HDFS-11926. Ozone: Implement a common helper to return a range of KVs in levelDB. Contributed by Weiwei Yang.

This commit is contained in:
Anu Engineer 2017-06-06 23:07:17 -07:00 committed by Owen O'Malley
parent f617b2dddb
commit 3c0ed97df9
7 changed files with 368 additions and 229 deletions

View File

@ -1,213 +0,0 @@
/*
* 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.
*/
package org.apache.hadoop.ozone.container.common.helpers;
import com.google.common.base.Preconditions;
import com.google.common.base.Strings;
import org.apache.hadoop.utils.LevelDBStore;
import org.iq80.leveldb.DBIterator;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.Closeable;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
/**
* An utility class to get a list of filtered keys.
*/
public class FilteredKeys implements Closeable {
private static final Logger LOG =
LoggerFactory.getLogger(FilteredKeys.class);
private final DBIterator dbIterator;
private final List<KeyFilter> filters;
private int count = 1000;
public FilteredKeys(LevelDBStore db, int count) {
Preconditions.checkNotNull(db, "LeveDBStore cannot be null.");
this.dbIterator = db.getIterator();
dbIterator.seekToFirst();
this.filters = new ArrayList<KeyFilter>();
if(count > 0) {
this.count = count;
}
}
/**
* Adds a key filter which filters keys by a certain criteria.
* Valid key filter is an implementation of {@link KeyFilter} class.
*
* @param filter
*/
public void addKeyFilter(KeyFilter filter) {
filter.setDbIterator(dbIterator);
filters.add(filter);
}
private boolean filter(String keyName) {
if(filters != null && !filters.isEmpty()) {
for(KeyFilter filter : filters) {
if(!filter.check(keyName)) {
return false;
}
}
}
return true;
}
public List<KeyData> getFilteredKeys() {
List<KeyData> result = new ArrayList<KeyData>();
while (dbIterator.hasNext() && result.size() < count) {
Map.Entry<byte[], byte[]> entry = dbIterator.next();
String keyName = KeyUtils.getKeyName(entry.getKey());
if (filter(keyName)) {
try {
KeyData value = KeyUtils.getKeyData(entry.getValue());
KeyData data = new KeyData(value.getContainerName(), keyName);
result.add(data);
} catch (IOException e) {
LOG.warn("Ignoring adding an invalid entry", e);
}
}
}
return result;
}
@Override public void close() {
if(dbIterator != null) {
try {
dbIterator.close();
} catch (IOException e) {
LOG.warn("Failed to close levelDB connection.", e);
}
}
}
/**
* An abstract class for all key filters.
*/
public static abstract class KeyFilter {
private DBIterator dbIterator;
/**
* Returns if this filter is enabled.
*
* @return true if this filter is enabled, false otherwise.
*/
abstract boolean isEnabled();
/**
* Filters the element by key name. Returns true if the key
* with the given key name complies with the criteria defined
* in this filter.
*
* @param keyName
* @return true if filter passes and false otherwise.
*/
abstract boolean filterKey(String keyName);
/**
* If this filter is enabled, returns true if the key with the
* given key name complies with the criteria defined in this filter;
* if this filter is disabled, always returns true.
*
* @param keyName
* @return true if filter passes and false otherwise.
*/
public boolean check(String keyName) {
return isEnabled()? filterKey(keyName) : true;
}
/**
* Set the {@link DBIterator} this filter used to iterate DB entries.
*
* @param dbIterator
*/
protected void setDbIterator(DBIterator dbIterator) {
this.dbIterator = dbIterator;
}
protected DBIterator getDbIterator() {
return this.dbIterator;
}
}
/**
* Filters keys with a previous key name,
* returns only the keys that whose position is behind the given key name.
*/
public static class PreKeyFilter extends KeyFilter{
private final String prevKey;
private boolean preKeyFound = false;
public PreKeyFilter(LevelDBStore db, String prevKey) {
Preconditions.checkNotNull(db, "LevelDB store cannot be null.");
this.prevKey = prevKey;
}
@Override
protected boolean isEnabled() {
return !Strings.isNullOrEmpty(prevKey);
}
@Override
protected boolean filterKey(String keyName) {
if (preKeyFound) {
return true;
} else {
if (getDbIterator().hasPrev()) {
byte[] prevKeyBytes = getDbIterator().peekPrev().getKey();
String prevKeyActual = KeyUtils.getKeyName(prevKeyBytes);
if (prevKeyActual.equals(prevKey)) {
preKeyFound = true;
}
}
return false;
}
}
}
/**
* Filters keys by a key name prefix.
*/
public static class KeyPrefixFilter extends KeyFilter{
private String prefix = null;
public KeyPrefixFilter(String prefix) {
this.prefix = prefix;
}
@Override
protected boolean isEnabled() {
return !Strings.isNullOrEmpty(prefix);
}
@Override
protected boolean filterKey(String keyName) {
return keyName.startsWith(prefix) ? true : false;
}
}
}

View File

@ -20,9 +20,9 @@ package org.apache.hadoop.ozone.container.common.impl;
import com.google.common.base.Preconditions;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hdfs.DFSUtil;
import org.apache.hadoop.hdfs.ozone.protocol.proto.ContainerProtos;
import org.apache.hadoop.ozone.container.common.helpers.ContainerData;
import org.apache.hadoop.ozone.container.common.helpers.FilteredKeys;
import org.apache.hadoop.ozone.container.common.helpers.KeyData;
import org.apache.hadoop.ozone.container.common.helpers.KeyUtils;
import org.apache.hadoop.ozone.container.common.interfaces.ContainerManager;
@ -30,14 +30,15 @@ import org.apache.hadoop.ozone.container.common.interfaces.KeyManager;
import org.apache.hadoop.ozone.container.common.utils.ContainerCache;
import org.apache.hadoop.scm.container.common.helpers.Pipeline;
import org.apache.hadoop.scm.container.common.helpers.StorageContainerException;
import org.apache.hadoop.ozone.container.common.helpers.FilteredKeys.KeyPrefixFilter;
import org.apache.hadoop.ozone.container.common.helpers.FilteredKeys.PreKeyFilter;
import org.apache.hadoop.utils.LevelDBKeyFilters.KeyPrefixFilter;
import org.apache.hadoop.utils.LevelDBStore;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import static org.apache.hadoop.hdfs.ozone.protocol.proto.ContainerProtos
.Result.IO_EXCEPTION;
@ -169,7 +170,7 @@ public class KeyManagerImpl implements KeyManager {
*/
@Override
public List<KeyData> listKey(
Pipeline pipeline, String prefix, String prevKey, int count)
Pipeline pipeline, String prefix, String startKey, int count)
throws StorageContainerException {
Preconditions.checkNotNull(pipeline,
"Pipeline cannot be null.");
@ -178,10 +179,23 @@ public class KeyManagerImpl implements KeyManager {
ContainerData cData = containerManager.readContainer(pipeline
.getContainerName());
LevelDBStore db = KeyUtils.getDB(cData, conf);
try (FilteredKeys filteredKeys = new FilteredKeys(db, count)) {
filteredKeys.addKeyFilter(new KeyPrefixFilter(prefix));
filteredKeys.addKeyFilter(new PreKeyFilter(db, prevKey));
return filteredKeys.getFilteredKeys();
try {
List<KeyData> result = new ArrayList<KeyData>();
byte[] startKeyInBytes = startKey == null ? null :
DFSUtil.string2Bytes(startKey);
KeyPrefixFilter prefixFilter = new KeyPrefixFilter(prefix);
List<Map.Entry<byte[], byte[]>> range =
db.getRangeKVs(startKeyInBytes, count, prefixFilter);
for(Map.Entry<byte[], byte[]> entry : range) {
String keyName = KeyUtils.getKeyName(entry.getKey());
KeyData value = KeyUtils.getKeyData(entry.getValue());
KeyData data = new KeyData(value.getContainerName(), keyName);
result.add(data);
}
return result;
} catch (IOException e) {
throw new StorageContainerException(e,
ContainerProtos.Result.IO_EXCEPTION);
}
}

View File

@ -60,11 +60,11 @@ public interface KeyManager {
*
* @param pipeline - pipeline.
* @param prefix - Prefix in needed.
* @param prevKey - Key to Start from, EMPTY_STRING to begin.
* @param startKey - Key to start from, EMPTY_STRING to begin.
* @param count - Number of keys to return.
* @return List of Keys that match the criteria.
*/
List<KeyData> listKey(Pipeline pipeline, String prefix, String prevKey,
List<KeyData> listKey(Pipeline pipeline, String prefix, String startKey,
int count) throws StorageContainerException;
/**

View File

@ -0,0 +1,65 @@
/*
* 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.
*/
package org.apache.hadoop.utils;
import com.google.common.base.Strings;
import org.apache.hadoop.hdfs.DFSUtil;
/**
* An utility class to filter levelDB keys.
*/
public class LevelDBKeyFilters {
/**
* Interface for levelDB key filters.
*/
public interface LevelDBKeyFilter {
/**
* Filter levelDB key with a certain condition.
*
* @param preKey previous key.
* @param currentKey current key.
* @param nextKey next key.
* @return true if a certain condition satisfied, return false otherwise.
*/
boolean filterKey(byte[] preKey, byte[] currentKey, byte[] nextKey);
}
/**
* Utility class to filter key by a string prefix. This filter
* assumes keys can be parsed to a string.
*/
public static class KeyPrefixFilter implements LevelDBKeyFilter {
private String keyPrefix = null;
public KeyPrefixFilter(String keyPrefix) {
this.keyPrefix = keyPrefix;
}
@Override public boolean filterKey(byte[] preKey, byte[] currentKey,
byte[] nextKey) {
if (Strings.isNullOrEmpty(keyPrefix)) {
return true;
} else {
return currentKey != null &&
DFSUtil.bytes2String(currentKey).startsWith(keyPrefix);
}
}
}
}

View File

@ -18,21 +18,34 @@
package org.apache.hadoop.utils;
import org.apache.hadoop.utils.LevelDBKeyFilters.LevelDBKeyFilter;
import org.fusesource.leveldbjni.JniDBFactory;
import org.iq80.leveldb.WriteBatch;
import org.iq80.leveldb.DB;
import org.iq80.leveldb.Options;
import org.iq80.leveldb.WriteOptions;
import org.iq80.leveldb.DBIterator;
import org.iq80.leveldb.Options;
import org.iq80.leveldb.Snapshot;
import org.iq80.leveldb.ReadOptions;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.Closeable;
import java.io.File;
import java.io.IOException;
import java.util.List;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Map.Entry;
/**
* LevelDB interface.
*/
public class LevelDBStore implements Closeable {
private static final Logger LOG =
LoggerFactory.getLogger(LevelDBStore.class);
private DB db;
private final File dbFile;
private final Options dbOptions;
@ -194,4 +207,99 @@ public class LevelDBStore implements Closeable {
db.compactRange(null, null);
}
}
/**
* Returns a certain range of key value pairs as a list based on a startKey
* or count.
*
* @param keyPrefix start key.
* @param count number of entries to return.
* @return a range of entries or an empty list if nothing found.
* @throws IOException
*
* @see #getRangeKVs(byte[], int, LevelDBKeyFilter...)
*/
public List<Entry<byte[], byte[]>> getRangeKVs(byte[] keyPrefix, int count)
throws IOException {
LevelDBKeyFilter emptyFilter = (preKey, currentKey, nextKey) -> true;
return getRangeKVs(keyPrefix, count, emptyFilter);
}
/**
* Returns a certain range of key value pairs as a list based on a
* startKey or count. Further a {@link LevelDBKeyFilter} can be added to
* filter keys if necessary. To prevent race conditions while listing
* entries, this implementation takes a snapshot and lists the entries from
* the snapshot. This may, on the other hand, cause the range result slight
* different with actual data if data is updating concurrently.
* <p>
* If the startKey is specified and found in levelDB, this key and the keys
* after this key will be included in the result. If the startKey is null
* all entries will be included as long as other conditions are satisfied.
* If the given startKey doesn't exist, an IOException will be thrown.
* <p>
* The count argument is to limit number of total entries to return,
* the value for count must be an integer greater than 0.
* <p>
* This method allows to specify one or more {@link LevelDBKeyFilter}
* to filter keys by certain condition. Once given, only the entries
* whose key passes all the filters will be included in the result.
*
* @param startKey a start key.
* @param count max number of entries to return.
* @param filters customized one or more {@link LevelDBKeyFilter}.
* @return a list of entries found in the database.
* @throws IOException if an invalid startKey is given or other I/O errors.
* @throws IllegalArgumentException if count is less than 0.
*/
public List<Entry<byte[], byte[]>> getRangeKVs(byte[] startKey,
int count, LevelDBKeyFilter... filters) throws IOException {
List<Entry<byte[], byte[]>> result = new ArrayList<>();
long start = System.currentTimeMillis();
if (count < 0) {
throw new IllegalArgumentException(
"Invalid count given " + count + ", count must be greater than 0");
}
Snapshot snapShot = null;
DBIterator dbIter = null;
try {
snapShot = db.getSnapshot();
ReadOptions readOptions = new ReadOptions().snapshot(snapShot);
dbIter = db.iterator(readOptions);
dbIter.seekToFirst();
if (startKey == null) {
dbIter.seekToFirst();
} else {
if (db.get(startKey) == null) {
throw new IOException("Invalid start key, not found in current db.");
}
dbIter.seek(startKey);
}
while (dbIter.hasNext() && result.size() < count) {
byte[] preKey = dbIter.hasPrev() ? dbIter.peekPrev().getKey() : null;
byte[] nextKey = dbIter.hasNext() ? dbIter.peekNext().getKey() : null;
Entry<byte[], byte[]> current = dbIter.next();
if (filters == null || Arrays.asList(filters).stream()
.allMatch(entry -> entry.filterKey(preKey,
current.getKey(), nextKey))) {
result.add(current);
}
}
} finally {
if (snapShot != null) {
snapShot.close();
}
if (dbIter != null) {
dbIter.close();
}
long end = System.currentTimeMillis();
long timeConsumed = end - start;
if (LOG.isDebugEnabled()) {
LOG.debug("Time consumed for getRangeKVs() is {},"
+ " result length is {}.",
timeConsumed, result.size());
}
}
return result;
}
}

View File

@ -0,0 +1,165 @@
/**
* 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
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* 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.
*/
package org.apache.hadoop.ozone;
import org.apache.commons.io.FileUtils;
import org.apache.hadoop.hdfs.DFSUtilClient;
import org.apache.hadoop.test.GenericTestUtils;
import org.apache.hadoop.utils.LevelDBKeyFilters.KeyPrefixFilter;
import org.apache.hadoop.utils.LevelDBKeyFilters.LevelDBKeyFilter;
import org.apache.hadoop.utils.LevelDBStore;
import org.junit.Rule;
import org.junit.Before;
import org.junit.After;
import org.junit.Test;
import org.junit.Assert;
import org.junit.rules.ExpectedException;
import java.io.File;
import java.io.IOException;
import java.util.List;
import java.util.Map;
/**
* Test class for {@link org.apache.hadoop.utils.LevelDBStore}.
*/
public class TestLevelDBStore {
private LevelDBStore store;
private File testDir;
private final static int MAX_GETRANGE_LENGTH = 100;
@Rule
public ExpectedException expectedException = ExpectedException.none();
@Before
public void init() throws IOException {
testDir = GenericTestUtils.getTestDir(getClass().getSimpleName());
store = new LevelDBStore(testDir, true);
// Add 20 entries.
// {a0 : a-value0} to {a9 : a-value9}
// {b0 : a-value0} to {b0 : b-value9}
for (int i=0; i<10; i++) {
store.put(getBytes("a" + i), getBytes("a-value" + i));
store.put(getBytes("b" + i), getBytes("b-value" + i));
}
}
@After
public void cleanup() throws IOException {
store.close();
store.destroy();
FileUtils.deleteDirectory(testDir);
}
private byte[] getBytes(String str) {
return DFSUtilClient.string2Bytes(str);
}
private String getString(byte[] bytes) {
return DFSUtilClient.bytes2String(bytes);
}
@Test
public void testGetRangeKVs() throws IOException {
List<Map.Entry<byte[], byte[]>> result = null;
// Set empty startKey will return values from beginning.
result = store.getRangeKVs(null, 5);
Assert.assertEquals(5, result.size());
Assert.assertEquals("a-value2", getString(result.get(2).getValue()));
// Returns max available entries after a valid startKey.
result = store.getRangeKVs(getBytes("b0"), MAX_GETRANGE_LENGTH);
Assert.assertEquals(10, result.size());
Assert.assertEquals("b0", getString(result.get(0).getKey()));
Assert.assertEquals("b-value0", getString(result.get(0).getValue()));
result = store.getRangeKVs(getBytes("b0"), 5);
Assert.assertEquals(5, result.size());
// Both startKey and count are honored.
result = store.getRangeKVs(getBytes("a9"), 2);
Assert.assertEquals(2, result.size());
Assert.assertEquals("a9", getString(result.get(0).getKey()));
Assert.assertEquals("a-value9", getString(result.get(0).getValue()));
Assert.assertEquals("b0", getString(result.get(1).getKey()));
Assert.assertEquals("b-value0", getString(result.get(1).getValue()));
// Filter keys by prefix.
// It should returns all "b*" entries.
LevelDBKeyFilter filter1 = new KeyPrefixFilter("b");
result = store.getRangeKVs(null, 100, filter1);
Assert.assertEquals(10, result.size());
Assert.assertTrue(result.stream().allMatch(entry ->
new String(entry.getKey()).startsWith("b")
));
result = store.getRangeKVs(null, 3, filter1);
Assert.assertEquals(3, result.size());
result = store.getRangeKVs(getBytes("b3"), 1, filter1);
Assert.assertEquals("b-value3", getString(result.get(0).getValue()));
// Define a customized filter that filters keys by suffix.
// Returns all "*2" entries.
LevelDBKeyFilter filter2 = (preKey, currentKey, nextKey)
-> getString(currentKey).endsWith("2");
result = store.getRangeKVs(null, MAX_GETRANGE_LENGTH, filter2);
Assert.assertEquals(2, result.size());
Assert.assertEquals("a2", getString(result.get(0).getKey()));
Assert.assertEquals("b2", getString(result.get(1).getKey()));
result = store.getRangeKVs(null, 1, filter2);
Assert.assertEquals(1, result.size());
Assert.assertEquals("a2", getString(result.get(0).getKey()));
// Apply multiple filters.
result = store.getRangeKVs(null, MAX_GETRANGE_LENGTH, filter1, filter2);
Assert.assertEquals(1, result.size());
Assert.assertEquals("b2", getString(result.get(0).getKey()));
Assert.assertEquals("b-value2", getString(result.get(0).getValue()));
// If filter is null, no effect.
result = store.getRangeKVs(null, 1, null);
Assert.assertEquals(1, result.size());
Assert.assertEquals("a0", getString(result.get(0).getKey()));
}
@Test
public void testGetRangeLength() throws IOException {
List<Map.Entry<byte[], byte[]>> result = null;
result = store.getRangeKVs(null, 0);
Assert.assertEquals(0, result.size());
result = store.getRangeKVs(null, 1);
Assert.assertEquals(1, result.size());
// Count less than zero is invalid.
expectedException.expect(IllegalArgumentException.class);
expectedException.expectMessage("Invalid count given");
store.getRangeKVs(null, -1);
}
@Test
public void testInvalidStartKey() throws IOException {
// If startKey is invalid, throws an invalid key exception.
expectedException.expect(IOException.class);
expectedException.expectMessage("Invalid start key");
store.getRangeKVs(getBytes("unknownKey"), MAX_GETRANGE_LENGTH);
}
}

View File

@ -760,17 +760,17 @@ public class TestContainerPersistence {
Assert.assertEquals(expectedKeys.get(1), result.get(0).getKeyName());
// List key with preKev filter
// List key with startKey filter
String k6 = expectedKeys.get(6);
result = keyManager.listKey(pipeline, null, k6, 100);
Assert.assertEquals(3, result.size());
for (int i = 7; i < 10; i++) {
Assert.assertEquals(4, result.size());
for (int i = 6; i < 10; i++) {
Assert.assertEquals(expectedKeys.get(i),
result.get(i - 7).getKeyName());
result.get(i - 6).getKeyName());
}
// List key with both prefix and preKey filter
// List key with both prefix and startKey filter
String k7 = expectedKeys.get(7);
result = keyManager.listKey(pipeline, "k3", k7, 100);
// k3 is after k7, enhance we get an empty result