From 25ffbc5f11cf0b281bb30185a0dd61e0d554efa6 Mon Sep 17 00:00:00 2001
From: Michael Stack <stack@apache.org>
Date: Thu, 5 Aug 2010 05:16:10 +0000
Subject: [PATCH] HBASE-2823 Entire Row Deletes not stored in Row+Col Bloom

git-svn-id: https://svn.apache.org/repos/asf/hbase/trunk@982472 13f79535-47bb-0310-9956-ffa450edef68
---
 CHANGES.txt                                   |  2 +
 .../hadoop/hbase/regionserver/StoreFile.java  | 20 ++++++++-
 .../hbase/regionserver/TestHRegion.java       | 41 +++++++++++++++++++
 3 files changed, 61 insertions(+), 2 deletions(-)

diff --git a/CHANGES.txt b/CHANGES.txt
index d6acd580870..989a13f45f4 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -469,6 +469,8 @@ Release 0.21.0 - Unreleased
    HBASE-2871  Make "start|stop" commands symmetric for Master & Cluster
                (Nicolas Spiegelberg via Stack)
    HBASE-2901  HBASE-2461 broke build
+   HBASE-2823  Entire Row Deletes not stored in Row+Col Bloom
+               (Alexander Georgiev via Stack)
 
   IMPROVEMENTS
    HBASE-1760  Cleanup TODOs in HTable
diff --git a/src/main/java/org/apache/hadoop/hbase/regionserver/StoreFile.java b/src/main/java/org/apache/hadoop/hbase/regionserver/StoreFile.java
index 757a50c44f9..107d641c6c0 100644
--- a/src/main/java/org/apache/hadoop/hbase/regionserver/StoreFile.java
+++ b/src/main/java/org/apache/hadoop/hbase/regionserver/StoreFile.java
@@ -694,6 +694,13 @@ public class StoreFile {
 
       if (bloomType != BloomType.NONE && conf != null) {
         float err = conf.getFloat(IO_STOREFILE_BLOOM_ERROR_RATE, (float)0.01);
+        // Since in row+col blooms we have 2 calls to shouldSeek() instead of 1
+        // and the false positives are adding up, we should keep the error rate
+        // twice as low in order to maintain the number of false positives as
+        // desired by the user
+        if (bloomType == BloomType.ROWCOL) {
+          err /= 2;
+        }
         int maxFold = conf.getInt(IO_STOREFILE_BLOOM_MAX_FOLD, 7);
 
         this.bloomFilter = new ByteBloomFilter(maxKeys, err,
@@ -800,7 +807,6 @@ public class StoreFile {
             byte [] result = new byte[rl + ql];
             System.arraycopy(kv.getBuffer(), ro, result, 0,  rl);
             System.arraycopy(kv.getBuffer(), qo, result, rl, ql);
-
             this.bloomFilter.add(result);
             break;
           default:
@@ -943,7 +949,17 @@ public class StoreFile {
       try {
         ByteBuffer bloom = reader.getMetaBlock(BLOOM_FILTER_DATA_KEY, true);
         if (bloom != null) {
-          return this.bloomFilter.contains(key, bloom);
+          if (this.bloomFilterType == BloomType.ROWCOL) {
+            // Since a Row Delete is essentially a DeleteFamily applied to all
+            // columns, a file might be skipped if using row+col Bloom filter.
+            // In order to ensure this file is included an additional check is
+            // required looking only for a row bloom.
+            return this.bloomFilter.contains(key, bloom) ||
+                this.bloomFilter.contains(row, bloom);
+          }
+          else {
+            return this.bloomFilter.contains(key, bloom);
+          }
         }
       } catch (IOException e) {
         LOG.error("Error reading bloom filter data -- proceeding without",
diff --git a/src/test/java/org/apache/hadoop/hbase/regionserver/TestHRegion.java b/src/test/java/org/apache/hadoop/hbase/regionserver/TestHRegion.java
index e9bdb4221bd..c425953ff04 100644
--- a/src/test/java/org/apache/hadoop/hbase/regionserver/TestHRegion.java
+++ b/src/test/java/org/apache/hadoop/hbase/regionserver/TestHRegion.java
@@ -2723,6 +2723,47 @@ public class TestHRegion extends HBaseTestCase {
     checkOneCell(kvs[3], FAMILY, 0, 0, 1);
   }
 
+  /**
+    * Testcase to cover bug-fix for HBASE-2823
+    * Ensures correct delete when issuing delete row
+    * on columns with bloom filter set to row+col (BloomType.ROWCOL)
+   */
+  public void testDeleteRowWithBloomFilter() throws IOException {
+    byte [] tableName = Bytes.toBytes("testDeleteRowWithBloomFilter");
+    byte [] familyName = Bytes.toBytes("familyName");
+
+    // Create Table
+    HColumnDescriptor hcd = new HColumnDescriptor(familyName, Integer.MAX_VALUE,
+        HColumnDescriptor.DEFAULT_COMPRESSION, false, true,
+        HColumnDescriptor.DEFAULT_TTL, "rowcol");
+
+    HTableDescriptor htd = new HTableDescriptor(tableName);
+    htd.addFamily(hcd);
+    HRegionInfo info = new HRegionInfo(htd, null, null, false);
+    Path path = new Path(DIR + "TestDeleteRowWithBloomFilter");
+    region = HRegion.createHRegion(info, path, conf);
+
+    // Insert some data
+    byte row[] = Bytes.toBytes("row1");
+    byte col[] = Bytes.toBytes("col1");
+
+    Put put = new Put(row);
+    put.add(familyName, col, 1, Bytes.toBytes("SomeRandomValue"));
+    region.put(put);
+    region.flushcache();
+
+    Delete del = new Delete(row);
+    region.delete(del, null, true);
+    region.flushcache();
+
+    // Get remaining rows (should have none)
+    Get get = new Get(row);
+    get.addColumn(familyName, col);
+
+    KeyValue[] keyValues = region.get(get, null).raw();
+    assertTrue(keyValues.length == 0);
+  }
+
   private void putData(int startRow, int numRows, byte [] qf,
       byte [] ...families)
   throws IOException {