/*
* 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.accumulo.core.client.mapred;
import java.io.IOException;
import java.net.InetAddress;
import java.nio.ByteBuffer;
import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
import java.util.Set;
import org.apache.accumulo.core.Constants;
import org.apache.accumulo.core.client.AccumuloException;
import org.apache.accumulo.core.client.AccumuloSecurityException;
import org.apache.accumulo.core.client.ClientSideIteratorScanner;
import org.apache.accumulo.core.client.Connector;
import org.apache.accumulo.core.client.Instance;
import org.apache.accumulo.core.client.IsolatedScanner;
import org.apache.accumulo.core.client.IteratorSetting;
import org.apache.accumulo.core.client.RowIterator;
import org.apache.accumulo.core.client.Scanner;
import org.apache.accumulo.core.client.TableDeletedException;
import org.apache.accumulo.core.client.TableNotFoundException;
import org.apache.accumulo.core.client.TableOfflineException;
import org.apache.accumulo.core.client.ZooKeeperInstance;
import org.apache.accumulo.core.client.impl.OfflineScanner;
import org.apache.accumulo.core.client.impl.Tables;
import org.apache.accumulo.core.client.impl.TabletLocator;
import org.apache.accumulo.core.client.mapreduce.lib.util.InputConfigurator;
import org.apache.accumulo.core.client.mock.MockInstance;
import org.apache.accumulo.core.client.security.tokens.AuthenticationToken;
import org.apache.accumulo.core.data.Key;
import org.apache.accumulo.core.data.KeyExtent;
import org.apache.accumulo.core.data.PartialKey;
import org.apache.accumulo.core.data.Range;
import org.apache.accumulo.core.data.Value;
import org.apache.accumulo.core.master.state.tables.TableState;
import org.apache.accumulo.core.security.Authorizations;
import org.apache.accumulo.core.security.CredentialHelper;
import org.apache.accumulo.core.security.thrift.TCredentials;
import org.apache.accumulo.core.util.Pair;
import org.apache.accumulo.core.util.UtilWaitThread;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.InputFormat;
import org.apache.hadoop.mapred.InputSplit;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.RecordReader;
import org.apache.hadoop.mapred.Reporter;
import org.apache.log4j.Level;
import org.apache.log4j.Logger;
/**
* This abstract {@link InputFormat} class allows MapReduce jobs to use Accumulo as the source of K,V pairs.
* <p>
* Subclasses must implement a {@link #getRecordReader(InputSplit, JobConf, Reporter)} to provide a {@link RecordReader} for K,V.
* <p>
* A static base class, RecordReaderBase, is provided to retrieve Accumulo {@link Key}/{@link Value} pairs, but one must implement its
* {@link RecordReaderBase#next(Object, Object)} to transform them to the desired generic types K,V.
* <p>
* See {@link AccumuloInputFormat} for an example implementation.
*/
public abstract class InputFormatBase<K,V> implements InputFormat<K,V> {
private static final Class<?> CLASS = AccumuloInputFormat.class;
protected static final Logger log = Logger.getLogger(CLASS);
/**
* Sets the connector information needed to communicate with Accumulo in this job.
*
* <p>
* <b>WARNING:</b> The serialized token is stored in the configuration and shared with all MapReduce tasks. It is BASE64 encoded to provide a charset safe
* conversion to a string, and is not intended to be secure.
*
* @param job
* the Hadoop job instance to be configured
* @param principal
* a valid Accumulo user name (user must have Table.CREATE permission)
* @param token
* the user's password
* @throws AccumuloSecurityException
* @since 1.5.0
*/
public static void setConnectorInfo(JobConf job, String principal, AuthenticationToken token) throws AccumuloSecurityException {
InputConfigurator.setConnectorInfo(CLASS, job, principal, token);
}
/**
* Determines if the connector has been configured.
*
* @param job
* the Hadoop context for the configured job
* @return true if the connector has been configured, false otherwise
* @since 1.5.0
* @see #setConnectorInfo(JobConf, String, AuthenticationToken)
*/
protected static Boolean isConnectorInfoSet(JobConf job) {
return InputConfigurator.isConnectorInfoSet(CLASS, job);
}
/**
* Gets the user name from the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return the user name
* @since 1.5.0
* @see #setConnectorInfo(JobConf, String, AuthenticationToken)
*/
protected static String getPrincipal(JobConf job) {
return InputConfigurator.getPrincipal(CLASS, job);
}
/**
* Gets the serialized token class from the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return the user name
* @since 1.5.0
* @see #setConnectorInfo(JobConf, String, AuthenticationToken)
*/
protected static String getTokenClass(JobConf job) {
return InputConfigurator.getTokenClass(CLASS, job);
}
/**
* Gets the password from the configuration. WARNING: The password is stored in the Configuration and shared with all MapReduce tasks; It is BASE64 encoded to
* provide a charset safe conversion to a string, and is not intended to be secure.
*
* @param job
* the Hadoop context for the configured job
* @return the decoded user password
* @since 1.5.0
* @see #setConnectorInfo(JobConf, String, AuthenticationToken)
*/
protected static byte[] getToken(JobConf job) {
return InputConfigurator.getToken(CLASS, job);
}
/**
* Configures a {@link ZooKeeperInstance} for this job.
*
* @param job
* the Hadoop job instance to be configured
* @param instanceName
* the Accumulo instance name
* @param zooKeepers
* a comma-separated list of zookeeper servers
* @since 1.5.0
*/
public static void setZooKeeperInstance(JobConf job, String instanceName, String zooKeepers) {
InputConfigurator.setZooKeeperInstance(CLASS, job, instanceName, zooKeepers);
}
/**
* Configures a {@link MockInstance} for this job.
*
* @param job
* the Hadoop job instance to be configured
* @param instanceName
* the Accumulo instance name
* @since 1.5.0
*/
public static void setMockInstance(JobConf job, String instanceName) {
InputConfigurator.setMockInstance(CLASS, job, instanceName);
}
/**
* Initializes an Accumulo {@link Instance} based on the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return an Accumulo instance
* @since 1.5.0
* @see #setZooKeeperInstance(JobConf, String, String)
* @see #setMockInstance(JobConf, String)
*/
protected static Instance getInstance(JobConf job) {
return InputConfigurator.getInstance(CLASS, job);
}
/**
* Sets the log level for this job.
*
* @param job
* the Hadoop job instance to be configured
* @param level
* the logging level
* @since 1.5.0
*/
public static void setLogLevel(JobConf job, Level level) {
InputConfigurator.setLogLevel(CLASS, job, level);
}
/**
* Gets the log level from this configuration.
*
* @param job
* the Hadoop context for the configured job
* @return the log level
* @since 1.5.0
* @see #setLogLevel(JobConf, Level)
*/
protected static Level getLogLevel(JobConf job) {
return InputConfigurator.getLogLevel(CLASS, job);
}
/**
* Sets the name of the input table, over which this job will scan.
*
* @param job
* the Hadoop job instance to be configured
* @param tableName
* the table to use when the tablename is null in the write call
* @since 1.5.0
*/
public static void setInputTableName(JobConf job, String tableName) {
InputConfigurator.setInputTableName(CLASS, job, tableName);
}
/**
* Gets the table name from the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return the table name
* @since 1.5.0
* @see #setInputTableName(JobConf, String)
*/
protected static String getInputTableName(JobConf job) {
return InputConfigurator.getInputTableName(CLASS, job);
}
/**
* Sets the {@link Authorizations} used to scan. Must be a subset of the user's authorization. Defaults to the empty set.
*
* @param job
* the Hadoop job instance to be configured
* @param auths
* the user's authorizations
* @since 1.5.0
*/
public static void setScanAuthorizations(JobConf job, Authorizations auths) {
InputConfigurator.setScanAuthorizations(CLASS, job, auths);
}
/**
* Gets the authorizations to set for the scans from the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return the Accumulo scan authorizations
* @since 1.5.0
* @see #setScanAuthorizations(JobConf, Authorizations)
*/
protected static Authorizations getScanAuthorizations(JobConf job) {
return InputConfigurator.getScanAuthorizations(CLASS, job);
}
/**
* Sets the input ranges to scan for this job. If not set, the entire table will be scanned.
*
* @param job
* the Hadoop job instance to be configured
* @param ranges
* the ranges that will be mapped over
* @since 1.5.0
*/
public static void setRanges(JobConf job, Collection<Range> ranges) {
InputConfigurator.setRanges(CLASS, job, ranges);
}
/**
* Gets the ranges to scan over from a job.
*
* @param job
* the Hadoop context for the configured job
* @return the ranges
* @throws IOException
* if the ranges have been encoded improperly
* @since 1.5.0
* @see #setRanges(JobConf, Collection)
*/
protected static List<Range> getRanges(JobConf job) throws IOException {
return InputConfigurator.getRanges(CLASS, job);
}
/**
* Restricts the columns that will be mapped over for this job.
*
* @param job
* the Hadoop job instance to be configured
* @param columnFamilyColumnQualifierPairs
* a pair of {@link Text} objects corresponding to column family and column qualifier. If the column qualifier is null, the entire column family is
* selected. An empty set is the default and is equivalent to scanning the all columns.
* @since 1.5.0
*/
public static void fetchColumns(JobConf job, Collection<Pair<Text,Text>> columnFamilyColumnQualifierPairs) {
InputConfigurator.fetchColumns(CLASS, job, columnFamilyColumnQualifierPairs);
}
/**
* Gets the columns to be mapped over from this job.
*
* @param job
* the Hadoop context for the configured job
* @return a set of columns
* @since 1.5.0
* @see #fetchColumns(JobConf, Collection)
*/
protected static Set<Pair<Text,Text>> getFetchedColumns(JobConf job) {
return InputConfigurator.getFetchedColumns(CLASS, job);
}
/**
* Encode an iterator on the input for this job.
*
* @param job
* the Hadoop job instance to be configured
* @param cfg
* the configuration of the iterator
* @since 1.5.0
*/
public static void addIterator(JobConf job, IteratorSetting cfg) {
InputConfigurator.addIterator(CLASS, job, cfg);
}
/**
* Gets a list of the iterator settings (for iterators to apply to a scanner) from this configuration.
*
* @param job
* the Hadoop context for the configured job
* @return a list of iterators
* @since 1.5.0
* @see #addIterator(JobConf, IteratorSetting)
*/
protected static List<IteratorSetting> getIterators(JobConf job) {
return InputConfigurator.getIterators(CLASS, job);
}
/**
* Controls the automatic adjustment of ranges for this job. This feature merges overlapping ranges, then splits them to align with tablet boundaries.
* Disabling this feature will cause exactly one Map task to be created for each specified range. The default setting is enabled. *
*
* <p>
* By default, this feature is <b>enabled</b>.
*
* @param job
* the Hadoop job instance to be configured
* @param enableFeature
* the feature is enabled if true, disabled otherwise
* @see #setRanges(JobConf, Collection)
* @since 1.5.0
*/
public static void setAutoAdjustRanges(JobConf job, boolean enableFeature) {
InputConfigurator.setAutoAdjustRanges(CLASS, job, enableFeature);
}
/**
* Determines whether a configuration has auto-adjust ranges enabled.
*
* @param job
* the Hadoop context for the configured job
* @return false if the feature is disabled, true otherwise
* @since 1.5.0
* @see #setAutoAdjustRanges(JobConf, boolean)
*/
protected static boolean getAutoAdjustRanges(JobConf job) {
return InputConfigurator.getAutoAdjustRanges(CLASS, job);
}
/**
* Controls the use of the {@link IsolatedScanner} in this job.
*
* <p>
* By default, this feature is <b>disabled</b>.
*
* @param job
* the Hadoop job instance to be configured
* @param enableFeature
* the feature is enabled if true, disabled otherwise
* @since 1.5.0
*/
public static void setScanIsolation(JobConf job, boolean enableFeature) {
InputConfigurator.setScanIsolation(CLASS, job, enableFeature);
}
/**
* Determines whether a configuration has isolation enabled.
*
* @param job
* the Hadoop context for the configured job
* @return true if the feature is enabled, false otherwise
* @since 1.5.0
* @see #setScanIsolation(JobConf, boolean)
*/
protected static boolean isIsolated(JobConf job) {
return InputConfigurator.isIsolated(CLASS, job);
}
/**
* Controls the use of the {@link ClientSideIteratorScanner} in this job. Enabling this feature will cause the iterator stack to be constructed within the Map
* task, rather than within the Accumulo TServer. To use this feature, all classes needed for those iterators must be available on the classpath for the task.
*
* <p>
* By default, this feature is <b>disabled</b>.
*
* @param job
* the Hadoop job instance to be configured
* @param enableFeature
* the feature is enabled if true, disabled otherwise
* @since 1.5.0
*/
public static void setLocalIterators(JobConf job, boolean enableFeature) {
InputConfigurator.setLocalIterators(CLASS, job, enableFeature);
}
/**
* Determines whether a configuration uses local iterators.
*
* @param job
* the Hadoop context for the configured job
* @return true if the feature is enabled, false otherwise
* @since 1.5.0
* @see #setLocalIterators(JobConf, boolean)
*/
protected static boolean usesLocalIterators(JobConf job) {
return InputConfigurator.usesLocalIterators(CLASS, job);
}
/**
* <p>
* Enable reading offline tables. By default, this feature is disabled and only online tables are scanned. This will make the map reduce job directly read the
* table's files. If the table is not offline, then the job will fail. If the table comes online during the map reduce job, it is likely that the job will
* fail.
*
* <p>
* To use this option, the map reduce user will need access to read the Accumulo directory in HDFS.
*
* <p>
* Reading the offline table will create the scan time iterator stack in the map process. So any iterators that are configured for the table will need to be
* on the mapper's classpath.
*
* <p>
* One way to use this feature is to clone a table, take the clone offline, and use the clone as the input table for a map reduce job. If you plan to map
* reduce over the data many times, it may be better to the compact the table, clone it, take it offline, and use the clone for all map reduce jobs. The
* reason to do this is that compaction will reduce each tablet in the table to one file, and it is faster to read from one file.
*
* <p>
* There are two possible advantages to reading a tables file directly out of HDFS. First, you may see better read performance. Second, it will support
* speculative execution better. When reading an online table speculative execution can put more load on an already slow tablet server.
*
* <p>
* By default, this feature is <b>disabled</b>.
*
* @param job
* the Hadoop job instance to be configured
* @param enableFeature
* the feature is enabled if true, disabled otherwise
* @since 1.5.0
*/
public static void setOfflineTableScan(JobConf job, boolean enableFeature) {
InputConfigurator.setOfflineTableScan(CLASS, job, enableFeature);
}
/**
* Determines whether a configuration has the offline table scan feature enabled.
*
* @param job
* the Hadoop context for the configured job
* @return true if the feature is enabled, false otherwise
* @since 1.5.0
* @see #setOfflineTableScan(JobConf, boolean)
*/
protected static boolean isOfflineScan(JobConf job) {
return InputConfigurator.isOfflineScan(CLASS, job);
}
/**
* Initializes an Accumulo {@link TabletLocator} based on the configuration.
*
* @param job
* the Hadoop context for the configured job
* @return an Accumulo tablet locator
* @throws TableNotFoundException
* if the table name set on the configuration doesn't exist
* @since 1.5.0
*/
protected static TabletLocator getTabletLocator(JobConf job) throws TableNotFoundException {
return InputConfigurator.getTabletLocator(CLASS, job);
}
// InputFormat doesn't have the equivalent of OutputFormat's checkOutputSpecs(JobContext job)
/**
* Check whether a configuration is fully configured to be used with an Accumulo {@link org.apache.hadoop.mapreduce.InputFormat}.
*
* @param job
* the Hadoop context for the configured job
* @throws IOException
* if the context is improperly configured
* @since 1.5.0
*/
protected static void validateOptions(JobConf job) throws IOException {
InputConfigurator.validateOptions(CLASS, job);
}
/**
* An abstract base class to be used to create {@link RecordReader} instances that convert from Accumulo {@link Key}/{@link Value} pairs to the user's K/V
* types.
*
* Subclasses must implement {@link #next(Object, Object)} to update key and value, and also to update the following variables:
* <ul>
* <li>Key {@link #currentKey} (used for progress reporting)</li>
* <li>int {@link #numKeysRead} (used for progress reporting)</li>
* </ul>
*/
protected abstract static class RecordReaderBase<K,V> implements RecordReader<K,V> {
protected long numKeysRead;
protected Iterator<Entry<Key,Value>> scannerIterator;
protected RangeInputSplit split;
/**
* Apply the configured iterators from the configuration to the scanner.
*
* @param scanner
* the scanner to configure
*/
protected void setupIterators(List<IteratorSetting> iterators, Scanner scanner) {
for (IteratorSetting iterator : iterators) {
scanner.addScanIterator(iterator);
}
}
/**
* Initialize a scanner over the given input split using this task attempt configuration.
*/
public void initialize(InputSplit inSplit, JobConf job) throws IOException {
Scanner scanner;
split = (RangeInputSplit) inSplit;
log.debug("Initializing input split: " + split.getRange());
Instance instance = split.getInstance();
if (null == instance) {
instance = getInstance(job);
}
String principal = split.getPrincipal();
if (null == principal) {
principal = getPrincipal(job);
}
AuthenticationToken token = split.getToken();
if (null == token) {
String tokenClass = getTokenClass(job);
byte[] tokenBytes = getToken(job);
try {
token = CredentialHelper.extractToken(tokenClass, tokenBytes);
} catch (AccumuloSecurityException e) {
throw new IOException(e);
}
}
Authorizations authorizations = split.getAuths();
if (null == authorizations) {
authorizations = getScanAuthorizations(job);
}
String table = split.getTable();
if (null == table) {
table = getInputTableName(job);
}
Boolean isOffline = split.isOffline();
if (null == isOffline) {
isOffline = isOfflineScan(job);
}
Boolean isIsolated = split.isIsolatedScan();
if (null == isIsolated) {
isIsolated = isIsolated(job);
}
Boolean usesLocalIterators = split.usesLocalIterators();
if (null == usesLocalIterators) {
usesLocalIterators = usesLocalIterators(job);
}
List<IteratorSetting> iterators = split.getIterators();
if (null == iterators) {
iterators = getIterators(job);
}
Set<Pair<Text,Text>> columns = split.getFetchedColumns();
if (null == columns) {
columns = getFetchedColumns(job);
}
try {
log.debug("Creating connector with user: " + principal);
Connector conn = instance.getConnector(principal, token);
log.debug("Creating scanner for table: " + table);
log.debug("Authorizations are: " + authorizations);
if (isOffline) {
String tokenClass = token.getClass().getCanonicalName();
ByteBuffer tokenBuffer = ByteBuffer.wrap(CredentialHelper.toBytes(token));
scanner = new OfflineScanner(instance, new TCredentials(principal, tokenClass, tokenBuffer, instance.getInstanceID()), Tables.getTableId(instance,
table), authorizations);
} else {
scanner = conn.createScanner(table, authorizations);
}
if (isIsolated) {
log.info("Creating isolated scanner");
scanner = new IsolatedScanner(scanner);
}
if (usesLocalIterators) {
log.info("Using local iterators");
scanner = new ClientSideIteratorScanner(scanner);
}
setupIterators(iterators, scanner);
} catch (Exception e) {
throw new IOException(e);
}
// setup a scanner within the bounds of this split
for (Pair<Text,Text> c : columns) {
if (c.getSecond() != null) {
log.debug("Fetching column " + c.getFirst() + ":" + c.getSecond());
scanner.fetchColumn(c.getFirst(), c.getSecond());
} else {
log.debug("Fetching column family " + c.getFirst());
scanner.fetchColumnFamily(c.getFirst());
}
}
scanner.setRange(split.getRange());
numKeysRead = 0;
// do this last after setting all scanner options
scannerIterator = scanner.iterator();
}
@Override
public void close() {}
@Override
public long getPos() throws IOException {
return numKeysRead;
}
@Override
public float getProgress() throws IOException {
if (numKeysRead > 0 && currentKey == null)
return 1.0f;
return split.getProgress(currentKey);
}
protected Key currentKey = null;
}
Map<String,Map<KeyExtent,List<Range>>> binOfflineTable(JobConf job, String tableName, List<Range> ranges) throws TableNotFoundException, AccumuloException,
AccumuloSecurityException {
Map<String,Map<KeyExtent,List<Range>>> binnedRanges = new HashMap<String,Map<KeyExtent,List<Range>>>();
Instance instance = getInstance(job);
Connector conn = instance.getConnector(getPrincipal(job), CredentialHelper.extractToken(getTokenClass(job), getToken(job)));
String tableId = Tables.getTableId(instance, tableName);
if (Tables.getTableState(instance, tableId) != TableState.OFFLINE) {
Tables.clearCache(instance);
if (Tables.getTableState(instance, tableId) != TableState.OFFLINE) {
throw new AccumuloException("Table is online " + tableName + "(" + tableId + ") cannot scan table in offline mode ");
}
}
for (Range range : ranges) {
Text startRow;
if (range.getStartKey() != null)
startRow = range.getStartKey().getRow();
else
startRow = new Text();
Range metadataRange = new Range(new KeyExtent(new Text(tableId), startRow, null).getMetadataEntry(), true, null, false);
Scanner scanner = conn.createScanner(Constants.METADATA_TABLE_NAME, Constants.NO_AUTHS);
Constants.METADATA_PREV_ROW_COLUMN.fetch(scanner);
scanner.fetchColumnFamily(Constants.METADATA_LAST_LOCATION_COLUMN_FAMILY);
scanner.fetchColumnFamily(Constants.METADATA_CURRENT_LOCATION_COLUMN_FAMILY);
scanner.fetchColumnFamily(Constants.METADATA_FUTURE_LOCATION_COLUMN_FAMILY);
scanner.setRange(metadataRange);
RowIterator rowIter = new RowIterator(scanner);
KeyExtent lastExtent = null;
while (rowIter.hasNext()) {
Iterator<Entry<Key,Value>> row = rowIter.next();
String last = "";
KeyExtent extent = null;
String location = null;
while (row.hasNext()) {
Entry<Key,Value> entry = row.next();
Key key = entry.getKey();
if (key.getColumnFamily().equals(Constants.METADATA_LAST_LOCATION_COLUMN_FAMILY)) {
last = entry.getValue().toString();
}
if (key.getColumnFamily().equals(Constants.METADATA_CURRENT_LOCATION_COLUMN_FAMILY)
|| key.getColumnFamily().equals(Constants.METADATA_FUTURE_LOCATION_COLUMN_FAMILY)) {
location = entry.getValue().toString();
}
if (Constants.METADATA_PREV_ROW_COLUMN.hasColumns(key)) {
extent = new KeyExtent(key.getRow(), entry.getValue());
}
}
if (location != null)
return null;
if (!extent.getTableId().toString().equals(tableId)) {
throw new AccumuloException("Saw unexpected table Id " + tableId + " " + extent);
}
if (lastExtent != null && !extent.isPreviousExtent(lastExtent)) {
throw new AccumuloException(" " + lastExtent + " is not previous extent " + extent);
}
Map<KeyExtent,List<Range>> tabletRanges = binnedRanges.get(last);
if (tabletRanges == null) {
tabletRanges = new HashMap<KeyExtent,List<Range>>();
binnedRanges.put(last, tabletRanges);
}
List<Range> rangeList = tabletRanges.get(extent);
if (rangeList == null) {
rangeList = new ArrayList<Range>();
tabletRanges.put(extent, rangeList);
}
rangeList.add(range);
if (extent.getEndRow() == null || range.afterEndKey(new Key(extent.getEndRow()).followingKey(PartialKey.ROW))) {
break;
}
lastExtent = extent;
}
}
return binnedRanges;
}
/**
* Read the metadata table to get tablets and match up ranges to them.
*/
@Override
public InputSplit[] getSplits(JobConf job, int numSplits) throws IOException {
Level logLevel = getLogLevel(job);
log.setLevel(logLevel);
validateOptions(job);
String tableName = getInputTableName(job);
boolean autoAdjust = getAutoAdjustRanges(job);
List<Range> ranges = autoAdjust ? Range.mergeOverlapping(getRanges(job)) : getRanges(job);
Instance instance = getInstance(job);
boolean offline = isOfflineScan(job);
boolean isolated = isIsolated(job);
boolean localIterators = usesLocalIterators(job);
boolean mockInstance = (null != instance && MockInstance.class.equals(instance.getClass()));
Set<Pair<Text,Text>> fetchedColumns = getFetchedColumns(job);
Authorizations auths = getScanAuthorizations(job);
String principal = getPrincipal(job);
String tokenClass = getTokenClass(job);
byte[] tokenBytes = getToken(job);
AuthenticationToken token;
try {
token = CredentialHelper.extractToken(tokenClass, tokenBytes);
} catch (AccumuloSecurityException e) {
throw new IOException(e);
}
List<IteratorSetting> iterators = getIterators(job);
if (ranges.isEmpty()) {
ranges = new ArrayList<Range>(1);
ranges.add(new Range());
}
// get the metadata information for these ranges
Map<String,Map<KeyExtent,List<Range>>> binnedRanges = new HashMap<String,Map<KeyExtent,List<Range>>>();
TabletLocator tl;
try {
if (isOfflineScan(job)) {
binnedRanges = binOfflineTable(job, tableName, ranges);
while (binnedRanges == null) {
// Some tablets were still online, try again
UtilWaitThread.sleep(100 + (int) (Math.random() * 100)); // sleep randomly between 100 and 200 ms
binnedRanges = binOfflineTable(job, tableName, ranges);
}
} else {
String tableId = null;
tl = getTabletLocator(job);
// its possible that the cache could contain complete, but old information about a tables tablets... so clear it
tl.invalidateCache();
while (!tl.binRanges(ranges, binnedRanges, new TCredentials(principal, tokenClass, ByteBuffer.wrap(tokenBytes), instance.getInstanceID())).isEmpty()) {
if (!(instance instanceof MockInstance)) {
if (tableId == null)
tableId = Tables.getTableId(instance, tableName);
if (!Tables.exists(instance, tableId))
throw new TableDeletedException(tableId);
if (Tables.getTableState(instance, tableId) == TableState.OFFLINE)
throw new TableOfflineException(instance, tableId);
}
binnedRanges.clear();
log.warn("Unable to locate bins for specified ranges. Retrying.");
UtilWaitThread.sleep(100 + (int) (Math.random() * 100)); // sleep randomly between 100 and 200 ms
tl.invalidateCache();
}
}
} catch (Exception e) {
throw new IOException(e);
}
ArrayList<RangeInputSplit> splits = new ArrayList<RangeInputSplit>(ranges.size());
HashMap<Range,ArrayList<String>> splitsToAdd = null;
if (!autoAdjust)
splitsToAdd = new HashMap<Range,ArrayList<String>>();
HashMap<String,String> hostNameCache = new HashMap<String,String>();
for (Entry<String,Map<KeyExtent,List<Range>>> tserverBin : binnedRanges.entrySet()) {
String ip = tserverBin.getKey().split(":", 2)[0];
String location = hostNameCache.get(ip);
if (location == null) {
InetAddress inetAddress = InetAddress.getByName(ip);
location = inetAddress.getHostName();
hostNameCache.put(ip, location);
}
for (Entry<KeyExtent,List<Range>> extentRanges : tserverBin.getValue().entrySet()) {
Range ke = extentRanges.getKey().toDataRange();
for (Range r : extentRanges.getValue()) {
if (autoAdjust) {
// divide ranges into smaller ranges, based on the tablets
splits.add(new RangeInputSplit(ke.clip(r), new String[] {location}));
} else {
// don't divide ranges
ArrayList<String> locations = splitsToAdd.get(r);
if (locations == null)
locations = new ArrayList<String>(1);
locations.add(location);
splitsToAdd.put(r, locations);
}
}
}
}
if (!autoAdjust)
for (Entry<Range,ArrayList<String>> entry : splitsToAdd.entrySet())
splits.add(new RangeInputSplit(entry.getKey(), entry.getValue().toArray(new String[0])));
for (RangeInputSplit split : splits) {
split.setTable(tableName);
split.setOffline(offline);
split.setIsolatedScan(isolated);
split.setUsesLocalIterators(localIterators);
split.setMockInstance(mockInstance);
split.setFetchedColumns(fetchedColumns);
split.setPrincipal(principal);
split.setToken(token);
split.setInstanceName(instance.getInstanceName());
split.setZooKeepers(instance.getZooKeepers());
split.setAuths(auths);
split.setIterators(iterators);
split.setLogLevel(logLevel);
}
return splits.toArray(new InputSplit[splits.size()]);
}
}