+ * Variant of {@link SpanNearQuery} which does not require the clauses to refer + * to the same field. This can be useful for denormalized relational data: for + * example, when indexing a document with conceptually many 'children': + *
+ * + *+ * teacherid: 1 + * studentfirstname: james + * studentsurname: jones + * + * teacherid: 2 + * studenfirstname: james + * studentsurname: smith + * studentfirstname: sally + * studentsurname: jones + *+ * + *
+ * a MultiFieldSpanNearQuery with a slop of 0 can be used to execute a search + * for 'studentfirstname:james studentsurname:jones' to find student James + * Jones, without matching teacherid 2 (which has a 'james' in position 0 and + * 'jones' in position 1. + *
+ *+ * Note: as this class does not search a sngle field, it cannot meaningfully + * implement {@link #getField()}; the first field only will be returned. This + * affects the behaviour of, e.g. + * {@link #createWeight(org.apache.lucene.search.Searcher)} and explanations. + *
+ */ +public class MultiFieldSpanNearQuery extends SpanNearQuery { + public MultiFieldSpanNearQuery(SpanQuery[] clauses, int slop, boolean inOrder) { + super(clauses, slop, inOrder, false); + } +} Property changes on: src/java/org/apache/lucene/search/spans/MultiFieldSpanNearQuery.java ___________________________________________________________________ Added: svn:mime-type + text/plain Index: src/java/org/apache/lucene/search/spans/SpanNearQuery.java =================================================================== --- src/java/org/apache/lucene/search/spans/SpanNearQuery.java (revision 727243) +++ src/java/org/apache/lucene/search/spans/SpanNearQuery.java (working copy) @@ -42,17 +42,25 @@ /** Construct a SpanNearQuery. Matches spans matching a span from each * clause, with up toslop total unmatched positions between
- * them. * When inOrder is true, the spans from each clause
- * must be * ordered as in clauses. */
+ * them. When inOrder is true, the spans from each clause
+ * must be ordered as in clauses. */
public SpanNearQuery(SpanQuery[] clauses, int slop, boolean inOrder) {
-
+ this(clauses, slop, inOrder, true);
+ }
+
+ /** Constructor for child classes to allow them to bypass the 'all clauses
+ * must refer to the same field' test.
+ * @see SpanNearQuery#SpanNearQuery(SpanQuery[], int, boolean)
+ */
+ protected SpanNearQuery(SpanQuery[] clauses, int slop, boolean inOrder,
+ boolean mustBeSameField) {
// copy clauses array into an ArrayList
this.clauses = new ArrayList(clauses.length);
for (int i = 0; i < clauses.length; i++) {
SpanQuery clause = clauses[i];
if (i == 0) { // check field
field = clause.getField();
- } else if (!clause.getField().equals(field)) {
+ } else if (mustBeSameField && !clause.getField().equals(field)) {
throw new IllegalArgumentException("Clauses must have same field.");
}
this.clauses.add(clause);
Index: src/test/org/apache/lucene/search/spans/TestBasics.java
===================================================================
--- src/test/org/apache/lucene/search/spans/TestBasics.java (revision 727243)
+++ src/test/org/apache/lucene/search/spans/TestBasics.java (working copy)
@@ -341,6 +341,18 @@
}
+ public void testSpanAcrossMultipleFields() {
+ SpanTermQuery term1 = new SpanTermQuery(new Term("field1", "eight"));
+ SpanTermQuery term2 = new SpanTermQuery(new Term("field2", "one"));
+ try {
+ new SpanNearQuery(new SpanQuery[] {term1, term2},
+ 0, true);
+ fail("Should not be able to create SpanNearQuery on 2 different fields");
+ } catch (IllegalArgumentException e) {
+ // Expected
+ }
+ }
+
private void checkHits(Query query, int[] results) throws IOException {
CheckHits.checkHits(query, "field", searcher, results);
}
Index: src/test/org/apache/lucene/search/spans/TestMultiFieldSpanNearQuery.java
===================================================================
--- src/test/org/apache/lucene/search/spans/TestMultiFieldSpanNearQuery.java (revision 0)
+++ src/test/org/apache/lucene/search/spans/TestMultiFieldSpanNearQuery.java (revision 0)
@@ -0,0 +1,91 @@
+package org.apache.lucene.search.spans;
+
+import java.io.IOException;
+
+import org.apache.lucene.analysis.standard.StandardAnalyzer;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.Field;
+import org.apache.lucene.index.IndexWriter;
+import org.apache.lucene.index.Term;
+import org.apache.lucene.search.CheckHits;
+import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.store.FSDirectory;
+import org.apache.lucene.store.RAMDirectory;
+import org.apache.lucene.util.LuceneTestCase;
+
+public class TestMultiFieldSpanNearQuery extends LuceneTestCase {
+ private IndexSearcher searcher;
+
+ private static final int INCREMENT_GAP = 100;
+
+ public void setUp() throws Exception {
+ super.setUp();
+ RAMDirectory directory = new RAMDirectory();
+ IndexWriter writer = new IndexWriter(directory, new StandardAnalyzer() {
+ @Override
+ public int getPositionIncrementGap(String fieldName) {
+ return INCREMENT_GAP;
+ }
+ }, true, IndexWriter.MaxFieldLength.LIMITED);
+
+ Document teacher1 = new Document();
+ addStudent(teacher1, "james", "jones");
+ addStudent(teacher1, "sally", "smith");
+ writer.addDocument(teacher1);
+
+ Document teacher2 = new Document();
+ addStudent(teacher2, "james", "smith");
+ addStudent(teacher2, "sally", "jones");
+ writer.addDocument(teacher2);
+
+ Document teacher3 = new Document();
+ addStudent(teacher3, "john jacob james", "jones");
+ writer.addDocument(teacher3);
+
+ addStudent(teacher1, "smith", "sally");
+
+ writer.close();
+
+ searcher = new IndexSearcher(directory);
+ }
+
+ public void testMultiFieldWithSpan0() throws IOException {
+ SpanNearQuery multiFieldSpan0 = new MultiFieldSpanNearQuery(
+ getSearchTerms(), 0, false);
+ // Only document 0 has 'james' and 'jones' in the same position
+ CheckHits.checkHits(multiFieldSpan0, "studentfirst", searcher,
+ new int[] { 0 });
+ }
+
+ public void testMultiFieldWithinFieldableBoundary() throws IOException {
+ SpanNearQuery multiFieldWithinFieldable = new MultiFieldSpanNearQuery(
+ getSearchTerms(), INCREMENT_GAP / 2, false);
+ // Only documents 0 & 2 have 'james' and 'jones' within the position
+ // increment gap
+ CheckHits.checkHits(multiFieldWithinFieldable, "studentfirst", searcher,
+ new int[] { 0, 2 });
+ }
+
+ public void testMultiFieldAcrossFieldableBoundary() throws IOException {
+ SpanNearQuery multiFieldAcrossFieldable = new MultiFieldSpanNearQuery(
+ getSearchTerms(), INCREMENT_GAP, false);
+ // All documents have both 'james' and 'jones' in adjacent terms
+ CheckHits.checkHits(multiFieldAcrossFieldable, "studentfirst", searcher,
+ new int[] { 0, 1, 2 });
+ }
+
+ private SpanQuery[] getSearchTerms() {
+ SpanTermQuery term1 = new SpanTermQuery(new Term("studentfirst", "james"));
+ SpanTermQuery term2 = new SpanTermQuery(new Term("studentsurname", "jones"));
+ return new SpanQuery[] { term1, term2 };
+ }
+
+ private void addStudent(Document teacher, String studentFirstName,
+ String studentSurname) {
+ teacher.add(new Field("studentfirst", studentFirstName, Field.Store.YES,
+ Field.Index.ANALYZED));
+ teacher.add(new Field("studentsurname", studentSurname, Field.Store.YES,
+ Field.Index.ANALYZED));
+ }
+
+}
\ No newline at end of file
Property changes on: src/test/org/apache/lucene/search/spans/TestMultiFieldSpanNearQuery.java
___________________________________________________________________
Added: svn:mime-type
+ text/plain