Return-Path: 
 <oak-commits-return-6681-apmail-jackrabbit-oak-commits-archive=jackrabbit.apache.org@jackrabbit.apache.org>
X-Original-To: apmail-jackrabbit-oak-commits-archive@minotaur.apache.org
Delivered-To: apmail-jackrabbit-oak-commits-archive@minotaur.apache.org
Received: from mail.apache.org (hermes.apache.org [140.211.11.3])
	by minotaur.apache.org (Postfix) with SMTP id 1F2C11148D
	for <apmail-jackrabbit-oak-commits-archive@minotaur.apache.org>;
 Wed,  2 Apr 2014 05:20:23 +0000 (UTC)
Received: (qmail 12649 invoked by uid 500); 2 Apr 2014 05:20:22 -0000
Delivered-To: apmail-jackrabbit-oak-commits-archive@jackrabbit.apache.org
Received: (qmail 12551 invoked by uid 500); 2 Apr 2014 05:20:16 -0000
Mailing-List: contact oak-commits-help@jackrabbit.apache.org; run by ezmlm
Precedence: bulk
List-Help: <mailto:oak-commits-help@jackrabbit.apache.org>
List-Unsubscribe: <mailto:oak-commits-unsubscribe@jackrabbit.apache.org>
List-Post: <mailto:oak-commits@jackrabbit.apache.org>
List-Id: <oak-commits.jackrabbit.apache.org>
Reply-To: oak-dev@jackrabbit.apache.org
Delivered-To: mailing list oak-commits@jackrabbit.apache.org
Received: (qmail 12428 invoked by uid 99); 2 Apr 2014 05:20:11 -0000
Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230)
    by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 02 Apr 2014 05:20:11 +0000
X-ASF-Spam-Status: No, hits=-2000.0 required=5.0
	tests=ALL_TRUSTED
X-Spam-Check-By: apache.org
Received: from [140.211.11.4] (HELO eris.apache.org) (140.211.11.4)
    by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 02 Apr 2014 05:20:08 +0000
Received: from eris.apache.org (localhost [127.0.0.1])
	by eris.apache.org (Postfix) with ESMTP id 40AED23888E4;
	Wed,  2 Apr 2014 05:19:46 +0000 (UTC)
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Subject: svn commit: r1583880 - in
 /jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment:
 ListRecord.java SegmentStream.java
Date: Wed, 02 Apr 2014 05:19:46 -0000
To: oak-commits@jackrabbit.apache.org
From: jukka@apache.org
X-Mailer: svnmailer-1.0.9
Message-Id: <20140402051946.40AED23888E4@eris.apache.org>
X-Virus-Checked: Checked by ClamAV on apache.org

Author: jukka
Date: Wed Apr  2 05:19:45 2014
New Revision: 1583880

URL: http://svn.apache.org/r1583880
Log:
OAK-1660 - SegmentMK: Optimize reading of large binaries

Merge adjacent block records into one big block when reading

Modified:
    jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/ListRecord.java
    jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/SegmentStream.java

Modified: jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/ListRecord.java
URL: http://svn.apache.org/viewvc/jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/ListRecord.java?rev=1583880&r1=1583879&r2=1583880&view=diff
==============================================================================
--- jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/ListRecord.java (original)
+++ jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/ListRecord.java Wed Apr  2 05:19:45 2014
@@ -50,7 +50,6 @@ class ListRecord extends Record {
 
     public RecordId getEntry(int index) {
         checkElementIndex(index, size);
-
         if (size == 1) {
             return getRecordId();
         } else {
@@ -65,24 +64,30 @@ class ListRecord extends Record {
     }
 
     public List<RecordId> getEntries() {
-        if (size == 0) {
+        return getEntries(0, size);
+    }
+
+    public List<RecordId> getEntries(int index, int count) {
+        if (index + count > size) {
+            count = size - index;
+        }
+        if (size == 0 || count == 0) {
             return emptyList();
         } else if (size == 1) {
             return singletonList(getRecordId());
         } else {
-            List<RecordId> list = newArrayListWithCapacity(size);
+            List<RecordId> list = newArrayListWithCapacity(count);
             Segment segment = getSegment();
-            int offset = getOffset();
-            for (int i = 0; i < size; i += bucketSize) {
-                RecordId id = segment.readRecordId(offset);
-                if (bucketSize == 1) {
-                    list.add(id);
-                } else {
-                    ListRecord bucket = new ListRecord(
-                            id, Math.min(bucketSize, size - i));
-                    list.addAll(bucket.getEntries());
-                }
-                offset += Segment.RECORD_ID_BYTES;
+            while (count > 0) {
+                int bucketIndex = index / bucketSize;
+                int bucketOffset = index % bucketSize;
+                RecordId id = segment.readRecordId(getOffset(0, bucketIndex));
+                ListRecord bucket = new ListRecord(
+                        id, Math.min(bucketSize, size - bucketIndex * bucketSize));
+                int n = Math.min(bucket.size() - bucketOffset, count);
+                list.addAll(bucket.getEntries(bucketOffset, n));
+                index += n;
+                count -= n;
             }
             return list;
         }

Modified: jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/SegmentStream.java
URL: http://svn.apache.org/viewvc/jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/SegmentStream.java?rev=1583880&r1=1583879&r2=1583880&view=diff
==============================================================================
--- jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/SegmentStream.java (original)
+++ jackrabbit/oak/trunk/oak-core/src/main/java/org/apache/jackrabbit/oak/plugins/segment/SegmentStream.java Wed Apr  2 05:19:45 2014
@@ -19,10 +19,12 @@ package org.apache.jackrabbit.oak.plugin
 import static com.google.common.base.Preconditions.checkArgument;
 import static com.google.common.base.Preconditions.checkNotNull;
 import static com.google.common.base.Preconditions.checkPositionIndexes;
+import static org.apache.jackrabbit.oak.plugins.segment.Segment.MAX_SEGMENT_SIZE;
 import static org.apache.jackrabbit.oak.plugins.segment.SegmentWriter.BLOCK_SIZE;
 
 import java.io.IOException;
 import java.io.InputStream;
+import java.util.List;
 
 import javax.annotation.CheckForNull;
 
@@ -113,7 +115,7 @@ public class SegmentStream extends Input
     @Override
     public int read() {
         byte[] b = new byte[1];
-        if (read(b) != -1) {
+        if (read(b, 0, 1) != -1) {
             return b[0] & 0xff;
         } else {
             return -1;
@@ -121,38 +123,51 @@ public class SegmentStream extends Input
     }
 
     @Override
-    public int read(byte[] b) {
-        return read(b, 0, b.length);
-    }
-
-    @Override
     public int read(byte[] b, int off, int len) {
         checkNotNull(b);
         checkPositionIndexes(off, off + len, b.length);
+
         if (len == 0) {
             return 0;
         } else if (position == length) {
             return -1;
-        } else if (inline != null) {
-            if (position + len > length) {
-                len = (int) (length - position);
-            }
+        }
+
+        if (position + len > length) {
+            len = (int) (length - position); // > 0 given the earlier check
+        }
+
+        if (inline != null) {
             System.arraycopy(inline, (int) position, b, off, len);
             position += len;
             return len;
         } else {
-            int blockIndex = (int) (position / SegmentWriter.BLOCK_SIZE);
-            int blockOffset = (int) (position % SegmentWriter.BLOCK_SIZE);
-
-            if (blockOffset + len > SegmentWriter.BLOCK_SIZE) {
-                len = SegmentWriter.BLOCK_SIZE - blockOffset;
+            int blockIndex = (int) (position / BLOCK_SIZE);
+            int blockOffset = (int) (position % BLOCK_SIZE);
+            int blockCount =
+                    Math.min(MAX_SEGMENT_SIZE, blockOffset + len + BLOCK_SIZE - 1) // round up
+                    / BLOCK_SIZE;
+
+            List<RecordId> ids = blocks.getEntries(blockIndex, blockCount);
+            RecordId first = ids.get(0); // guaranteed to contain at least one
+            SegmentId segmentId = first.getSegmentId();
+            int offset = first.getOffset();
+            int count = 1;
+            while (count < ids.size()) {
+                RecordId id = ids.get(count);
+                if (id.getSegmentId() == segmentId
+                        && id.getOffset() == offset + count * BLOCK_SIZE) {
+                    count++;
+                } else {
+                    break;
+                }
             }
-            if (position + len > length) {
-                len = (int) (length - position);
+
+            if (blockOffset + len > count * BLOCK_SIZE) {
+                len = count * BLOCK_SIZE - blockOffset;
             }
 
-            BlockRecord block =
-                    new BlockRecord(blocks.getEntry(blockIndex), BLOCK_SIZE);
+            BlockRecord block = new BlockRecord(first, blockOffset + len);
             len = block.read(blockOffset, b, off, len);
             position += len;
             return len;