SOLR-540: globbing in hl.fl

git-svn-id: https://svn.apache.org/repos/asf/lucene/solr/trunk@730008 13f79535-47bb-0310-9956-ffa450edef68

CHANGES.txt

@@ -128,6 +128,11 @@ New Features
 27. SOLR-847: Enhance the snappull command in ReplicationHandler to accept masterUrl.
     (Noble Paul, Preetam Rao via shalin)
 
+28. SOLR-540: Add support for globbing in field names to highlight.
+    For example, hl.fl=*_text will highlight all fieldnames ending with
+    _text.  (Lars Kotthoff via yonik)
+
+
 Optimizations
 ----------------------
  1. SOLR-374: Use IndexReader.reopen to save resources by re-using parts of the

src/java/org/apache/solr/highlight/SolrHighlighter.java

@@ -17,8 +17,11 @@ package org.apache.solr.highlight;
  */
 
 import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Collection;
 import java.util.Collections;
 import java.util.HashMap;
+import java.util.List;
 import java.util.Map;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
@@ -79,9 +82,22 @@ public abstract class SolrHighlighter
       }
     }
     else if (fields.length == 1) {
+      if (fields[0].contains("*")) {
+        // create a Java regular expression from the wildcard string
+        String fieldRegex = fields[0].replaceAll("\\*", ".*");
+        Collection<String> storedHighlightFieldNames = request.getSearcher().getStoredHighlightFieldNames();
+        List<String> storedFieldsToHighlight = new ArrayList<String>();
+        for (String storedFieldName: storedHighlightFieldNames) {
+            if (storedFieldName.matches(fieldRegex)) {
+              storedFieldsToHighlight.add(storedFieldName);
+            }
+        }
+        fields = storedFieldsToHighlight.toArray(new String[] {});
+      } else {
         // if there's a single request/handler value, it may be a space/comma separated list
         fields = SolrPluginUtils.split(fields[0]);
       }
+    }
 
     return fields;
   }

src/java/org/apache/solr/search/SolrIndexSearcher.java

@@ -31,6 +31,7 @@ import org.apache.solr.core.SolrConfig;
 import org.apache.solr.core.SolrCore;
 import org.apache.solr.core.SolrInfoMBean;
 import org.apache.solr.schema.IndexSchema;
+import org.apache.solr.schema.SchemaField;
 import org.apache.solr.request.UnInvertedField;
 import org.apache.lucene.util.OpenBitSet;
 
@@ -93,6 +94,9 @@ public class SolrIndexSearcher extends Searcher implements SolrInfoMBean {
   private final SolrCache[] cacheList;
   private static final SolrCache[] noCaches = new SolrCache[0];
   
+  private final Collection<String> fieldNames;
+  private Collection<String> storedHighlightFieldNames;
+
   /** Creates a searcher searching the index in the named directory.
    * 
    * @deprecated use alternate constructor
@@ -179,6 +183,8 @@ public class SolrIndexSearcher extends Searcher implements SolrInfoMBean {
     // for DocSets
     HASHSET_INVERSE_LOAD_FACTOR = solrConfig.hashSetInverseLoadFactor;
     HASHDOCSET_MAXSIZE = solrConfig.hashDocSetMaxSize;
+
+    fieldNames = r.getFieldNames(IndexReader.FieldOption.ALL);
   }
 
 
@@ -233,6 +239,36 @@ public class SolrIndexSearcher extends Searcher implements SolrInfoMBean {
   public IndexReader getReader() { return reader; }
   /** Direct access to the IndexSchema for use with this searcher */
   public IndexSchema getSchema() { return schema; }
+  
+  /**
+   * Returns a collection of all field names the index reader knows about.
+   */
+  public Collection<String> getFieldNames() {
+    return fieldNames;
+  }
+
+  /**
+   * Returns a collection of the names of all stored fields which can be
+   * highlighted the index reader knows about.
+   */
+  public Collection<String> getStoredHighlightFieldNames() {
+    if (storedHighlightFieldNames == null) {
+      storedHighlightFieldNames = new LinkedList<String>();
+      for (String fieldName : fieldNames) {
+        try {
+          SchemaField field = schema.getField(fieldName);
+          if (field.stored() &&
+                  ((field.getType() instanceof org.apache.solr.schema.TextField) ||
+                  (field.getType() instanceof org.apache.solr.schema.StrField))) {
+            storedHighlightFieldNames.add(fieldName);
+          }
+        } catch (RuntimeException e) { // getField() throws a SolrException, but it arrives as a RuntimeException
+            log.warn("Field \"" + fieldName + "\" found in index, but not defined in schema.");
+        }
+      }
+    }
+    return storedHighlightFieldNames;
+  }
   //
   // Set default regenerators on filter and query caches if they don't have any
   //

src/test/org/apache/solr/highlight/HighlighterTest.java

@@ -22,12 +22,15 @@ import org.apache.lucene.analysis.Token;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.WhitespaceAnalyzer;
 import org.apache.solr.core.SolrCore;
+import org.apache.solr.request.SolrQueryRequest;
 import org.apache.solr.util.*;
 import org.apache.solr.common.params.HighlightParams;
 
 import java.io.IOException;
 import java.io.StringReader;
+import java.util.Arrays;
 import java.util.HashMap;
+import java.util.List;
 
 /**
  * Tests some basic functionality of Solr while demonstrating good
@@ -630,4 +633,43 @@ public class HighlighterTest extends AbstractSolrTestCase {
         oldHighlight1, oldHighlight2, oldHighlight3
         );
   }
+  
+  public void testGetHighlightFields() {
+    HashMap<String, String> args = new HashMap<String, String>();
+    args.put("fl", "id score");
+    args.put("hl", "true");
+    args.put("hl.fl", "t*");
+
+    assertU(adoc("id", "0", "title", "test", // static stored
+        "text", "test", // static not stored
+        "foo_s", "test", // dynamic stored
+        "foo_sI", "test", // dynamic not stored
+        "weight", "1.0")); // stored but not text
+    assertU(commit());
+    assertU(optimize());
+
+    TestHarness.LocalRequestFactory lrf = h.getRequestFactory("standard", 0,
+        10, args);
+    SolrQueryRequest request = lrf.makeRequest("test");
+    SolrHighlighter highlighter = request.getCore().getHighlighter();
+    List<String> highlightFieldNames = Arrays.asList(highlighter
+        .getHighlightFields(null, request, new String[] {}));
+    assertTrue("Expected to highlight on field \"title\"", highlightFieldNames
+        .contains("title"));
+    assertFalse("Expected to not highlight on field \"text\"",
+        highlightFieldNames.contains("text"));
+    assertFalse("Expected to not highlight on field \"weight\"",
+        highlightFieldNames.contains("weight"));
+
+    args.put("hl.fl", "foo_*");
+    lrf = h.getRequestFactory("standard", 0, 10, args);
+    request = lrf.makeRequest("test");
+    highlighter = request.getCore().getHighlighter();
+    highlightFieldNames = Arrays.asList(highlighter.getHighlightFields(null,
+        request, new String[] {}));
+    assertEquals("Expected one field to highlight on", 1, highlightFieldNames
+        .size());
+    assertEquals("Expected to highlight on field \"foo_s\"", "foo_s",
+        highlightFieldNames.get(0));
+  }
 }