Mailing-List: contact java-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: java-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01CAB004.F38D0D01"
Subject: Question on highlighting of nested SpanQuery instances
Date: Wed, 17 Feb 2010 14:11:04 -0500
Message-Id: 
 <3B01AF82880E6947A069AA17FF5CFE1503D2693C@0015-its-exmb04.us.saic.com>
Thread-Topic: Question on highlighting of nested SpanQuery instances
Thread-Index: AcqwBPOI+y3Jb+dBRDiwziMWUSIbyg==
From: "Goddard, Michael J." <MICHAEL.J.GODDARD@saic.com>
To: <java-user@lucene.apache.org>

------_=_NextPart_001_01CAB004.F38D0D01
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hello,

I'm seeking some help with a highlighting issue involving the SpanQuery =
family.  To illustrate my issue, I added a test to the existing =
HighlighterTest (see diff, below, against tags/lucene_2_9_1).  When this =
test runs, it fails and the System.out.println yields this:

Expected: "Sam dislikes most of the food and has to order <B>fish</B> =
and <B>chips</B> - however the fish is <B>frozen</B>, not fresh.
Observed: "Sam dislikes most of the food and has to order <B>fish</B> =
and <B>chips</B> - however the <B>fish</B> is <B>frozen</B>, not fresh.

That second "fish" doesn't satisfy the query, so I don't expect it to be =
highlighted.  Can anyone out there offer a good starting point on this =
one?

Regards,

  Mike


Index: =
contrib/highlighter/src/test/org/apache/lucene/search/highlight/Highlight=
erTest.java
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
--- =
contrib/highlighter/src/test/org/apache/lucene/search/highlight/Highlight=
erTest.java	(revision 908726)
+++ =
contrib/highlighter/src/test/org/apache/lucene/search/highlight/Highlight=
erTest.java	(working copy)
@@ -173,7 +173,40 @@
         "Query in a named field does not result in highlighting when =
that field isn't in the query",
         s1, highlightField(q, FIELD_NAME, s1));
   }
+ =20
+  /*
+   * TODO: Why is that second instance of the term "fish" highlighted?  =
It is not
+   * followed by the term "chips", so it should not be highlighted.
+   */
+  public void testHighlightingNestedSpans() throws Exception {
=20
+	    String pubText =3D "Sam dislikes most of the food and has to =
order"
+			+ " fish and chips - however the fish is frozen, not fresh.";
+	   =20
+	    String fieldName =3D "SOME_FIELD_NAME";
+
+	    SpanOrQuery spanOr =3D new SpanOrQuery(
+				new SpanTermQuery[] {
+						new SpanTermQuery(new Term(fieldName, "fish")),
+						new SpanTermQuery(new Term(fieldName, "term1")),
+						new SpanTermQuery(new Term(fieldName, "term2")),
+						new SpanTermQuery(new Term(fieldName, "term3")) });
+	   =20
+		SpanNearQuery innerSpanNear =3D new SpanNearQuery(new SpanQuery[] {
+				spanOr,
+				new SpanTermQuery(new Term(fieldName, "chips")) }, 2, true);
+	=09
+		SpanNearQuery query =3D new SpanNearQuery(new SpanQuery[] {
+				innerSpanNear,
+				new SpanTermQuery(new Term(fieldName, "frozen")) }, 8, true);
+	=09
+	    String expected =3D "Sam dislikes most of the food and has to =
order"
+			+ " <B>fish</B> and <B>chips</B> - however the fish is =
<B>frozen</B>, not fresh.";
+	    String observed =3D highlightField(query, fieldName, pubText);
+	    System.out.println("Expected: \"" + expected + "\n" + "Observed: =
\"" + observed);
+	    assertEquals("Why is that second instance of the term \"fish\" =
highlighted?", expected, observed);
+  }
+
   /**
    * This method intended for use with =
<tt>testHighlightingWithDefaultField()</tt>
  * @throws InvalidTokenOffsetsException=20

------_=_NextPart_001_01CAB004.F38D0D01--