--- lucene/contrib/CHANGES.txt	2011-06-16 19:56:08.496774800 +0200
+++ lucene/contrib/CHANGES.txt	2011-06-18 12:49:39.980077700 +0200
@@ -29,6 +29,10 @@
    facilitate doing grouping in a distributed environment (Uwe
    Schindler, Mike McCandless)
 
+ * LUCENE-2919: Added PKIndexSplitter, that splits an index according
+   to a middle term in a specified field.  (Jason Rutherglen via Mike
+   McCandless, Uwe Schindler)
+
 API Changes
 
  * LUCENE-3141: add getter method to access fragInfos in FieldFragList.
--- lucene/contrib/misc/src/java/org/apache/lucene/index/PKIndexSplitter.java	1970-01-01 01:00:00.000000000 +0100
+++ lucene/contrib/misc/src/java/org/apache/lucene/index/PKIndexSplitter.java	2011-06-18 12:53:48.528293900 +0200
@@ -0,0 +1,155 @@
+package org.apache.lucene.index;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.io.IOException;
+
+import org.apache.lucene.index.IndexWriterConfig.OpenMode;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.search.DocIdSet;
+import org.apache.lucene.search.DocIdSetIterator;
+import org.apache.lucene.search.Filter;
+import org.apache.lucene.search.TermRangeFilter;
+import org.apache.lucene.util.OpenBitSetDISI;
+import org.apache.lucene.util.IOUtils;
+import org.apache.lucene.util.Version;
+
+/**
+ * Split an index based on a {@link Filter}.
+ */
+public class PKIndexSplitter {
+  private final Filter docsInFirstIndex;
+  private final Directory input;
+  private final Directory dir1;
+  private final Directory dir2; 
+  
+  /**
+   * Split an index based on a {@link Filter}. All documents that match the filter
+   * are sent to dir1, remaining ones to dir2.
+   */
+  public PKIndexSplitter(Directory input, Directory dir1, Directory dir2, Filter docsInFirstIndex) {
+    this.input = input;
+    this.dir1 = dir1;
+    this.dir2 = dir2;
+    this.docsInFirstIndex = docsInFirstIndex;
+  }
+  
+  /**
+   * Split an index based on a  given primary key term 
+   * and a 'middle' term.  If the middle term is present, it's
+   * sent to dir2.
+   */
+  public PKIndexSplitter(Directory input, Directory dir1, Directory dir2, Term midTerm) {
+    this(input, dir1, dir2,
+      new TermRangeFilter(midTerm.field(), null, midTerm.text(), true, false));
+  }
+  
+  public void split() throws IOException {
+    boolean success = false;
+    IndexReader reader = IndexReader.open(input);
+    try {
+      createIndex(dir1, reader, docsInFirstIndex, false);
+      createIndex(dir2, reader, docsInFirstIndex, true);
+      success = true;
+    } finally {
+      IOUtils.closeSafely(!success, reader);
+    }
+  }
+  
+  private void createIndex(Directory target, IndexReader reader, Filter preserveFilter, boolean negateFilter) throws IOException {
+    boolean success = false;
+    IndexWriter w = new IndexWriter(target, new IndexWriterConfig(
+        Version.LUCENE_CURRENT, null).setOpenMode(OpenMode.CREATE));
+    try {
+      w.addIndexes(new DocumentFilteredIndexReader(reader, preserveFilter, negateFilter));
+      success = true;
+    } finally {
+      IOUtils.closeSafely(!success, w);
+    }
+  }
+    
+  public static class DocumentFilteredIndexReader extends FilterIndexReader {
+    final OpenBitSetDISI readerDels;
+    final int numDocs;
+    
+    public DocumentFilteredIndexReader(IndexReader reader, Filter preserveFilter, boolean negateFilter) throws IOException {
+      super(reader);
+      
+      final OpenBitSetDISI bits = new OpenBitSetDISI(in.maxDoc());
+      final DocIdSet docs = preserveFilter.getDocIdSet(in);
+      if (docs != null) {
+        final DocIdSetIterator it = docs.iterator();
+        if (it != null) {
+          bits.inPlaceOr(it);
+        }
+      }
+      // this is somehow inverse, if we negate the filter, we delete all documents it matches!
+      if (!negateFilter) {
+        bits.flip(0, in.maxDoc());
+      }
+
+      if (in.hasDeletions()) {
+        for (int i = 0; i < in.maxDoc(); i++) {
+          if (in.isDeleted(i)) {
+            bits.set(i);
+          }
+        }
+      }
+      
+      this.readerDels = bits;
+      this.numDocs = in.maxDoc() - (int) bits.cardinality();
+    }
+    
+    @Override
+    public int numDocs() {
+      return numDocs;
+    }
+    
+    @Override
+    public boolean hasDeletions() {
+      return (in.maxDoc() != numDocs);
+    }
+
+    @Override
+    public boolean isDeleted(int n) {
+      return readerDels.get(n);
+    }
+
+    @Override
+    public IndexReader[] getSequentialSubReaders() {
+      return null;
+    }
+
+    @Override
+    public TermPositions termPositions() throws IOException {
+      return new FilterTermPositions(in.termPositions()) {
+
+        @Override
+        public boolean next() throws IOException {
+          boolean res;
+          while ((res = super.next())) {
+            if (!readerDels.get(doc())) {
+              break;
+            }
+          }
+          return res;
+        }        
+      };
+    }
+  }
+}
--- lucene/contrib/misc/src/test/org/apache/lucene/index/TestPKIndexSplitter.java	1970-01-01 01:00:00.000000000 +0100
+++ lucene/contrib/misc/src/test/org/apache/lucene/index/TestPKIndexSplitter.java	2011-06-18 12:54:55.592129700 +0200
@@ -0,0 +1,113 @@
+package org.apache.lucene.index;
+
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with this
+ * work for additional information regarding copyright ownership. The ASF
+ * licenses this file to You under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ * 
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * 
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
+ * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
+ * License for the specific language governing permissions and limitations under
+ * the License.
+ */
+
+import java.text.DecimalFormat;
+import java.text.NumberFormat;
+
+import org.apache.lucene.analysis.MockAnalyzer;
+import org.apache.lucene.analysis.MockTokenizer;
+import org.apache.lucene.document.Document;
+import org.apache.lucene.document.Field.Index;
+import org.apache.lucene.document.Field.Store;
+import org.apache.lucene.index.IndexWriterConfig.OpenMode;
+import org.apache.lucene.store.Directory;
+import org.apache.lucene.util.LuceneTestCase;
+
+public class TestPKIndexSplitter extends LuceneTestCase {
+
+  public void testSplit() throws Exception {    
+    NumberFormat format = new DecimalFormat("000000000");
+    Directory dir = newDirectory();
+    IndexWriter w = new IndexWriter(dir, newIndexWriterConfig(
+        TEST_VERSION_CURRENT, new MockAnalyzer(random, MockTokenizer.WHITESPACE, false))
+        .setOpenMode(OpenMode.CREATE));
+    for (int x = 0; x < 11; x++) {
+      Document doc = createDocument(x, "1", 3, format);
+      w.addDocument(doc);
+    }
+    for (int x = 11; x < 20; x++) {
+      Document doc = createDocument(x, "2", 3, format);
+      w.addDocument(doc);
+    }
+    w.close();
+    
+    final Term midTerm = new Term("id", format.format(11));
+    
+    checkSplitting(dir, midTerm, 11, 9);
+    
+    // delete some documents
+    w = new IndexWriter(dir, newIndexWriterConfig(
+        TEST_VERSION_CURRENT, new MockAnalyzer(random, MockTokenizer.WHITESPACE, false))
+        .setOpenMode(OpenMode.APPEND));
+    w.deleteDocuments(midTerm);
+    w.deleteDocuments(new Term("id", format.format(2)));
+    w.close();
+    
+    checkSplitting(dir, midTerm, 10, 8);
+    
+    dir.close();
+  }
+  
+  private void checkSplitting(Directory dir, Term splitTerm, int leftCount, int rightCount) throws Exception {
+    Directory dir1 = newDirectory();
+    Directory dir2 = newDirectory();
+    PKIndexSplitter splitter = new PKIndexSplitter(dir, dir1, dir2, splitTerm);
+    splitter.split();
+    
+    IndexReader ir1 = IndexReader.open(dir1);
+    IndexReader ir2 = IndexReader.open(dir2);
+    assertEquals(leftCount, ir1.numDocs());
+    assertEquals(rightCount, ir2.numDocs());
+    
+    checkContents(ir1, "1");
+    checkContents(ir2, "2");
+    
+    ir1.close();
+    ir2.close();
+    
+    dir1.close();
+    dir2.close();
+  }
+  
+  private void checkContents(IndexReader ir, String indexname) throws Exception {
+    for (int i = 0; i < ir.maxDoc(); i++) {
+      if (!ir.isDeleted(i)) {
+        assertEquals(indexname, ir.document(i).get("indexname"));
+      }
+    }
+  }
+  
+  private Document createDocument(int n, String indexName, 
+      int numFields, NumberFormat format) {
+    StringBuilder sb = new StringBuilder();
+    Document doc = new Document();
+    String id = format.format(n);
+    doc.add(newField("id", id, Store.YES, Index.NOT_ANALYZED));
+    doc.add(newField("indexname", indexName, Store.YES, Index.NOT_ANALYZED));
+    sb.append("a");
+    sb.append(n);
+    doc.add(newField("field1", sb.toString(), Store.YES, Index.ANALYZED));
+    sb.append(" b");
+    sb.append(n);
+    for (int i = 1; i < numFields; i++) {
+      doc.add(newField("field" + (i + 1), sb.toString(), Store.YES, Index.ANALYZED));
+    }
+    return doc;
+  }
+}