Repository: accumulo Updated Branches: refs/heads/1.7 0bbc67594 -> 83d1b8388
ACCUMULO-3756 Remove extension of impl type by API type Project: http://git-wip-us.apache.org/repos/asf/accumulo/repo Commit: http://git-wip-us.apache.org/repos/asf/accumulo/commit/83d1b838 Tree: http://git-wip-us.apache.org/repos/asf/accumulo/tree/83d1b838 Diff: http://git-wip-us.apache.org/repos/asf/accumulo/diff/83d1b838 Branch: refs/heads/1.7 Commit: 83d1b8388ad807d678c9a3a922e5025faa9a5933 Parents: 0bbc675 Author: Keith Turner <ke...@deenlo.com> Authored: Thu Apr 30 19:37:02 2015 -0400 Committer: Keith Turner <ktur...@apache.org> Committed: Mon May 4 15:56:27 2015 -0400 ---------------------------------------------------------------------- .../core/client/mapred/AbstractInputFormat.java | 83 ++-- .../core/client/mapred/AccumuloInputFormat.java | 4 +- .../client/mapreduce/AbstractInputFormat.java | 78 ++-- .../client/mapreduce/AccumuloInputFormat.java | 5 +- .../core/client/mapreduce/RangeInputSplit.java | 411 +++++++++++++++-- .../mapreduce/impl/AccumuloInputSplit.java | 436 ------------------- .../client/mapreduce/impl/BatchInputSplit.java | 47 +- .../core/client/mapreduce/impl/SplitUtils.java | 43 +- 8 files changed, 540 insertions(+), 567 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapred/AbstractInputFormat.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapred/AbstractInputFormat.java b/core/src/main/java/org/apache/accumulo/core/client/mapred/AbstractInputFormat.java index f2e3a79..cb3ef11 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapred/AbstractInputFormat.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapred/AbstractInputFormat.java @@ -53,7 +53,6 @@ import org.apache.accumulo.core.client.impl.Tables; import org.apache.accumulo.core.client.impl.TabletLocator; import org.apache.accumulo.core.client.mapred.impl.BatchInputSplit; import org.apache.accumulo.core.client.mapreduce.InputTableConfig; -import org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit; import org.apache.accumulo.core.client.mapreduce.impl.SplitUtils; import org.apache.accumulo.core.client.mapreduce.lib.impl.ConfiguratorBase; import org.apache.accumulo.core.client.mapreduce.lib.impl.InputConfigurator; @@ -396,7 +395,7 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { protected long numKeysRead; protected Iterator<Map.Entry<Key,Value>> scannerIterator; protected RangeInputSplit split; - private org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit aiSplit; + private org.apache.accumulo.core.client.mapreduce.RangeInputSplit baseSplit; protected ScannerBase scannerBase; @@ -423,7 +422,7 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { * the table name for which the scanner is configured * @since 1.7.0 */ - private void setupIterators(JobConf job, ScannerBase scanner, String tableName, AccumuloInputSplit split) { + private void setupIterators(JobConf job, ScannerBase scanner, String tableName, org.apache.accumulo.core.client.mapreduce.RangeInputSplit split) { List<IteratorSetting> iterators = null; if (null == split) { @@ -453,60 +452,75 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { */ @Deprecated protected void setupIterators(JobConf job, Scanner scanner, String tableName, RangeInputSplit split) { - setupIterators(job, (ScannerBase) scanner, tableName, (AccumuloInputSplit) split); + setupIterators(job, (ScannerBase) scanner, tableName, split); } /** * Initialize a scanner over the given input split using this task attempt configuration. */ public void initialize(InputSplit inSplit, JobConf job) throws IOException { - aiSplit = (AccumuloInputSplit) inSplit; - log.debug("Initializing input split: " + aiSplit.toString()); + baseSplit = (org.apache.accumulo.core.client.mapreduce.RangeInputSplit) inSplit; + log.debug("Initializing input split: " + baseSplit.toString()); - Instance instance = aiSplit.getInstance(getClientConfiguration(job)); + Instance instance = baseSplit.getInstance(getClientConfiguration(job)); if (null == instance) { instance = getInstance(job); } - String principal = aiSplit.getPrincipal(); + String principal = baseSplit.getPrincipal(); if (null == principal) { principal = getPrincipal(job); } - AuthenticationToken token = aiSplit.getToken(); + AuthenticationToken token = baseSplit.getToken(); if (null == token) { token = getAuthenticationToken(job); } - Authorizations authorizations = aiSplit.getAuths(); + Authorizations authorizations = baseSplit.getAuths(); if (null == authorizations) { authorizations = getScanAuthorizations(job); } - String table = aiSplit.getTableName(); + String table = baseSplit.getTableName(); // in case the table name changed, we can still use the previous name for terms of configuration, // but the scanner will use the table id resolved at job setup time - InputTableConfig tableConfig = getInputTableConfig(job, aiSplit.getTableName()); + InputTableConfig tableConfig = getInputTableConfig(job, baseSplit.getTableName()); log.debug("Creating connector with user: " + principal); log.debug("Creating scanner for table: " + table); log.debug("Authorizations are: " + authorizations); - if (aiSplit instanceof RangeInputSplit) { - RangeInputSplit rangeSplit = (RangeInputSplit) aiSplit; - split = rangeSplit; - Boolean isOffline = rangeSplit.isOffline(); + if (baseSplit instanceof BatchInputSplit) { + BatchScanner scanner; + BatchInputSplit multiRangeSplit = (BatchInputSplit) baseSplit; + + try{ + // Note: BatchScanner will use at most one thread per tablet, currently BatchInputSplit will not span tablets + int scanThreads = 1; + scanner = instance.getConnector(principal, token).createBatchScanner(baseSplit.getTableName(), authorizations, scanThreads); + setupIterators(job, scanner, baseSplit.getTableName(), baseSplit); + } catch (Exception e) { + throw new IOException(e); + } + + scanner.setRanges(multiRangeSplit.getRanges()); + scannerBase = scanner; + + } else if (baseSplit instanceof RangeInputSplit) { + split = (RangeInputSplit) baseSplit; + Boolean isOffline = baseSplit.isOffline(); if (null == isOffline) { isOffline = tableConfig.isOfflineScan(); } - Boolean isIsolated = rangeSplit.isIsolatedScan(); + Boolean isIsolated = baseSplit.isIsolatedScan(); if (null == isIsolated) { isIsolated = tableConfig.shouldUseIsolatedScanners(); } - Boolean usesLocalIterators = rangeSplit.usesLocalIterators(); + Boolean usesLocalIterators = baseSplit.usesLocalIterators(); if (null == usesLocalIterators) { usesLocalIterators = tableConfig.shouldUseLocalIterators(); } @@ -515,13 +529,13 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { try { if (isOffline) { - scanner = new OfflineScanner(instance, new Credentials(principal, token), aiSplit.getTableId(), authorizations); + scanner = new OfflineScanner(instance, new Credentials(principal, token), baseSplit.getTableId(), authorizations); } else if (instance instanceof MockInstance) { - scanner = instance.getConnector(principal, token).createScanner(aiSplit.getTableName(), authorizations); + scanner = instance.getConnector(principal, token).createScanner(baseSplit.getTableName(), authorizations); } else { ClientConfiguration clientConf = getClientConfiguration(job); ClientContext context = new ClientContext(instance, new Credentials(principal, token), clientConf); - scanner = new ScannerImpl(context, aiSplit.getTableId(), authorizations); + scanner = new ScannerImpl(context, baseSplit.getTableId(), authorizations); } if (isIsolated) { log.info("Creating isolated scanner"); @@ -531,35 +545,18 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { log.info("Using local iterators"); scanner = new ClientSideIteratorScanner(scanner); } - setupIterators(job, scanner, aiSplit.getTableName(), aiSplit); + setupIterators(job, scanner, baseSplit.getTableName(), baseSplit); } catch (Exception e) { throw new IOException(e); } - scanner.setRange(rangeSplit.getRange()); + scanner.setRange(baseSplit.getRange()); scannerBase = scanner; - - } else if (aiSplit instanceof BatchInputSplit) { - BatchScanner scanner; - BatchInputSplit multiRangeSplit = (BatchInputSplit) aiSplit; - - try{ - // Note: BatchScanner will use at most one thread per tablet, currently BatchInputSplit will not span tablets - int scanThreads = 1; - scanner = instance.getConnector(principal, token).createBatchScanner(aiSplit.getTableName(), authorizations, scanThreads); - setupIterators(job, scanner, aiSplit.getTableName(), aiSplit); - } catch (Exception e) { - throw new IOException(e); - } - - scanner.setRanges(multiRangeSplit.getRanges()); - scannerBase = scanner; - } else { - throw new IllegalArgumentException("Can not initialize from " + aiSplit.getClass().toString()); + throw new IllegalArgumentException("Can not initialize from " + baseSplit.getClass().toString()); } - Collection<Pair<Text,Text>> columns = aiSplit.getFetchedColumns(); + Collection<Pair<Text,Text>> columns = baseSplit.getFetchedColumns(); if (null == columns) { columns = tableConfig.getFetchedColumns(); } @@ -595,7 +592,7 @@ public abstract class AbstractInputFormat<K,V> implements InputFormat<K,V> { public float getProgress() throws IOException { if (numKeysRead > 0 && currentKey == null) return 1.0f; - return aiSplit.getProgress(currentKey); + return baseSplit.getProgress(currentKey); } protected Key currentKey = null; http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapred/AccumuloInputFormat.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapred/AccumuloInputFormat.java b/core/src/main/java/org/apache/accumulo/core/client/mapred/AccumuloInputFormat.java index 2c23552..856a11a 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapred/AccumuloInputFormat.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapred/AccumuloInputFormat.java @@ -54,8 +54,8 @@ public class AccumuloInputFormat extends InputFormatBase<Key,Value> { log.setLevel(getLogLevel(job)); // Override the log level from the configuration as if the RangeInputSplit has one it's the more correct one to use. - if (split instanceof org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit) { - org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit accSplit = (org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit) split; + if (split instanceof org.apache.accumulo.core.client.mapreduce.RangeInputSplit) { + org.apache.accumulo.core.client.mapreduce.RangeInputSplit accSplit = (org.apache.accumulo.core.client.mapreduce.RangeInputSplit) split; Level level = accSplit.getLogLevel(); if (null != level) { log.setLevel(level); http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AbstractInputFormat.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AbstractInputFormat.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AbstractInputFormat.java index d402bb0..b2f1035 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AbstractInputFormat.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AbstractInputFormat.java @@ -51,7 +51,6 @@ import org.apache.accumulo.core.client.impl.OfflineScanner; import org.apache.accumulo.core.client.impl.ScannerImpl; import org.apache.accumulo.core.client.impl.Tables; import org.apache.accumulo.core.client.impl.TabletLocator; -import org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit; import org.apache.accumulo.core.client.mapreduce.impl.BatchInputSplit; import org.apache.accumulo.core.client.mapreduce.impl.SplitUtils; import org.apache.accumulo.core.client.mapreduce.lib.impl.ConfiguratorBase; @@ -427,7 +426,6 @@ public abstract class AbstractInputFormat<K,V> extends InputFormat<K,V> { protected Iterator<Map.Entry<Key,Value>> scannerIterator; protected ScannerBase scannerBase; protected RangeInputSplit split; - private AccumuloInputSplit aiSplit; /** * Extracts Iterators settings from the context to be used by RecordReader. @@ -455,7 +453,7 @@ public abstract class AbstractInputFormat<K,V> extends InputFormat<K,V> { * InputSplit containing configurations * @since 1.7.0 */ - private void setupIterators(TaskAttemptContext context, ScannerBase scanner, String tableName, AccumuloInputSplit split) { + private void setupIterators(TaskAttemptContext context, ScannerBase scanner, String tableName, RangeInputSplit split) { List<IteratorSetting> iterators = null; if (null == split) { @@ -485,74 +483,88 @@ public abstract class AbstractInputFormat<K,V> extends InputFormat<K,V> { */ @Deprecated protected void setupIterators(TaskAttemptContext context, Scanner scanner, String tableName, RangeInputSplit split) { - setupIterators(context, (ScannerBase) scanner, tableName, (AccumuloInputSplit) split); + setupIterators(context, (ScannerBase) scanner, tableName, split); } @Override public void initialize(InputSplit inSplit, TaskAttemptContext attempt) throws IOException { - aiSplit = (AccumuloInputSplit) inSplit; - log.debug("Initializing input split: " + aiSplit.toString()); + split = (RangeInputSplit) inSplit; + log.debug("Initializing input split: " + split.toString()); - Instance instance = aiSplit.getInstance(getClientConfiguration(attempt)); + Instance instance = split.getInstance(getClientConfiguration(attempt)); if (null == instance) { instance = getInstance(attempt); } - String principal = aiSplit.getPrincipal(); + String principal = split.getPrincipal(); if (null == principal) { principal = getPrincipal(attempt); } - AuthenticationToken token = aiSplit.getToken(); + AuthenticationToken token = split.getToken(); if (null == token) { token = getAuthenticationToken(attempt); } - Authorizations authorizations = aiSplit.getAuths(); + Authorizations authorizations = split.getAuths(); if (null == authorizations) { authorizations = getScanAuthorizations(attempt); } - String table = aiSplit.getTableName(); + String table = split.getTableName(); // in case the table name changed, we can still use the previous name for terms of configuration, // but the scanner will use the table id resolved at job setup time - InputTableConfig tableConfig = getInputTableConfig(attempt, aiSplit.getTableName()); + InputTableConfig tableConfig = getInputTableConfig(attempt, split.getTableName()); log.debug("Creating connector with user: " + principal); log.debug("Creating scanner for table: " + table); log.debug("Authorizations are: " + authorizations); - if (aiSplit instanceof RangeInputSplit) { - RangeInputSplit rangeSplit = (RangeInputSplit) aiSplit; - split = rangeSplit; + if (split instanceof BatchInputSplit) { + BatchInputSplit batchSplit = (BatchInputSplit) split; + + BatchScanner scanner; + try{ + // Note: BatchScanner will use at most one thread per tablet, currently BatchInputSplit will not span tablets + int scanThreads = 1; + scanner = instance.getConnector(principal, token).createBatchScanner(split.getTableName(), authorizations, scanThreads); + setupIterators(attempt, scanner, split.getTableName(), split); + } catch (Exception e) { + e.printStackTrace(); + throw new IOException(e); + } + + scanner.setRanges(batchSplit.getRanges()); + scannerBase = scanner; + } else { Scanner scanner; - Boolean isOffline = rangeSplit.isOffline(); + Boolean isOffline = split.isOffline(); if (null == isOffline) { isOffline = tableConfig.isOfflineScan(); } - Boolean isIsolated = rangeSplit.isIsolatedScan(); + Boolean isIsolated = split.isIsolatedScan(); if (null == isIsolated) { isIsolated = tableConfig.shouldUseIsolatedScanners(); } - Boolean usesLocalIterators = rangeSplit.usesLocalIterators(); + Boolean usesLocalIterators = split.usesLocalIterators(); if (null == usesLocalIterators) { usesLocalIterators = tableConfig.shouldUseLocalIterators(); } try { if (isOffline) { - scanner = new OfflineScanner(instance, new Credentials(principal, token), aiSplit.getTableId(), authorizations); + scanner = new OfflineScanner(instance, new Credentials(principal, token), split.getTableId(), authorizations); } else if (instance instanceof MockInstance) { - scanner = instance.getConnector(principal, token).createScanner(aiSplit.getTableName(), authorizations); + scanner = instance.getConnector(principal, token).createScanner(split.getTableName(), authorizations); } else { ClientConfiguration clientConf = getClientConfiguration(attempt); ClientContext context = new ClientContext(instance, new Credentials(principal, token), clientConf); - scanner = new ScannerImpl(context, aiSplit.getTableId(), authorizations); + scanner = new ScannerImpl(context, split.getTableId(), authorizations); } if (isIsolated) { log.info("Creating isolated scanner"); @@ -563,33 +575,17 @@ public abstract class AbstractInputFormat<K,V> extends InputFormat<K,V> { scanner = new ClientSideIteratorScanner(scanner); } - setupIterators(attempt, scanner, aiSplit.getTableName(), aiSplit); + setupIterators(attempt, scanner, split.getTableName(), split); } catch (Exception e) { throw new IOException(e); } - scanner.setRange(rangeSplit.getRange()); + scanner.setRange(split.getRange()); scannerBase = scanner; - } else if (aiSplit instanceof BatchInputSplit) { - BatchInputSplit batchSplit = (BatchInputSplit) aiSplit; - - BatchScanner scanner; - try{ - // Note: BatchScanner will use at most one thread per tablet, currently BatchInputSplit will not span tablets - int scanThreads = 1; - scanner = instance.getConnector(principal, token).createBatchScanner(aiSplit.getTableName(), authorizations, scanThreads); - setupIterators(attempt, scanner, aiSplit.getTableName(), aiSplit); - } catch (Exception e) { - e.printStackTrace(); - throw new IOException(e); - } - - scanner.setRanges(batchSplit.getRanges()); - scannerBase = scanner; } - Collection<Pair<Text,Text>> columns = aiSplit.getFetchedColumns(); + Collection<Pair<Text,Text>> columns = split.getFetchedColumns(); if (null == columns) { columns = tableConfig.getFetchedColumns(); } @@ -620,7 +616,7 @@ public abstract class AbstractInputFormat<K,V> extends InputFormat<K,V> { public float getProgress() throws IOException { if (numKeysRead > 0 && currentKey == null) return 1.0f; - return aiSplit.getProgress(currentKey); + return split.getProgress(currentKey); } /** http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AccumuloInputFormat.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AccumuloInputFormat.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AccumuloInputFormat.java index 7af5e66..45315db 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AccumuloInputFormat.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/AccumuloInputFormat.java @@ -21,7 +21,6 @@ import java.util.Map.Entry; import org.apache.accumulo.core.client.ClientConfiguration; import org.apache.accumulo.core.client.security.tokens.AuthenticationToken; -import org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit; import org.apache.accumulo.core.data.Key; import org.apache.accumulo.core.data.Value; import org.apache.accumulo.core.security.Authorizations; @@ -54,8 +53,8 @@ public class AccumuloInputFormat extends InputFormatBase<Key,Value> { log.setLevel(getLogLevel(context)); // Override the log level from the configuration as if the InputSplit has one it's the more correct one to use. - if (split instanceof AccumuloInputSplit) { - AccumuloInputSplit accSplit = (AccumuloInputSplit) split; + if (split instanceof org.apache.accumulo.core.client.mapreduce.RangeInputSplit) { + org.apache.accumulo.core.client.mapreduce.RangeInputSplit accSplit = (org.apache.accumulo.core.client.mapreduce.RangeInputSplit) split; Level level = accSplit.getLogLevel(); if (null != level) { log.setLevel(level); http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/RangeInputSplit.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/RangeInputSplit.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/RangeInputSplit.java index 9851192..b55e2cf 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/RangeInputSplit.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/RangeInputSplit.java @@ -19,50 +19,96 @@ package org.apache.accumulo.core.client.mapreduce; import java.io.DataInput; import java.io.DataOutput; import java.io.IOException; +import java.util.ArrayList; +import java.util.Arrays; +import java.util.Collection; +import java.util.HashSet; +import java.util.List; +import java.util.Set; -import org.apache.accumulo.core.client.mapreduce.impl.AccumuloInputSplit; +import org.apache.accumulo.core.client.ClientConfiguration; +import org.apache.accumulo.core.client.Instance; +import org.apache.accumulo.core.client.IteratorSetting; +import org.apache.accumulo.core.client.ZooKeeperInstance; import org.apache.accumulo.core.client.mapreduce.impl.SplitUtils; +import org.apache.accumulo.core.client.mapreduce.lib.impl.ConfiguratorBase.TokenSource; +import org.apache.accumulo.core.client.mapreduce.lib.impl.InputConfigurator; +import org.apache.accumulo.core.client.mock.MockInstance; +import org.apache.accumulo.core.client.security.tokens.AuthenticationToken; +import org.apache.accumulo.core.client.security.tokens.AuthenticationToken.AuthenticationTokenSerializer; +import org.apache.accumulo.core.data.ByteSequence; import org.apache.accumulo.core.data.Key; import org.apache.accumulo.core.data.PartialKey; import org.apache.accumulo.core.data.Range; +import org.apache.accumulo.core.security.Authorizations; +import org.apache.accumulo.core.util.Base64; +import org.apache.accumulo.core.util.Pair; +import org.apache.hadoop.io.Text; +import org.apache.hadoop.io.Writable; +import org.apache.hadoop.mapreduce.InputSplit; +import org.apache.log4j.Level; + +import static java.nio.charset.StandardCharsets.UTF_8; /** * The Class RangeInputSplit. Encapsulates an Accumulo range for use in Map Reduce jobs. */ -public class RangeInputSplit extends AccumuloInputSplit { +public class RangeInputSplit extends InputSplit implements Writable { private Range range; - private Boolean offline, isolatedScan, localIterators; + private String[] locations; + private String tableId, tableName, instanceName, zooKeepers, principal; + private TokenSource tokenSource; + private String tokenFile; + private AuthenticationToken token; + private Boolean offline, mockInstance, isolatedScan, localIterators; + private Authorizations auths; + private Set<Pair<Text,Text>> fetchedColumns; + private List<IteratorSetting> iterators; + private Level level; public RangeInputSplit() { range = new Range(); + locations = new String[0]; + tableName = ""; + tableId = ""; } public RangeInputSplit(RangeInputSplit split) throws IOException { - super(split); - this.setRange(split.getRange()); + this.range = split.getRange(); + this.setLocations(split.getLocations()); + this.setTableName(split.getTableName()); + this.setTableId(split.getTableId()); } protected RangeInputSplit(String table, String tableId, Range range, String[] locations) { - super(table, tableId, locations); this.range = range; + setLocations(locations); + this.tableName = table; + this.tableId = tableId; + } + + public Range getRange() { + return range; + } + + public static float getProgress(ByteSequence start, ByteSequence end, ByteSequence position) { + return SplitUtils.getProgress(start, end, position); } - @Override public float getProgress(Key currentKey) { if (currentKey == null) return 0f; - if (range.contains(currentKey)) { - // find the current range and report as if that is the single range + if(range.contains(currentKey)){ if (range.getStartKey() != null && range.getEndKey() != null) { if (range.getStartKey().compareTo(range.getEndKey(), PartialKey.ROW) != 0) { // just look at the row progress - return SplitUtils.getProgress(range.getStartKey().getRowData(), range.getEndKey().getRowData(), currentKey.getRowData()); + return getProgress(range.getStartKey().getRowData(), range.getEndKey().getRowData(), currentKey.getRowData()); } else if (range.getStartKey().compareTo(range.getEndKey(), PartialKey.ROW_COLFAM) != 0) { // just look at the column family progress - return SplitUtils.getProgress(range.getStartKey().getColumnFamilyData(), range.getEndKey().getColumnFamilyData(), currentKey.getColumnFamilyData()); + return getProgress(range.getStartKey().getColumnFamilyData(), range.getEndKey().getColumnFamilyData(), currentKey.getColumnFamilyData()); } else if (range.getStartKey().compareTo(range.getEndKey(), PartialKey.ROW_COLFAM_COLQUAL) != 0) { // just look at the column qualifier progress - return SplitUtils.getProgress(range.getStartKey().getColumnQualifierData(), range.getEndKey().getColumnQualifierData(), currentKey.getColumnQualifierData()); + return getProgress(range.getStartKey().getColumnQualifierData(), range.getEndKey().getColumnQualifierData(), currentKey.getColumnQualifierData()); } } } @@ -75,15 +121,23 @@ public class RangeInputSplit extends AccumuloInputSplit { */ @Override public long getLength() throws IOException { - return getRangeLength(range); + return SplitUtils.getRangeLength(range); } + @Override + public String[] getLocations() throws IOException { + return Arrays.copyOf(locations, locations.length); + } @Override public void readFields(DataInput in) throws IOException { - super.readFields(in); - range.readFields(in); + tableName = in.readUTF(); + tableId = in.readUTF(); + int numLocs = in.readInt(); + locations = new String[numLocs]; + for (int i = 0; i < numLocs; ++i) + locations[i] = in.readUTF(); if (in.readBoolean()) { isolatedScan = in.readBoolean(); @@ -96,13 +150,81 @@ public class RangeInputSplit extends AccumuloInputSplit { if (in.readBoolean()) { localIterators = in.readBoolean(); } + + if (in.readBoolean()) { + mockInstance = in.readBoolean(); + } + + if (in.readBoolean()) { + int numColumns = in.readInt(); + List<String> columns = new ArrayList<String>(numColumns); + for (int i = 0; i < numColumns; i++) { + columns.add(in.readUTF()); + } + + fetchedColumns = InputConfigurator.deserializeFetchedColumns(columns); + } + + if (in.readBoolean()) { + String strAuths = in.readUTF(); + auths = new Authorizations(strAuths.getBytes(UTF_8)); + } + + if (in.readBoolean()) { + principal = in.readUTF(); + } + + if (in.readBoolean()) { + int ordinal = in.readInt(); + this.tokenSource = TokenSource.values()[ordinal]; + + switch (this.tokenSource) { + case INLINE: + String tokenClass = in.readUTF(); + byte[] base64TokenBytes = in.readUTF().getBytes(UTF_8); + byte[] tokenBytes = Base64.decodeBase64(base64TokenBytes); + + this.token = AuthenticationTokenSerializer.deserialize(tokenClass, tokenBytes); + break; + + case FILE: + this.tokenFile = in.readUTF(); + + break; + default: + throw new IOException("Cannot parse unknown TokenSource ordinal"); + } + } + + if (in.readBoolean()) { + instanceName = in.readUTF(); + } + + if (in.readBoolean()) { + zooKeepers = in.readUTF(); + } + + if (in.readBoolean()) { + int numIterators = in.readInt(); + iterators = new ArrayList<IteratorSetting>(numIterators); + for (int i = 0; i < numIterators; i++) { + iterators.add(new IteratorSetting(in)); + } + } + + if (in.readBoolean()) { + level = Level.toLevel(in.readInt()); + } } @Override public void write(DataOutput out) throws IOException { - super.write(out); - range.write(out); + out.writeUTF(tableName); + out.writeUTF(tableId); + out.writeInt(locations.length); + for (int i = 0; i < locations.length; ++i) + out.writeUTF(locations[i]); out.writeBoolean(null != isolatedScan); if (null != isolatedScan) { @@ -118,34 +240,166 @@ public class RangeInputSplit extends AccumuloInputSplit { if (null != localIterators) { out.writeBoolean(localIterators); } + + out.writeBoolean(null != mockInstance); + if (null != mockInstance) { + out.writeBoolean(mockInstance); + } + + out.writeBoolean(null != fetchedColumns); + if (null != fetchedColumns) { + String[] cols = InputConfigurator.serializeColumns(fetchedColumns); + out.writeInt(cols.length); + for (String col : cols) { + out.writeUTF(col); + } + } + + out.writeBoolean(null != auths); + if (null != auths) { + out.writeUTF(auths.serialize()); + } + + out.writeBoolean(null != principal); + if (null != principal) { + out.writeUTF(principal); + } + + out.writeBoolean(null != tokenSource); + if (null != tokenSource) { + out.writeInt(tokenSource.ordinal()); + + if (null != token && null != tokenFile) { + throw new IOException("Cannot use both inline AuthenticationToken and file-based AuthenticationToken"); + } else if (null != token) { + out.writeUTF(token.getClass().getCanonicalName()); + out.writeUTF(Base64.encodeBase64String(AuthenticationTokenSerializer.serialize(token))); + } else { + out.writeUTF(tokenFile); + } + } + + out.writeBoolean(null != instanceName); + if (null != instanceName) { + out.writeUTF(instanceName); + } + + out.writeBoolean(null != zooKeepers); + if (null != zooKeepers) { + out.writeUTF(zooKeepers); + } + + out.writeBoolean(null != iterators); + if (null != iterators) { + out.writeInt(iterators.size()); + for (IteratorSetting iterator : iterators) { + iterator.write(out); + } + } + + out.writeBoolean(null != level); + if (null != level) { + out.writeInt(level.toInt()); + } } - @Override - public String toString() { - StringBuilder sb = new StringBuilder(256); - sb.append("RangeInputSplit:"); - sb.append(" Range: ").append(range); - sb.append(super.toString()); - sb.append(" offlineScan: ").append(offline); - sb.append(" isolatedScan: ").append(isolatedScan); - sb.append(" localIterators: ").append(localIterators); - return sb.toString(); + /** + * Use {@link #getTableName} + * + * @deprecated since 1.6.1, use getTableName() instead. + */ + @Deprecated + public String getTable() { + return getTableName(); } - public Range getRange() { - return range; + public String getTableName() { + return tableName; } - public void setRange(Range range) { - this.range = range; + /** + * Use {@link #setTableName} + * + * @deprecated since 1.6.1, use setTableName() instead. + */ + @Deprecated + public void setTable(String table) { + setTableName(table); } - public Boolean isIsolatedScan() { - return isolatedScan; + public void setTableName(String table) { + this.tableName = table; } - public void setIsolatedScan(Boolean isolatedScan) { - this.isolatedScan = isolatedScan; + public void setTableId(String tableId) { + this.tableId = tableId; + } + + public String getTableId() { + return tableId; + } + + /** + * @see #getInstance(ClientConfiguration) + * @deprecated since 1.7.0, use getInstance(ClientConfiguration) instead. + */ + @Deprecated + public Instance getInstance() { + return getInstance(ClientConfiguration.loadDefault()); + } + + public Instance getInstance(ClientConfiguration base) { + if (null == instanceName) { + return null; + } + + if (isMockInstance()) { + return new MockInstance(getInstanceName()); + } + + if (null == zooKeepers) { + return null; + } + + return new ZooKeeperInstance(base.withInstance(getInstanceName()).withZkHosts(getZooKeepers())); + } + + public String getInstanceName() { + return instanceName; + } + + public void setInstanceName(String instanceName) { + this.instanceName = instanceName; + } + + public String getZooKeepers() { + return zooKeepers; + } + + public void setZooKeepers(String zooKeepers) { + this.zooKeepers = zooKeepers; + } + + public String getPrincipal() { + return principal; + } + + public void setPrincipal(String principal) { + this.principal = principal; + } + + public AuthenticationToken getToken() { + return token; + } + + public void setToken(AuthenticationToken token) { + this.tokenSource = TokenSource.INLINE; + this.token = token; + } + + public void setToken(String tokenFile) { + this.tokenSource = TokenSource.FILE; + this.tokenFile = tokenFile; } public Boolean isOffline() { @@ -156,6 +410,38 @@ public class RangeInputSplit extends AccumuloInputSplit { this.offline = offline; } + public void setLocations(String[] locations) { + this.locations = Arrays.copyOf(locations, locations.length); + } + + public Boolean isMockInstance() { + return mockInstance; + } + + public void setMockInstance(Boolean mockInstance) { + this.mockInstance = mockInstance; + } + + public Boolean isIsolatedScan() { + return isolatedScan; + } + + public void setIsolatedScan(Boolean isolatedScan) { + this.isolatedScan = isolatedScan; + } + + public Authorizations getAuths() { + return auths; + } + + public void setAuths(Authorizations auths) { + this.auths = auths; + } + + public void setRange(Range range) { + this.range = range; + } + public Boolean usesLocalIterators() { return localIterators; } @@ -163,4 +449,59 @@ public class RangeInputSplit extends AccumuloInputSplit { public void setUsesLocalIterators(Boolean localIterators) { this.localIterators = localIterators; } + + public Set<Pair<Text,Text>> getFetchedColumns() { + return fetchedColumns; + } + + public void setFetchedColumns(Collection<Pair<Text,Text>> fetchedColumns) { + this.fetchedColumns = new HashSet<Pair<Text,Text>>(); + for (Pair<Text,Text> columns : fetchedColumns) { + this.fetchedColumns.add(columns); + } + } + + public void setFetchedColumns(Set<Pair<Text,Text>> fetchedColumns) { + this.fetchedColumns = fetchedColumns; + } + + public List<IteratorSetting> getIterators() { + return iterators; + } + + public void setIterators(List<IteratorSetting> iterators) { + this.iterators = iterators; + } + + public Level getLogLevel() { + return level; + } + + public void setLogLevel(Level level) { + this.level = level; + } + + @Override + public String toString() { + StringBuilder sb = new StringBuilder(256); + sb.append("Range: ").append(range); + sb.append(" Locations: ").append(Arrays.asList(locations)); + sb.append(" Table: ").append(tableName); + sb.append(" TableID: ").append(tableId); + sb.append(" InstanceName: ").append(instanceName); + sb.append(" zooKeepers: ").append(zooKeepers); + sb.append(" principal: ").append(principal); + sb.append(" tokenSource: ").append(tokenSource); + sb.append(" authenticationToken: ").append(token); + sb.append(" authenticationTokenFile: ").append(tokenFile); + sb.append(" Authorizations: ").append(auths); + sb.append(" offlineScan: ").append(offline); + sb.append(" mockInstance: ").append(mockInstance); + sb.append(" isolatedScan: ").append(isolatedScan); + sb.append(" localIterators: ").append(localIterators); + sb.append(" fetchColumns: ").append(fetchedColumns); + sb.append(" iterators: ").append(iterators); + sb.append(" logLevel: ").append(level); + return sb.toString(); + } } http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/AccumuloInputSplit.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/AccumuloInputSplit.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/AccumuloInputSplit.java deleted file mode 100644 index 7f83936..0000000 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/AccumuloInputSplit.java +++ /dev/null @@ -1,436 +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.accumulo.core.client.mapreduce.impl; - -import static java.nio.charset.StandardCharsets.UTF_8; - -import java.io.DataInput; -import java.io.DataOutput; -import java.io.IOException; -import java.util.ArrayList; -import java.util.Arrays; -import java.util.Collection; -import java.util.HashSet; -import java.util.List; -import java.util.Set; - -import org.apache.accumulo.core.client.ClientConfiguration; -import org.apache.accumulo.core.client.Instance; -import org.apache.accumulo.core.client.IteratorSetting; -import org.apache.accumulo.core.client.ZooKeeperInstance; -import org.apache.accumulo.core.client.mapreduce.lib.impl.ConfiguratorBase.TokenSource; -import org.apache.accumulo.core.client.mapreduce.lib.impl.InputConfigurator; -import org.apache.accumulo.core.client.mock.MockInstance; -import org.apache.accumulo.core.client.security.tokens.AuthenticationToken; -import org.apache.accumulo.core.client.security.tokens.AuthenticationToken.AuthenticationTokenSerializer; -import org.apache.accumulo.core.data.ByteSequence; -import org.apache.accumulo.core.data.Key; -import org.apache.accumulo.core.data.Range; -import org.apache.accumulo.core.security.Authorizations; -import org.apache.accumulo.core.util.Base64; -import org.apache.accumulo.core.util.Pair; -import org.apache.hadoop.io.Text; -import org.apache.hadoop.io.Writable; -import org.apache.hadoop.mapreduce.InputSplit; -import org.apache.log4j.Level; - -/** - * Abstracts over configurations common to all InputSplits. Specifically it leaves out methods - * related to number of ranges and locations per InputSplit as those vary by implementation. - * - * @see org.apache.accumulo.core.client.mapreduce.RangeInputSplit - * @see org.apache.accumulo.core.client.mapreduce.impl.BatchInputSplit - */ -public abstract class AccumuloInputSplit extends InputSplit implements Writable { - private String[] locations; - private String tableId, tableName, instanceName, zooKeepers, principal; - private TokenSource tokenSource; - private String tokenFile; - private AuthenticationToken token; - private Boolean mockInstance; - private Authorizations auths; - private Set<Pair<Text,Text>> fetchedColumns; - private List<IteratorSetting> iterators; - private Level level; - - public abstract float getProgress(Key currentKey); - - public AccumuloInputSplit() { - locations = new String[0]; - tableName = ""; - tableId = ""; - } - - public AccumuloInputSplit(AccumuloInputSplit split) throws IOException { - this.setLocations(split.getLocations()); - this.setTableName(split.getTableName()); - this.setTableId(split.getTableId()); - } - - protected AccumuloInputSplit(String table, String tableId, String[] locations) { - setLocations(locations); - this.tableName = table; - this.tableId = tableId; - } - - static byte[] extractBytes(ByteSequence seq, int numBytes) { - byte[] bytes = new byte[numBytes + 1]; - bytes[0] = 0; - for (int i = 0; i < numBytes; i++) { - if (i >= seq.length()) - bytes[i + 1] = 0; - else - bytes[i + 1] = seq.byteAt(i); - } - return bytes; - } - - public long getRangeLength(Range range) throws IOException { - Text startRow = range.isInfiniteStartKey() ? new Text(new byte[] {Byte.MIN_VALUE}) : range.getStartKey().getRow(); - Text stopRow = range.isInfiniteStopKey() ? new Text(new byte[] {Byte.MAX_VALUE}) : range.getEndKey().getRow(); - int maxCommon = Math.min(7, Math.min(startRow.getLength(), stopRow.getLength())); - long diff = 0; - - byte[] start = startRow.getBytes(); - byte[] stop = stopRow.getBytes(); - for (int i = 0; i < maxCommon; ++i) { - diff |= 0xff & (start[i] ^ stop[i]); - diff <<= Byte.SIZE; - } - - if (startRow.getLength() != stopRow.getLength()) - diff |= 0xff; - - return diff + 1; - } - - @Override - public String[] getLocations() throws IOException { - return Arrays.copyOf(locations, locations.length); - } - - @Override - public void readFields(DataInput in) throws IOException { - tableName = in.readUTF(); - tableId = in.readUTF(); - int numLocs = in.readInt(); - locations = new String[numLocs]; - for (int i = 0; i < numLocs; ++i) - locations[i] = in.readUTF(); - - if (in.readBoolean()) { - mockInstance = in.readBoolean(); - } - - if (in.readBoolean()) { - int numColumns = in.readInt(); - List<String> columns = new ArrayList<String>(numColumns); - for (int i = 0; i < numColumns; i++) { - columns.add(in.readUTF()); - } - - fetchedColumns = InputConfigurator.deserializeFetchedColumns(columns); - } - - if (in.readBoolean()) { - String strAuths = in.readUTF(); - auths = new Authorizations(strAuths.getBytes(UTF_8)); - } - - if (in.readBoolean()) { - principal = in.readUTF(); - } - - if (in.readBoolean()) { - int ordinal = in.readInt(); - this.tokenSource = TokenSource.values()[ordinal]; - - switch (this.tokenSource) { - case INLINE: - String tokenClass = in.readUTF(); - byte[] base64TokenBytes = in.readUTF().getBytes(UTF_8); - byte[] tokenBytes = Base64.decodeBase64(base64TokenBytes); - - this.token = AuthenticationTokenSerializer.deserialize(tokenClass, tokenBytes); - break; - - case FILE: - this.tokenFile = in.readUTF(); - - break; - default: - throw new IOException("Cannot parse unknown TokenSource ordinal"); - } - } - - if (in.readBoolean()) { - instanceName = in.readUTF(); - } - - if (in.readBoolean()) { - zooKeepers = in.readUTF(); - } - - if (in.readBoolean()) { - int numIterators = in.readInt(); - iterators = new ArrayList<IteratorSetting>(numIterators); - for (int i = 0; i < numIterators; i++) { - iterators.add(new IteratorSetting(in)); - } - } - - if (in.readBoolean()) { - level = Level.toLevel(in.readInt()); - } - } - - @Override - public void write(DataOutput out) throws IOException { - out.writeUTF(tableName); - out.writeUTF(tableId); - out.writeInt(locations.length); - for (int i = 0; i < locations.length; ++i) - out.writeUTF(locations[i]); - - out.writeBoolean(null != mockInstance); - if (null != mockInstance) { - out.writeBoolean(mockInstance); - } - - out.writeBoolean(null != fetchedColumns); - if (null != fetchedColumns) { - String[] cols = InputConfigurator.serializeColumns(fetchedColumns); - out.writeInt(cols.length); - for (String col : cols) { - out.writeUTF(col); - } - } - - out.writeBoolean(null != auths); - if (null != auths) { - out.writeUTF(auths.serialize()); - } - - out.writeBoolean(null != principal); - if (null != principal) { - out.writeUTF(principal); - } - - out.writeBoolean(null != tokenSource); - if (null != tokenSource) { - out.writeInt(tokenSource.ordinal()); - - if (null != token && null != tokenFile) { - throw new IOException("Cannot use both inline AuthenticationToken and file-based AuthenticationToken"); - } else if (null != token) { - out.writeUTF(token.getClass().getCanonicalName()); - out.writeUTF(Base64.encodeBase64String(AuthenticationTokenSerializer.serialize(token))); - } else { - out.writeUTF(tokenFile); - } - } - - out.writeBoolean(null != instanceName); - if (null != instanceName) { - out.writeUTF(instanceName); - } - - out.writeBoolean(null != zooKeepers); - if (null != zooKeepers) { - out.writeUTF(zooKeepers); - } - - out.writeBoolean(null != iterators); - if (null != iterators) { - out.writeInt(iterators.size()); - for (IteratorSetting iterator : iterators) { - iterator.write(out); - } - } - - out.writeBoolean(null != level); - if (null != level) { - out.writeInt(level.toInt()); - } - } - - /** - * Use {@link #getTableName} - */ - @Deprecated - public String getTable() { - return getTableName(); - } - - public String getTableName() { - return tableName; - } - - /** - * Use {@link #setTableName} - */ - @Deprecated - public void setTable(String table) { - setTableName(table); - } - - public void setTableName(String table) { - this.tableName = table; - } - - public void setTableId(String tableId) { - this.tableId = tableId; - } - - public String getTableId() { - return tableId; - } - - /** - * @see #getInstance(ClientConfiguration) - */ - @Deprecated - public Instance getInstance() { - return getInstance(ClientConfiguration.loadDefault()); - } - - public Instance getInstance(ClientConfiguration base) { - if (null == instanceName) { - return null; - } - - if (isMockInstance()) { - return new MockInstance(getInstanceName()); - } - - if (null == zooKeepers) { - return null; - } - - return new ZooKeeperInstance(base.withInstance(getInstanceName()).withZkHosts(getZooKeepers())); - } - - public String getInstanceName() { - return instanceName; - } - - public void setInstanceName(String instanceName) { - this.instanceName = instanceName; - } - - public String getZooKeepers() { - return zooKeepers; - } - - public void setZooKeepers(String zooKeepers) { - this.zooKeepers = zooKeepers; - } - - public String getPrincipal() { - return principal; - } - - public void setPrincipal(String principal) { - this.principal = principal; - } - - public AuthenticationToken getToken() { - return token; - } - - public void setToken(AuthenticationToken token) { - this.tokenSource = TokenSource.INLINE; - this.token = token; - } - - public void setToken(String tokenFile) { - this.tokenSource = TokenSource.FILE; - this.tokenFile = tokenFile; - } - - public void setLocations(String[] locations) { - this.locations = Arrays.copyOf(locations, locations.length); - } - - public Boolean isMockInstance() { - return mockInstance; - } - - public void setMockInstance(Boolean mockInstance) { - this.mockInstance = mockInstance; - } - - public Authorizations getAuths() { - return auths; - } - - public void setAuths(Authorizations auths) { - this.auths = auths; - } - - - public Set<Pair<Text,Text>> getFetchedColumns() { - return fetchedColumns; - } - - public void setFetchedColumns(Collection<Pair<Text,Text>> fetchedColumns) { - this.fetchedColumns = new HashSet<Pair<Text,Text>>(); - for (Pair<Text,Text> columns : fetchedColumns) { - this.fetchedColumns.add(columns); - } - } - - public void setFetchedColumns(Set<Pair<Text,Text>> fetchedColumns) { - this.fetchedColumns = fetchedColumns; - } - - public List<IteratorSetting> getIterators() { - return iterators; - } - - public void setIterators(List<IteratorSetting> iterators) { - this.iterators = iterators; - } - - public Level getLogLevel() { - return level; - } - - public void setLogLevel(Level level) { - this.level = level; - } - - @Override - public String toString(){ - StringBuilder sb = new StringBuilder(256); - sb.append(" Locations: ").append(Arrays.asList(locations)); - sb.append(" Table: ").append(tableName); - sb.append(" TableID: ").append(tableId); - sb.append(" InstanceName: ").append(instanceName); - sb.append(" zooKeepers: ").append(zooKeepers); - sb.append(" principal: ").append(principal); - sb.append(" tokenSource: ").append(tokenSource); - sb.append(" authenticationToken: ").append(token); - sb.append(" authenticationTokenFile: ").append(tokenFile); - sb.append(" Authorizations: ").append(auths); - sb.append(" mockInstance: ").append(mockInstance); - sb.append(" fetchColumns: ").append(fetchedColumns); - sb.append(" iterators: ").append(iterators); - sb.append(" logLevel: ").append(level); - return sb.toString(); - } -} http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/BatchInputSplit.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/BatchInputSplit.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/BatchInputSplit.java index 24b9ef3..f674485 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/BatchInputSplit.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/BatchInputSplit.java @@ -24,6 +24,7 @@ import java.util.Arrays; import java.util.Collection; import java.util.Collections; +import org.apache.accumulo.core.client.mapreduce.RangeInputSplit; import org.apache.accumulo.core.data.Key; import org.apache.accumulo.core.data.PartialKey; import org.apache.accumulo.core.data.Range; @@ -32,7 +33,7 @@ import org.apache.accumulo.core.data.Range; * The Class BatchInputSplit. Encapsulates a set of Accumulo ranges on a single tablet for use in Map Reduce jobs. * Can contain several Ranges per split. */ -public class BatchInputSplit extends AccumuloInputSplit { +public class BatchInputSplit extends RangeInputSplit { private Collection<Range> ranges; private float[] rangeProgress = null; @@ -46,7 +47,7 @@ public class BatchInputSplit extends AccumuloInputSplit { } public BatchInputSplit(String table, String tableId, Collection<Range> ranges, String[] locations) { - super(table, tableId, locations); + super(table, tableId, new Range(), locations); this.ranges = ranges; } @@ -96,7 +97,7 @@ public class BatchInputSplit extends AccumuloInputSplit { public long getLength() throws IOException { long sum = 0; for (Range range : ranges) - sum += getRangeLength(range); + sum += SplitUtils.getRangeLength(range); return sum; } @@ -138,4 +139,44 @@ public class BatchInputSplit extends AccumuloInputSplit { public Collection<Range> getRanges() { return ranges; } + + @Override + public Range getRange(){ + throw new UnsupportedOperationException(); + } + + @Override + public void setRange(Range range){ + throw new UnsupportedOperationException(); + } + + @Override + public Boolean isIsolatedScan() { + throw new UnsupportedOperationException(); + } + + @Override + public void setIsolatedScan(Boolean isolatedScan) { + throw new UnsupportedOperationException(); + } + + @Override + public Boolean isOffline() { + throw new UnsupportedOperationException(); + } + + @Override + public void setOffline(Boolean offline) { + throw new UnsupportedOperationException(); + } + + @Override + public Boolean usesLocalIterators() { + throw new UnsupportedOperationException(); + } + + @Override + public void setUsesLocalIterators(Boolean localIterators) { + throw new UnsupportedOperationException(); + } } \ No newline at end of file http://git-wip-us.apache.org/repos/asf/accumulo/blob/83d1b838/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/SplitUtils.java ---------------------------------------------------------------------- diff --git a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/SplitUtils.java b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/SplitUtils.java index 0aee665..742660d 100644 --- a/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/SplitUtils.java +++ b/core/src/main/java/org/apache/accumulo/core/client/mapreduce/impl/SplitUtils.java @@ -17,14 +17,18 @@ package org.apache.accumulo.core.client.mapreduce.impl; +import java.io.IOException; import java.math.BigInteger; import org.apache.accumulo.core.client.Instance; import org.apache.accumulo.core.client.mapreduce.InputTableConfig; +import org.apache.accumulo.core.client.mapreduce.RangeInputSplit; import org.apache.accumulo.core.client.mock.MockInstance; import org.apache.accumulo.core.client.security.tokens.AuthenticationToken; import org.apache.accumulo.core.data.ByteSequence; +import org.apache.accumulo.core.data.Range; import org.apache.accumulo.core.security.Authorizations; +import org.apache.hadoop.io.Text; import org.apache.log4j.Level; public class SplitUtils { @@ -33,7 +37,7 @@ public class SplitUtils { * Central place to set common split configuration not handled by split constructors. * The intention is to make it harder to miss optional setters in future refactor. */ - public static void updateSplit(AccumuloInputSplit split, Instance instance, InputTableConfig tableConfig, + public static void updateSplit(RangeInputSplit split, Instance instance, InputTableConfig tableConfig, String principal, AuthenticationToken token, Authorizations auths, Level logLevel) { split.setInstanceName(instance.getInstanceName()); split.setZooKeepers(instance.getZooKeepers()); @@ -50,10 +54,41 @@ public class SplitUtils { public static float getProgress(ByteSequence start, ByteSequence end, ByteSequence position) { int maxDepth = Math.min(Math.max(end.length(), start.length()), position.length()); - BigInteger startBI = new BigInteger(AccumuloInputSplit.extractBytes(start, maxDepth)); - BigInteger endBI = new BigInteger(AccumuloInputSplit.extractBytes(end, maxDepth)); - BigInteger positionBI = new BigInteger(AccumuloInputSplit.extractBytes(position, maxDepth)); + BigInteger startBI = new BigInteger(SplitUtils.extractBytes(start, maxDepth)); + BigInteger endBI = new BigInteger(SplitUtils.extractBytes(end, maxDepth)); + BigInteger positionBI = new BigInteger(SplitUtils.extractBytes(position, maxDepth)); return (float) (positionBI.subtract(startBI).doubleValue() / endBI.subtract(startBI).doubleValue()); } + public static long getRangeLength(Range range) throws IOException { + Text startRow = range.isInfiniteStartKey() ? new Text(new byte[] {Byte.MIN_VALUE}) : range.getStartKey().getRow(); + Text stopRow = range.isInfiniteStopKey() ? new Text(new byte[] {Byte.MAX_VALUE}) : range.getEndKey().getRow(); + int maxCommon = Math.min(7, Math.min(startRow.getLength(), stopRow.getLength())); + long diff = 0; + + byte[] start = startRow.getBytes(); + byte[] stop = stopRow.getBytes(); + for (int i = 0; i < maxCommon; ++i) { + diff |= 0xff & (start[i] ^ stop[i]); + diff <<= Byte.SIZE; + } + + if (startRow.getLength() != stopRow.getLength()) + diff |= 0xff; + + return diff + 1; + } + + static byte[] extractBytes(ByteSequence seq, int numBytes) { + byte[] bytes = new byte[numBytes + 1]; + bytes[0] = 0; + for (int i = 0; i < numBytes; i++) { + if (i >= seq.length()) + bytes[i + 1] = 0; + else + bytes[i + 1] = seq.byteAt(i); + } + return bytes; + } + }