.search.TopDocs;
+import org.apache.lucene.search.spans.SpanNearQuery;
+import org.apache.lucene.search.spans.SpanQuery;
import org.apache.lucene.search.spans.SpanTermQuery;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.LockObtainFailedException;
@@ -86,12 +89,12 @@ public class TokenSourcesTest extends Lu
public void reset() {
this.i = -1;
this.tokens = new Token[] {
- new Token(new char[] { 't', 'h', 'e' }, 0, 3, 0, 3),
- new Token(new char[] { '{', 'f', 'o', 'x', '}' }, 0, 5, 0, 7),
- new Token(new char[] { 'f', 'o', 'x' }, 0, 3, 4, 7),
- new Token(new char[] { 'd', 'i', 'd' }, 0, 3, 8, 11),
- new Token(new char[] { 'n', 'o', 't' }, 0, 3, 12, 15),
- new Token(new char[] { 'j', 'u', 'm', 'p' }, 0, 4, 16, 20) };
+ new Token(new char[] {'t', 'h', 'e'}, 0, 3, 0, 3),
+ new Token(new char[] {'{', 'f', 'o', 'x', '}'}, 0, 5, 0, 7),
+ new Token(new char[] {'f', 'o', 'x'}, 0, 3, 4, 7),
+ new Token(new char[] {'d', 'i', 'd'}, 0, 3, 8, 11),
+ new Token(new char[] {'n', 'o', 't'}, 0, 3, 12, 15),
+ new Token(new char[] {'j', 'u', 'm', 'p'}, 0, 4, 16, 20)};
this.tokens[1].setPositionIncrement(0);
}
}
@@ -188,4 +191,97 @@ public class TokenSourcesTest extends Lu
}
}
+ public void testOverlapWithOffsetExactPhrase() throws CorruptIndexException,
+ LockObtainFailedException, IOException, InvalidTokenOffsetsException {
+ final String TEXT = "the fox did not jump";
+ final Directory directory = newDirectory();
+ final IndexWriter indexWriter = new IndexWriter(directory,
+ newIndexWriterConfig(TEST_VERSION_CURRENT, new OverlapAnalyzer()));
+ try {
+ final Document document = new Document();
+ document.add(new Field(FIELD, new TokenStreamOverlap(),
+ TermVector.WITH_OFFSETS));
+ indexWriter.addDocument(document);
+ } finally {
+ indexWriter.close();
+ }
+ final IndexReader indexReader = IndexReader.open(directory, true);
+ try {
+ assertEquals(1, indexReader.numDocs());
+ final IndexSearcher indexSearcher = newSearcher(indexReader);
+ try {
+ // final DisjunctionMaxQuery query = new DisjunctionMaxQuery(1);
+ // query.add(new SpanTermQuery(new Term(FIELD, "{fox}")));
+ // query.add(new SpanTermQuery(new Term(FIELD, "fox")));
+ final Query phraseQuery = new SpanNearQuery(new SpanQuery[] {
+ new SpanTermQuery(new Term(FIELD, "the")),
+ new SpanTermQuery(new Term(FIELD, "fox"))}, 0, true);
+
+ TopDocs hits = indexSearcher.search(phraseQuery, 1);
+ assertEquals(1, hits.totalHits);
+ final Highlighter highlighter = new Highlighter(
+ new SimpleHTMLFormatter(), new SimpleHTMLEncoder(),
+ new QueryScorer(phraseQuery));
+ final TokenStream tokenStream = TokenSources
+ .getTokenStream(
+ (TermPositionVector) indexReader.getTermFreqVector(0, FIELD),
+ false);
+ assertEquals("<B>the fox</B> did not jump",
+ highlighter.getBestFragment(tokenStream, TEXT));
+ } finally {
+ indexSearcher.close();
+ }
+ } finally {
+ indexReader.close();
+ directory.close();
+ }
+ }
+
+ public void testOverlapWithPositionsAndOffsetExactPhrase()
+ throws CorruptIndexException, LockObtainFailedException, IOException,
+ InvalidTokenOffsetsException {
+ final String TEXT = "the fox did not jump";
+ final Directory directory = newDirectory();
+ final IndexWriter indexWriter = new IndexWriter(directory,
+ newIndexWriterConfig(TEST_VERSION_CURRENT, new OverlapAnalyzer()));
+ try {
+ final Document document = new Document();
+ document.add(new Field(FIELD, new TokenStreamOverlap(),
+ TermVector.WITH_POSITIONS_OFFSETS));
+ indexWriter.addDocument(document);
+ } finally {
+ indexWriter.close();
+ }
+ final IndexReader indexReader = IndexReader.open(directory, true);
+ try {
+ assertEquals(1, indexReader.numDocs());
+ final IndexSearcher indexSearcher = newSearcher(indexReader);
+ try {
+ // final DisjunctionMaxQuery query = new DisjunctionMaxQuery(1);
+ // query.add(new SpanTermQuery(new Term(FIELD, "the")));
+ // query.add(new SpanTermQuery(new Term(FIELD, "fox")));
+ final Query phraseQuery = new SpanNearQuery(new SpanQuery[] {
+ new SpanTermQuery(new Term(FIELD, "the")),
+ new SpanTermQuery(new Term(FIELD, "fox"))}, 0, true);
+
+ TopDocs hits = indexSearcher.search(phraseQuery, 1);
+ assertEquals(1, hits.totalHits);
+ final Highlighter highlighter = new Highlighter(
+ new SimpleHTMLFormatter(), new SimpleHTMLEncoder(),
+ new QueryScorer(phraseQuery));
+ final TokenStream tokenStream = TokenSources
+ .getTokenStream(
+ (TermPositionVector) indexReader.getTermFreqVector(0, FIELD),
+ false);
+ assertEquals("<B>the fox</B> did not jump",
+ highlighter.getBestFragment(tokenStream, TEXT));
+ } finally {
+ indexSearcher.close();
+ }
+ } finally {
+ indexReader.close();
+ directory.close();
+ }
+ }
+
}