Added Binary Preclustered Group Operator. Preclustered and Hash based group operators now use the same aggregator interface. Added a Frame Writing Operator.
git-svn-id: https://hyracks.googlecode.com/svn/trunk/hyracks@132 123451ca-8445-de46-9d55-352943316053
diff --git a/hyracks-examples/text-example/.project b/hyracks-examples/text-example/.project
new file mode 100644
index 0000000..4e057cd
--- /dev/null
+++ b/hyracks-examples/text-example/.project
@@ -0,0 +1,17 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+ <name>text-example</name>
+ <comment></comment>
+ <projects>
+ </projects>
+ <buildSpec>
+ <buildCommand>
+ <name>org.maven.ide.eclipse.maven2Builder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ </buildSpec>
+ <natures>
+ <nature>org.maven.ide.eclipse.maven2Nature</nature>
+ </natures>
+</projectDescription>
diff --git a/hyracks-examples/text-example/.settings/org.maven.ide.eclipse.prefs b/hyracks-examples/text-example/.settings/org.maven.ide.eclipse.prefs
new file mode 100644
index 0000000..4562b1a
--- /dev/null
+++ b/hyracks-examples/text-example/.settings/org.maven.ide.eclipse.prefs
@@ -0,0 +1,9 @@
+#Tue Sep 28 14:37:42 PDT 2010
+activeProfiles=
+eclipse.preferences.version=1
+fullBuildGoals=process-test-resources
+includeModules=false
+resolveWorkspaceProjects=true
+resourceFilterGoals=process-resources resources\:testResources
+skipCompilerPlugin=true
+version=1
diff --git a/hyracks-examples/text-example/pom.xml b/hyracks-examples/text-example/pom.xml
new file mode 100644
index 0000000..f6b8d8f
--- /dev/null
+++ b/hyracks-examples/text-example/pom.xml
@@ -0,0 +1,19 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+ <modelVersion>4.0.0</modelVersion>
+ <groupId>edu.uci.ics.hyracks.examples</groupId>
+ <artifactId>text-example</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <packaging>pom</packaging>
+
+ <parent>
+ <groupId>edu.uci.ics.hyracks</groupId>
+ <artifactId>hyracks-examples</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ </parent>
+
+ <modules>
+ <module>texthelper</module>
+ <module>textclient</module>
+ <module>textapp</module>
+ </modules>
+</project>
diff --git a/hyracks-examples/text-example/textapp/.classpath b/hyracks-examples/text-example/textapp/.classpath
new file mode 100644
index 0000000..3f62785
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/.classpath
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+ <classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/J2SE-1.4"/>
+ <classpathentry kind="con" path="org.maven.ide.eclipse.MAVEN2_CLASSPATH_CONTAINER"/>
+ <classpathentry kind="output" path="target/classes"/>
+</classpath>
diff --git a/hyracks-examples/text-example/textapp/.project b/hyracks-examples/text-example/textapp/.project
new file mode 100644
index 0000000..4f3af14
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/.project
@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+ <name>textapp</name>
+ <comment></comment>
+ <projects>
+ </projects>
+ <buildSpec>
+ <buildCommand>
+ <name>org.eclipse.jdt.core.javabuilder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ <buildCommand>
+ <name>org.maven.ide.eclipse.maven2Builder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ </buildSpec>
+ <natures>
+ <nature>org.eclipse.jdt.core.javanature</nature>
+ <nature>org.maven.ide.eclipse.maven2Nature</nature>
+ </natures>
+</projectDescription>
diff --git a/hyracks-examples/text-example/textapp/.settings/org.eclipse.jdt.core.prefs b/hyracks-examples/text-example/textapp/.settings/org.eclipse.jdt.core.prefs
new file mode 100644
index 0000000..37272d9
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/.settings/org.eclipse.jdt.core.prefs
@@ -0,0 +1,6 @@
+#Tue Sep 28 14:37:42 PDT 2010
+eclipse.preferences.version=1
+org.eclipse.jdt.core.compiler.codegen.targetPlatform=1.4
+org.eclipse.jdt.core.compiler.compliance=1.4
+org.eclipse.jdt.core.compiler.problem.forbiddenReference=warning
+org.eclipse.jdt.core.compiler.source=1.4
diff --git a/hyracks-examples/text-example/textapp/.settings/org.maven.ide.eclipse.prefs b/hyracks-examples/text-example/textapp/.settings/org.maven.ide.eclipse.prefs
new file mode 100644
index 0000000..4562b1a
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/.settings/org.maven.ide.eclipse.prefs
@@ -0,0 +1,9 @@
+#Tue Sep 28 14:37:42 PDT 2010
+activeProfiles=
+eclipse.preferences.version=1
+fullBuildGoals=process-test-resources
+includeModules=false
+resolveWorkspaceProjects=true
+resourceFilterGoals=process-resources resources\:testResources
+skipCompilerPlugin=true
+version=1
diff --git a/hyracks-examples/text-example/textapp/pom.xml b/hyracks-examples/text-example/textapp/pom.xml
new file mode 100644
index 0000000..706ae72
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/pom.xml
@@ -0,0 +1,58 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+ <modelVersion>4.0.0</modelVersion>
+ <groupId>edu.uci.ics.hyracks.examples.text</groupId>
+ <artifactId>textapp</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+
+ <parent>
+ <groupId>edu.uci.ics.hyracks.examples</groupId>
+ <artifactId>text-example</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ </parent>
+
+ <build>
+ <plugins>
+ <plugin>
+ <groupId>org.apache.maven.plugins</groupId>
+ <artifactId>maven-dependency-plugin</artifactId>
+ <executions>
+ <execution>
+ <id>copy-dependencies</id>
+ <phase>package</phase>
+ <goals>
+ <goal>copy-dependencies</goal>
+ </goals>
+ <configuration>
+ <outputDirectory>target/application/lib</outputDirectory>
+ </configuration>
+ </execution>
+ </executions>
+ </plugin>
+ <plugin>
+ <artifactId>maven-assembly-plugin</artifactId>
+ <version>2.2-beta-5</version>
+ <executions>
+ <execution>
+ <configuration>
+ <descriptors>
+ <descriptor>src/main/assembly/app-assembly.xml</descriptor>
+ </descriptors>
+ </configuration>
+ <phase>package</phase>
+ <goals>
+ <goal>attached</goal>
+ </goals>
+ </execution>
+ </executions>
+ </plugin>
+ </plugins>
+ </build>
+ <dependencies>
+ <dependency>
+ <groupId>edu.uci.ics.hyracks.examples.text</groupId>
+ <artifactId>texthelper</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <scope>compile</scope>
+ </dependency>
+ </dependencies>
+</project>
diff --git a/hyracks-examples/text-example/textapp/src/main/assembly/app-assembly.xml b/hyracks-examples/text-example/textapp/src/main/assembly/app-assembly.xml
new file mode 100644
index 0000000..43ace6c
--- /dev/null
+++ b/hyracks-examples/text-example/textapp/src/main/assembly/app-assembly.xml
@@ -0,0 +1,13 @@
+<assembly>
+ <id>app-assembly</id>
+ <formats>
+ <format>zip</format>
+ </formats>
+ <includeBaseDirectory>false</includeBaseDirectory>
+ <fileSets>
+ <fileSet>
+ <directory>target/application/lib</directory>
+ <outputDirectory>lib</outputDirectory>
+ </fileSet>
+ </fileSets>
+</assembly>
diff --git a/hyracks-examples/text-example/textclient/.classpath b/hyracks-examples/text-example/textclient/.classpath
new file mode 100644
index 0000000..1f3c1ff
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/.classpath
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+ <classpathentry kind="src" output="target/classes" path="src/main/java"/>
+ <classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.6"/>
+ <classpathentry kind="con" path="org.maven.ide.eclipse.MAVEN2_CLASSPATH_CONTAINER"/>
+ <classpathentry kind="output" path="target/classes"/>
+</classpath>
diff --git a/hyracks-examples/text-example/textclient/.project b/hyracks-examples/text-example/textclient/.project
new file mode 100644
index 0000000..04307d3
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/.project
@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+ <name>textclient</name>
+ <comment></comment>
+ <projects>
+ </projects>
+ <buildSpec>
+ <buildCommand>
+ <name>org.eclipse.jdt.core.javabuilder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ <buildCommand>
+ <name>org.maven.ide.eclipse.maven2Builder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ </buildSpec>
+ <natures>
+ <nature>org.eclipse.jdt.core.javanature</nature>
+ <nature>org.maven.ide.eclipse.maven2Nature</nature>
+ </natures>
+</projectDescription>
diff --git a/hyracks-examples/text-example/textclient/.settings/org.eclipse.jdt.core.prefs b/hyracks-examples/text-example/textclient/.settings/org.eclipse.jdt.core.prefs
new file mode 100644
index 0000000..8599738
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/.settings/org.eclipse.jdt.core.prefs
@@ -0,0 +1,6 @@
+#Tue Sep 28 14:37:42 PDT 2010
+eclipse.preferences.version=1
+org.eclipse.jdt.core.compiler.codegen.targetPlatform=1.6
+org.eclipse.jdt.core.compiler.compliance=1.6
+org.eclipse.jdt.core.compiler.problem.forbiddenReference=warning
+org.eclipse.jdt.core.compiler.source=1.6
diff --git a/hyracks-examples/text-example/textclient/.settings/org.maven.ide.eclipse.prefs b/hyracks-examples/text-example/textclient/.settings/org.maven.ide.eclipse.prefs
new file mode 100644
index 0000000..4562b1a
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/.settings/org.maven.ide.eclipse.prefs
@@ -0,0 +1,9 @@
+#Tue Sep 28 14:37:42 PDT 2010
+activeProfiles=
+eclipse.preferences.version=1
+fullBuildGoals=process-test-resources
+includeModules=false
+resolveWorkspaceProjects=true
+resourceFilterGoals=process-resources resources\:testResources
+skipCompilerPlugin=true
+version=1
diff --git a/hyracks-examples/text-example/textclient/pom.xml b/hyracks-examples/text-example/textclient/pom.xml
new file mode 100644
index 0000000..c56a6b8
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/pom.xml
@@ -0,0 +1,80 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+ <modelVersion>4.0.0</modelVersion>
+ <groupId>edu.uci.ics.hyracks.examples.text</groupId>
+ <artifactId>textclient</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+
+ <parent>
+ <groupId>edu.uci.ics.hyracks.examples</groupId>
+ <artifactId>text-example</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ </parent>
+
+ <dependencies>
+ <dependency>
+ <groupId>edu.uci.ics.hyracks</groupId>
+ <artifactId>hyracks-dataflow-std</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <scope>compile</scope>
+ </dependency>
+ <dependency>
+ <groupId>edu.uci.ics.hyracks.examples.text</groupId>
+ <artifactId>texthelper</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <type>jar</type>
+ <scope>compile</scope>
+ </dependency>
+ </dependencies>
+ <build>
+ <plugins>
+ <plugin>
+ <groupId>org.apache.maven.plugins</groupId>
+ <artifactId>maven-compiler-plugin</artifactId>
+ <version>2.0.2</version>
+ <configuration>
+ <source>1.6</source>
+ <target>1.6</target>
+ </configuration>
+ </plugin>
+ <plugin>
+ <groupId>org.codehaus.mojo</groupId>
+ <artifactId>appassembler-maven-plugin</artifactId>
+ <executions>
+ <execution>
+ <configuration>
+ <programs>
+ <program>
+ <mainClass>edu.uci.ics.hyracks.examples.text.client.WordCountMain</mainClass>
+ <name>textclient</name>
+ </program>
+ </programs>
+ <repositoryLayout>flat</repositoryLayout>
+ <repositoryName>lib</repositoryName>
+ </configuration>
+ <phase>package</phase>
+ <goals>
+ <goal>assemble</goal>
+ </goals>
+ </execution>
+ </executions>
+ </plugin>
+ <plugin>
+ <artifactId>maven-assembly-plugin</artifactId>
+ <version>2.2-beta-5</version>
+ <executions>
+ <execution>
+ <configuration>
+ <descriptors>
+ <descriptor>src/main/assembly/binary-assembly.xml</descriptor>
+ </descriptors>
+ </configuration>
+ <phase>package</phase>
+ <goals>
+ <goal>attached</goal>
+ </goals>
+ </execution>
+ </executions>
+ </plugin>
+ </plugins>
+ </build>
+</project>
diff --git a/hyracks-examples/text-example/textclient/src/main/assembly/binary-assembly.xml b/hyracks-examples/text-example/textclient/src/main/assembly/binary-assembly.xml
new file mode 100644
index 0000000..0500499
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/src/main/assembly/binary-assembly.xml
@@ -0,0 +1,19 @@
+<assembly>
+ <id>binary-assembly</id>
+ <formats>
+ <format>zip</format>
+ <format>dir</format>
+ </formats>
+ <includeBaseDirectory>false</includeBaseDirectory>
+ <fileSets>
+ <fileSet>
+ <directory>target/appassembler/bin</directory>
+ <outputDirectory>bin</outputDirectory>
+ <fileMode>0755</fileMode>
+ </fileSet>
+ <fileSet>
+ <directory>target/appassembler/lib</directory>
+ <outputDirectory>lib</outputDirectory>
+ </fileSet>
+ </fileSets>
+</assembly>
diff --git a/hyracks-examples/text-example/textclient/src/main/java/edu/uci/ics/hyracks/examples/text/client/WordCountMain.java b/hyracks-examples/text-example/textclient/src/main/java/edu/uci/ics/hyracks/examples/text/client/WordCountMain.java
new file mode 100644
index 0000000..96045d2
--- /dev/null
+++ b/hyracks-examples/text-example/textclient/src/main/java/edu/uci/ics/hyracks/examples/text/client/WordCountMain.java
@@ -0,0 +1,180 @@
+/*
+ * Copyright 2009-2010 by The Regents of the University of California
+ * Licensed 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 from
+ *
+ * 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 edu.uci.ics.hyracks.examples.text.client;
+
+import java.io.File;
+import java.util.UUID;
+
+import org.kohsuke.args4j.CmdLineParser;
+import org.kohsuke.args4j.Option;
+
+import edu.uci.ics.hyracks.api.client.HyracksRMIConnection;
+import edu.uci.ics.hyracks.api.client.IHyracksClientConnection;
+import edu.uci.ics.hyracks.api.constraints.AbsoluteLocationConstraint;
+import edu.uci.ics.hyracks.api.constraints.ExplicitPartitionConstraint;
+import edu.uci.ics.hyracks.api.constraints.LocationConstraint;
+import edu.uci.ics.hyracks.api.constraints.PartitionConstraint;
+import edu.uci.ics.hyracks.api.dataflow.IConnectorDescriptor;
+import edu.uci.ics.hyracks.api.dataflow.IOperatorDescriptor;
+import edu.uci.ics.hyracks.api.dataflow.value.IBinaryComparatorFactory;
+import edu.uci.ics.hyracks.api.dataflow.value.IBinaryHashFunctionFactory;
+import edu.uci.ics.hyracks.api.dataflow.value.ISerializerDeserializer;
+import edu.uci.ics.hyracks.api.dataflow.value.RecordDescriptor;
+import edu.uci.ics.hyracks.api.job.JobSpecification;
+import edu.uci.ics.hyracks.dataflow.common.data.comparators.UTF8StringBinaryComparatorFactory;
+import edu.uci.ics.hyracks.dataflow.common.data.hash.UTF8StringBinaryHashFunctionFactory;
+import edu.uci.ics.hyracks.dataflow.common.data.marshalling.IntegerSerializerDeserializer;
+import edu.uci.ics.hyracks.dataflow.common.data.marshalling.UTF8StringSerializerDeserializer;
+import edu.uci.ics.hyracks.dataflow.common.data.partition.FieldHashPartitionComputerFactory;
+import edu.uci.ics.hyracks.dataflow.std.aggregators.CountAggregatorFactory;
+import edu.uci.ics.hyracks.dataflow.std.aggregators.IFieldValueResultingAggregatorFactory;
+import edu.uci.ics.hyracks.dataflow.std.aggregators.MultiAggregatorFactory;
+import edu.uci.ics.hyracks.dataflow.std.connectors.MToNHashPartitioningConnectorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.connectors.OneToOneConnectorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.file.ConstantFileSplitProvider;
+import edu.uci.ics.hyracks.dataflow.std.file.FileScanOperatorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.file.FileSplit;
+import edu.uci.ics.hyracks.dataflow.std.file.FrameFileWriterOperatorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.file.IFileSplitProvider;
+import edu.uci.ics.hyracks.dataflow.std.group.HashGroupOperatorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.group.PreclusteredGroupOperatorDescriptor;
+import edu.uci.ics.hyracks.dataflow.std.sort.ExternalSortOperatorDescriptor;
+import edu.uci.ics.hyracks.examples.text.WordTupleParserFactory;
+
+public class WordCountMain {
+ private static class Options {
+ @Option(name = "-host", usage = "Hyracks Cluster Controller Host name", required = true)
+ public String host;
+
+ @Option(name = "-port", usage = "Hyracks Cluster Controller Port (default: 1099)")
+ public int port = 1099;
+
+ @Option(name = "-app", usage = "Hyracks Application name", required = true)
+ public String app;
+
+ @Option(name = "-infile-splits", usage = "Comma separated list of file-splits for the input. A file-split is <node-name>:<path>", required = true)
+ public String inFileSplits;
+
+ @Option(name = "-outfile-splits", usage = "Comma separated list of file-splits for the output", required = true)
+ public String outFileSplits;
+
+ @Option(name = "-use-hash", usage = "Use Hash based grouping", required = true)
+ public boolean useHash;
+
+ @Option(name = "-hashtable-size", usage = "Hash table size (default: 8191)", required = false)
+ public int htSize = 8191;
+
+ @Option(name = "-sortbuffer-size", usage = "Sort buffer size in frames (default: 32768)", required = false)
+ public int sbSize = 32768;
+ }
+
+ public static void main(String[] args) throws Exception {
+ Options options = new Options();
+ CmdLineParser parser = new CmdLineParser(options);
+ parser.parseArgument(args);
+
+ IHyracksClientConnection hcc = new HyracksRMIConnection(options.host, options.port);
+
+ JobSpecification job = createJob(parseFileSplits(options.inFileSplits), parseFileSplits(options.outFileSplits),
+ options.useHash, options.htSize, options.sbSize);
+
+ long start = System.currentTimeMillis();
+ UUID jobId = hcc.createJob(options.app, job);
+ hcc.start(jobId);
+ hcc.waitForCompletion(jobId);
+ long end = System.currentTimeMillis();
+ System.err.println(start + " " + end + " " + (end - start));
+ }
+
+ private static FileSplit[] parseFileSplits(String fileSplits) {
+ String[] splits = fileSplits.split(",");
+ FileSplit[] fSplits = new FileSplit[splits.length];
+ for (int i = 0; i < splits.length; ++i) {
+ String s = splits[i].trim();
+ int idx = s.indexOf(':');
+ if (idx < 0) {
+ throw new IllegalArgumentException("File split " + s + " not well formed");
+ }
+ fSplits[i] = new FileSplit(s.substring(0, idx), new File(s.substring(idx + 1)));
+ }
+ return fSplits;
+ }
+
+ private static JobSpecification createJob(FileSplit[] inSplits, FileSplit[] outSplits, boolean useHash, int htSize,
+ int sbSize) {
+ JobSpecification spec = new JobSpecification();
+
+ IFileSplitProvider splitsProvider = new ConstantFileSplitProvider(inSplits);
+ RecordDescriptor wordDesc = new RecordDescriptor(
+ new ISerializerDeserializer[] { UTF8StringSerializerDeserializer.INSTANCE });
+
+ FileScanOperatorDescriptor wordScanner = new FileScanOperatorDescriptor(spec, splitsProvider,
+ new WordTupleParserFactory(), wordDesc);
+ wordScanner.setPartitionConstraint(createPartitionConstraint(inSplits));
+
+ RecordDescriptor groupResultDesc = new RecordDescriptor(new ISerializerDeserializer[] {
+ UTF8StringSerializerDeserializer.INSTANCE, IntegerSerializerDeserializer.INSTANCE });
+
+ IOperatorDescriptor gBy;
+ if (useHash) {
+ gBy = new HashGroupOperatorDescriptor(spec, new int[] { 0 },
+ new FieldHashPartitionComputerFactory(new int[] { 0 },
+ new IBinaryHashFunctionFactory[] { UTF8StringBinaryHashFunctionFactory.INSTANCE }),
+ new IBinaryComparatorFactory[] { UTF8StringBinaryComparatorFactory.INSTANCE },
+ new MultiAggregatorFactory(
+ new IFieldValueResultingAggregatorFactory[] { new CountAggregatorFactory() }),
+ groupResultDesc, htSize);
+ gBy.setPartitionConstraint(createPartitionConstraint(outSplits));
+ IConnectorDescriptor scanGroupConn = new MToNHashPartitioningConnectorDescriptor(spec,
+ new FieldHashPartitionComputerFactory(new int[] { 0 },
+ new IBinaryHashFunctionFactory[] { UTF8StringBinaryHashFunctionFactory.INSTANCE }));
+ spec.connect(scanGroupConn, wordScanner, 0, gBy, 0);
+ } else {
+ ExternalSortOperatorDescriptor sorter = new ExternalSortOperatorDescriptor(spec, sbSize, new int[] { 0 },
+ new IBinaryComparatorFactory[] { UTF8StringBinaryComparatorFactory.INSTANCE }, wordDesc);
+
+ IConnectorDescriptor scanSortConn = new MToNHashPartitioningConnectorDescriptor(spec,
+ new FieldHashPartitionComputerFactory(new int[] { 0 },
+ new IBinaryHashFunctionFactory[] { UTF8StringBinaryHashFunctionFactory.INSTANCE }));
+ spec.connect(scanSortConn, wordScanner, 0, sorter, 0);
+
+ gBy = new PreclusteredGroupOperatorDescriptor(spec, new int[] { 0 },
+ new IBinaryComparatorFactory[] { UTF8StringBinaryComparatorFactory.INSTANCE },
+ new MultiAggregatorFactory(
+ new IFieldValueResultingAggregatorFactory[] { new CountAggregatorFactory() }),
+ groupResultDesc);
+ OneToOneConnectorDescriptor sortGroupConn = new OneToOneConnectorDescriptor(spec);
+ spec.connect(sortGroupConn, sorter, 0, gBy, 0);
+ }
+
+ IFileSplitProvider outSplitProvider = new ConstantFileSplitProvider(outSplits);
+ FrameFileWriterOperatorDescriptor writer = new FrameFileWriterOperatorDescriptor(spec, outSplitProvider);
+ writer.setPartitionConstraint(createPartitionConstraint(outSplits));
+
+ IConnectorDescriptor gbyPrinterConn = new OneToOneConnectorDescriptor(spec);
+ spec.connect(gbyPrinterConn, gBy, 0, writer, 0);
+
+ spec.addRoot(writer);
+ return spec;
+ }
+
+ private static PartitionConstraint createPartitionConstraint(FileSplit[] splits) {
+ LocationConstraint[] lConstraints = new LocationConstraint[splits.length];
+ for (int i = 0; i < splits.length; ++i) {
+ lConstraints[i] = new AbsoluteLocationConstraint(splits[i].getNodeName());
+ }
+ return new ExplicitPartitionConstraint(lConstraints);
+ }
+}
\ No newline at end of file
diff --git a/hyracks-examples/text-example/texthelper/.classpath b/hyracks-examples/text-example/texthelper/.classpath
new file mode 100644
index 0000000..1f3c1ff
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/.classpath
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<classpath>
+ <classpathentry kind="src" output="target/classes" path="src/main/java"/>
+ <classpathentry kind="con" path="org.eclipse.jdt.launching.JRE_CONTAINER/org.eclipse.jdt.internal.debug.ui.launcher.StandardVMType/JavaSE-1.6"/>
+ <classpathentry kind="con" path="org.maven.ide.eclipse.MAVEN2_CLASSPATH_CONTAINER"/>
+ <classpathentry kind="output" path="target/classes"/>
+</classpath>
diff --git a/hyracks-examples/text-example/texthelper/.project b/hyracks-examples/text-example/texthelper/.project
new file mode 100644
index 0000000..19ce234
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/.project
@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<projectDescription>
+ <name>texthelper</name>
+ <comment></comment>
+ <projects>
+ </projects>
+ <buildSpec>
+ <buildCommand>
+ <name>org.eclipse.jdt.core.javabuilder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ <buildCommand>
+ <name>org.maven.ide.eclipse.maven2Builder</name>
+ <arguments>
+ </arguments>
+ </buildCommand>
+ </buildSpec>
+ <natures>
+ <nature>org.eclipse.jdt.core.javanature</nature>
+ <nature>org.maven.ide.eclipse.maven2Nature</nature>
+ </natures>
+</projectDescription>
diff --git a/hyracks-examples/text-example/texthelper/.settings/org.eclipse.jdt.core.prefs b/hyracks-examples/text-example/texthelper/.settings/org.eclipse.jdt.core.prefs
new file mode 100644
index 0000000..8599738
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/.settings/org.eclipse.jdt.core.prefs
@@ -0,0 +1,6 @@
+#Tue Sep 28 14:37:42 PDT 2010
+eclipse.preferences.version=1
+org.eclipse.jdt.core.compiler.codegen.targetPlatform=1.6
+org.eclipse.jdt.core.compiler.compliance=1.6
+org.eclipse.jdt.core.compiler.problem.forbiddenReference=warning
+org.eclipse.jdt.core.compiler.source=1.6
diff --git a/hyracks-examples/text-example/texthelper/.settings/org.maven.ide.eclipse.prefs b/hyracks-examples/text-example/texthelper/.settings/org.maven.ide.eclipse.prefs
new file mode 100644
index 0000000..4562b1a
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/.settings/org.maven.ide.eclipse.prefs
@@ -0,0 +1,9 @@
+#Tue Sep 28 14:37:42 PDT 2010
+activeProfiles=
+eclipse.preferences.version=1
+fullBuildGoals=process-test-resources
+includeModules=false
+resolveWorkspaceProjects=true
+resourceFilterGoals=process-resources resources\:testResources
+skipCompilerPlugin=true
+version=1
diff --git a/hyracks-examples/text-example/texthelper/pom.xml b/hyracks-examples/text-example/texthelper/pom.xml
new file mode 100644
index 0000000..b0dae86
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/pom.xml
@@ -0,0 +1,40 @@
+<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
+ <modelVersion>4.0.0</modelVersion>
+ <groupId>edu.uci.ics.hyracks.examples.text</groupId>
+ <artifactId>texthelper</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+
+ <parent>
+ <groupId>edu.uci.ics.hyracks.examples</groupId>
+ <artifactId>text-example</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ </parent>
+
+ <dependencies>
+ <dependency>
+ <groupId>edu.uci.ics.hyracks</groupId>
+ <artifactId>hyracks-dataflow-std</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <scope>compile</scope>
+ </dependency>
+ <dependency>
+ <groupId>edu.uci.ics.hyracks</groupId>
+ <artifactId>hyracks-api</artifactId>
+ <version>0.1.3-SNAPSHOT</version>
+ <scope>compile</scope>
+ </dependency>
+ </dependencies>
+ <build>
+ <plugins>
+ <plugin>
+ <groupId>org.apache.maven.plugins</groupId>
+ <artifactId>maven-compiler-plugin</artifactId>
+ <version>2.0.2</version>
+ <configuration>
+ <source>1.6</source>
+ <target>1.6</target>
+ </configuration>
+ </plugin>
+ </plugins>
+ </build>
+</project>
diff --git a/hyracks-examples/text-example/texthelper/src/main/java/edu/uci/ics/hyracks/examples/text/WordTupleParserFactory.java b/hyracks-examples/text-example/texthelper/src/main/java/edu/uci/ics/hyracks/examples/text/WordTupleParserFactory.java
new file mode 100644
index 0000000..9825fdf
--- /dev/null
+++ b/hyracks-examples/text-example/texthelper/src/main/java/edu/uci/ics/hyracks/examples/text/WordTupleParserFactory.java
@@ -0,0 +1,128 @@
+package edu.uci.ics.hyracks.examples.text;
+
+import java.io.DataOutput;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.io.Reader;
+import java.nio.ByteBuffer;
+import java.util.Arrays;
+
+import edu.uci.ics.hyracks.api.comm.IFrameWriter;
+import edu.uci.ics.hyracks.api.context.IHyracksContext;
+import edu.uci.ics.hyracks.api.exceptions.HyracksDataException;
+import edu.uci.ics.hyracks.dataflow.common.comm.io.ArrayTupleBuilder;
+import edu.uci.ics.hyracks.dataflow.common.comm.io.FrameTupleAppender;
+import edu.uci.ics.hyracks.dataflow.common.comm.util.FrameUtils;
+import edu.uci.ics.hyracks.dataflow.common.data.parsers.IValueParser;
+import edu.uci.ics.hyracks.dataflow.common.data.parsers.UTF8StringParserFactory;
+import edu.uci.ics.hyracks.dataflow.std.file.ITupleParser;
+import edu.uci.ics.hyracks.dataflow.std.file.ITupleParserFactory;
+
+public class WordTupleParserFactory implements ITupleParserFactory {
+ private static final long serialVersionUID = 1L;
+
+ @Override
+ public ITupleParser createTupleParser(final IHyracksContext ctx) {
+ return new ITupleParser() {
+ @Override
+ public void parse(InputStream in, IFrameWriter writer) throws HyracksDataException {
+ try {
+ ByteBuffer frame = ctx.getResourceManager().allocateFrame();
+ FrameTupleAppender appender = new FrameTupleAppender(ctx);
+ appender.reset(frame, true);
+ ArrayTupleBuilder tb = new ArrayTupleBuilder(1);
+ DataOutput dos = tb.getDataOutput();
+
+ IValueParser utf8StringParser = UTF8StringParserFactory.INSTANCE.createValueParser();
+ WordCursor cursor = new WordCursor(new InputStreamReader(in));
+ while (cursor.nextWord()) {
+ tb.reset();
+ utf8StringParser.parse(cursor.buffer, cursor.fStart, cursor.fEnd - cursor.fStart, dos);
+ tb.addFieldEndOffset();
+ if (!appender.append(tb.getFieldEndOffsets(), tb.getByteArray(), 0, tb.getSize())) {
+ FrameUtils.flushFrame(frame, writer);
+ appender.reset(frame, true);
+ if (!appender.append(tb.getFieldEndOffsets(), tb.getByteArray(), 0, tb.getSize())) {
+ throw new IllegalStateException();
+ }
+ }
+ }
+ if (appender.getTupleCount() > 0) {
+ FrameUtils.flushFrame(frame, writer);
+ }
+ } catch (IOException e) {
+ throw new HyracksDataException(e);
+ }
+ }
+ };
+ }
+
+ private static class WordCursor {
+ private static final int INITIAL_BUFFER_SIZE = 4096;
+ private static final int INCREMENT = 4096;
+
+ private char[] buffer;
+
+ private int start;
+ private int end;
+ private boolean eof;
+
+ private int fStart;
+ private int fEnd;
+ private Reader in;
+
+ public WordCursor(Reader in) {
+ this.in = in;
+ buffer = new char[INITIAL_BUFFER_SIZE];
+ start = 0;
+ end = 0;
+ eof = false;
+ }
+
+ public boolean nextWord() throws IOException {
+ if (eof) {
+ return false;
+ }
+
+ int p = start;
+ while (true) {
+ if (p >= end) {
+ int s = start;
+ eof = !readMore();
+ if (eof) {
+ return true;
+ }
+ p -= (s - start);
+ }
+ char ch = buffer[p];
+ if (Character.isWhitespace(ch)) {
+ fStart = start;
+ fEnd = p;
+ start = p + 1;
+ return true;
+ }
+ ++p;
+ }
+ }
+
+ private boolean readMore() throws IOException {
+ if (start > 0) {
+ System.arraycopy(buffer, start, buffer, 0, end - start);
+ }
+ end -= start;
+ start = 0;
+
+ if (end == buffer.length) {
+ buffer = Arrays.copyOf(buffer, buffer.length + INCREMENT);
+ }
+
+ int n = in.read(buffer, end, buffer.length - end);
+ if (n < 0) {
+ return false;
+ }
+ end += n;
+ return true;
+ }
+ }
+}
\ No newline at end of file