use of org.apache.commons.io.output.ByteArrayOutputStream in project Gaffer by gchq.
the class ImportJavaRDDOfElementsHandlerTest method checkImportJavaRDDOfElements.
@Test
public void checkImportJavaRDDOfElements() throws OperationException, IOException, InterruptedException {
final Graph graph1 = new Graph.Builder().addSchema(getClass().getResourceAsStream("/schema/dataSchema.json")).addSchema(getClass().getResourceAsStream("/schema/dataTypes.json")).addSchema(getClass().getResourceAsStream("/schema/storeTypes.json")).storeProperties(getClass().getResourceAsStream("/store.properties")).build();
final List<Element> elements = new ArrayList<>();
for (int i = 0; i < 10; i++) {
final Entity entity = new Entity(TestGroups.ENTITY);
entity.setVertex("" + i);
final Edge edge1 = new Edge(TestGroups.EDGE);
edge1.setSource("" + i);
edge1.setDestination("B");
edge1.setDirected(false);
edge1.putProperty(TestPropertyNames.COUNT, 2);
final Edge edge2 = new Edge(TestGroups.EDGE);
edge2.setSource("" + i);
edge2.setDestination("C");
edge2.setDirected(false);
edge2.putProperty(TestPropertyNames.COUNT, 4);
elements.add(edge1);
elements.add(edge2);
elements.add(entity);
}
final User user = new User();
final SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("testCheckGetCorrectElementsInJavaRDDForEntitySeed").set("spark.serializer", "org.apache.spark.serializer.KryoSerializer").set("spark.kryo.registrator", "uk.gov.gchq.gaffer.spark.serialisation.kryo.Registrator").set("spark.driver.allowMultipleContexts", "true");
final JavaSparkContext sparkContext = new JavaSparkContext(sparkConf);
// Create Hadoop configuration and serialise to a string
final Configuration configuration = new Configuration();
final ByteArrayOutputStream baos = new ByteArrayOutputStream();
configuration.write(new DataOutputStream(baos));
final String configurationString = new String(baos.toByteArray(), CommonConstants.UTF_8);
final String outputPath = this.getClass().getResource("/").getPath().toString() + "load";
final String failurePath = this.getClass().getResource("/").getPath().toString() + "failure";
final File file = new File(outputPath);
if (file.exists()) {
FileUtils.forceDelete(file);
}
final JavaRDD<Element> elementJavaRDD = sparkContext.parallelize(elements);
final ImportJavaRDDOfElements addRdd = new ImportJavaRDDOfElements.Builder().javaSparkContext(sparkContext).input(elementJavaRDD).option("outputPath", outputPath).option("failurePath", failurePath).build();
graph1.execute(addRdd, user);
FileUtils.forceDelete(file);
// Check all elements were added
final GetJavaRDDOfAllElements rddQuery = new GetJavaRDDOfAllElements.Builder().javaSparkContext(sparkContext).option(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString).build();
final JavaRDD<Element> rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
final Set<Element> results = new HashSet<>(rdd.collect());
assertEquals(elements.size(), results.size());
sparkContext.stop();
}
use of org.apache.commons.io.output.ByteArrayOutputStream in project Gaffer by gchq.
the class ImportKeyValueJavaPairRDDToAccumuloHandlerTest method checkImportKeyValueJavaPairRDD.
@Test
public void checkImportKeyValueJavaPairRDD() throws OperationException, IOException, InterruptedException {
final Graph graph1 = new Graph.Builder().addSchema(getClass().getResourceAsStream("/schema/dataSchema.json")).addSchema(getClass().getResourceAsStream("/schema/dataTypes.json")).addSchema(getClass().getResourceAsStream("/schema/storeSchema.json")).addSchema(getClass().getResourceAsStream("/schema/storeTypes.json")).storeProperties(getClass().getResourceAsStream("/store.properties")).build();
final List<Element> elements = new ArrayList<>();
for (int i = 0; i < 10; i++) {
final Entity entity = new Entity(TestGroups.ENTITY);
entity.setVertex("" + i);
final Edge edge1 = new Edge(TestGroups.EDGE);
edge1.setSource("" + i);
edge1.setDestination("B");
edge1.setDirected(false);
edge1.putProperty(TestPropertyNames.COUNT, 2);
final Edge edge2 = new Edge(TestGroups.EDGE);
edge2.setSource("" + i);
edge2.setDestination("C");
edge2.setDirected(false);
edge2.putProperty(TestPropertyNames.COUNT, 4);
elements.add(edge1);
elements.add(edge2);
elements.add(entity);
}
final User user = new User();
final SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("testCheckGetCorrectElementsInJavaRDDForEntitySeed").set("spark.serializer", "org.apache.spark.serializer.KryoSerializer").set("spark.kryo.registrator", "uk.gov.gchq.gaffer.spark.serialisation.kryo.Registrator").set("spark.driver.allowMultipleContexts", "true");
final JavaSparkContext sparkContext = new JavaSparkContext(sparkConf);
// Create Hadoop configuration and serialise to a string
final Configuration configuration = new Configuration();
final ByteArrayOutputStream baos = new ByteArrayOutputStream();
configuration.write(new DataOutputStream(baos));
final String configurationString = new String(baos.toByteArray(), CommonConstants.UTF_8);
final String outputPath = this.getClass().getResource("/").getPath().toString() + "load";
final String failurePath = this.getClass().getResource("/").getPath().toString() + "failure";
final File file = new File(outputPath);
if (file.exists()) {
FileUtils.forceDelete(file);
}
final ElementConverterFunction func = new ElementConverterFunction(sparkContext.broadcast(new ByteEntityAccumuloElementConverter(graph1.getSchema())));
final JavaPairRDD<Key, Value> elementJavaRDD = sparkContext.parallelize(elements).flatMapToPair(func);
final ImportKeyValueJavaPairRDDToAccumulo addRdd = new ImportKeyValueJavaPairRDDToAccumulo.Builder().input(elementJavaRDD).outputPath(outputPath).failurePath(failurePath).build();
graph1.execute(addRdd, user);
FileUtils.forceDelete(file);
// Check all elements were added
final GetJavaRDDOfAllElements rddQuery = new GetJavaRDDOfAllElements.Builder().javaSparkContext(sparkContext).option(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString).build();
final JavaRDD<Element> rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
final Set<Element> results = new HashSet<>(rdd.collect());
assertEquals(elements.size(), results.size());
sparkContext.stop();
}
use of org.apache.commons.io.output.ByteArrayOutputStream in project Gaffer by gchq.
the class GetJavaRDDOfElementsHandlerTest method checkGetCorrectElementsInJavaRDDForEntitySeed.
@Test
public void checkGetCorrectElementsInJavaRDDForEntitySeed() throws OperationException, IOException {
final Graph graph1 = new Graph.Builder().addSchema(getClass().getResourceAsStream("/schema/dataSchema.json")).addSchema(getClass().getResourceAsStream("/schema/dataTypes.json")).addSchema(getClass().getResourceAsStream("/schema/storeTypes.json")).storeProperties(getClass().getResourceAsStream("/store.properties")).build();
final List<Element> elements = new ArrayList<>();
for (int i = 0; i < 10; i++) {
final Entity entity = new Entity(ENTITY_GROUP);
entity.setVertex("" + i);
final Edge edge1 = new Edge(EDGE_GROUP);
edge1.setSource("" + i);
edge1.setDestination("B");
edge1.setDirected(false);
edge1.putProperty("count", 2);
final Edge edge2 = new Edge(EDGE_GROUP);
edge2.setSource("" + i);
edge2.setDestination("C");
edge2.setDirected(false);
edge2.putProperty("count", 4);
elements.add(edge1);
elements.add(edge2);
elements.add(entity);
}
final User user = new User();
graph1.execute(new AddElements(elements), user);
final SparkConf sparkConf = new SparkConf().setMaster("local").setAppName("testCheckGetCorrectElementsInJavaRDDForEntitySeed").set("spark.serializer", "org.apache.spark.serializer.KryoSerializer").set("spark.kryo.registrator", "uk.gov.gchq.gaffer.spark.serialisation.kryo.Registrator").set("spark.driver.allowMultipleContexts", "true");
final JavaSparkContext sparkContext = new JavaSparkContext(sparkConf);
// Create Hadoop configuration and serialise to a string
final Configuration configuration = new Configuration();
final ByteArrayOutputStream baos = new ByteArrayOutputStream();
configuration.write(new DataOutputStream(baos));
final String configurationString = new String(baos.toByteArray(), CommonConstants.UTF_8);
// Check get correct edges for "1"
GetJavaRDDOfElements<EntitySeed> rddQuery = new GetJavaRDDOfElements.Builder<EntitySeed>().javaSparkContext(sparkContext).seeds(Collections.singleton(new EntitySeed("1"))).build();
rddQuery.addOption(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString);
JavaRDD<Element> rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
final Set<Element> results = new HashSet<>(rdd.collect());
final Set<Element> expectedElements = new HashSet<>();
final Entity entity1 = new Entity(ENTITY_GROUP);
entity1.setVertex("1");
final Edge edge1B = new Edge(EDGE_GROUP);
edge1B.setSource("1");
edge1B.setDestination("B");
edge1B.setDirected(false);
edge1B.putProperty("count", 2);
final Edge edge1C = new Edge(EDGE_GROUP);
edge1C.setSource("1");
edge1C.setDestination("C");
edge1C.setDirected(false);
edge1C.putProperty("count", 4);
expectedElements.add(entity1);
expectedElements.add(edge1B);
expectedElements.add(edge1C);
assertEquals(expectedElements, results);
// Check get correct edges for "1" when specify entities only
rddQuery = new GetJavaRDDOfElements.Builder<EntitySeed>().javaSparkContext(sparkContext).seeds(Collections.singleton(new EntitySeed("1"))).view(new View.Builder().entity(ENTITY_GROUP).build()).build();
rddQuery.addOption(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString);
rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
results.clear();
results.addAll(rdd.collect());
expectedElements.clear();
expectedElements.add(entity1);
assertEquals(expectedElements, results);
// Check get correct edges for "1" when specify edges only
rddQuery = new GetJavaRDDOfElements.Builder<EntitySeed>().javaSparkContext(sparkContext).seeds(Collections.singleton(new EntitySeed("1"))).view(new View.Builder().edge(EDGE_GROUP).build()).build();
rddQuery.addOption(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString);
rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
results.clear();
results.addAll(rdd.collect());
expectedElements.clear();
expectedElements.add(edge1B);
expectedElements.add(edge1C);
assertEquals(expectedElements, results);
// Check get correct edges for "1" and "5"
Set<EntitySeed> seeds = new HashSet<>();
seeds.add(new EntitySeed("1"));
seeds.add(new EntitySeed("5"));
rddQuery = new GetJavaRDDOfElements.Builder<EntitySeed>().javaSparkContext(sparkContext).seeds(seeds).build();
rddQuery.addOption(AbstractGetRDDHandler.HADOOP_CONFIGURATION_KEY, configurationString);
rdd = graph1.execute(rddQuery, user);
if (rdd == null) {
fail("No RDD returned");
}
results.clear();
results.addAll(rdd.collect());
final Entity entity5 = new Entity(ENTITY_GROUP);
entity5.setVertex("5");
final Edge edge5B = new Edge(EDGE_GROUP);
edge5B.setSource("5");
edge5B.setDestination("B");
edge5B.setDirected(false);
edge5B.putProperty("count", 2);
final Edge edge5C = new Edge(EDGE_GROUP);
edge5C.setSource("5");
edge5C.setDestination("C");
edge5C.setDirected(false);
edge5C.putProperty("count", 4);
expectedElements.clear();
expectedElements.add(entity1);
expectedElements.add(edge1B);
expectedElements.add(edge1C);
expectedElements.add(entity5);
expectedElements.add(edge5B);
expectedElements.add(edge5C);
assertEquals(expectedElements, results);
sparkContext.stop();
}
use of org.apache.commons.io.output.ByteArrayOutputStream in project hudson-2.x by hudson.
the class ZFSInstaller method doStart.
/**
* Called from the confirmation screen to actually initiate the migration.
*/
public void doStart(StaplerRequest req, StaplerResponse rsp, @QueryParameter String username, @QueryParameter String password) throws ServletException, IOException {
requirePOST();
Hudson hudson = Hudson.getInstance();
hudson.checkPermission(Hudson.ADMINISTER);
final String datasetName;
ByteArrayOutputStream log = new ByteArrayOutputStream();
StreamTaskListener listener = new StreamTaskListener(log);
try {
datasetName = createZfsFileSystem(listener, username, password);
} catch (Exception e) {
e.printStackTrace(listener.error(e.getMessage()));
if (e instanceof ZFSException) {
ZFSException ze = (ZFSException) e;
if (ze.getCode() == ErrorCode.EZFS_PERM) {
// permission problem. ask the user to give us the root password
req.setAttribute("message", log.toString());
rsp.forward(this, "askRootPassword", req);
return;
}
}
// for other kinds of problems, report and bail out
req.setAttribute("pre", true);
sendError(log.toString(), req, rsp);
return;
}
// file system creation successful, so restart
WebAppController.get().install(new HudsonIsRestarting());
// redirect the user to the manage page
rsp.sendRedirect2(req.getContextPath() + "/manage");
// asynchronously restart, so that we can give a bit of time to the browser to load "restarting..." screen.
new Thread("restart thread") {
@Override
public void run() {
try {
Thread.sleep(5000);
// close all descriptors on exec except stdin,out,err
int sz = LIBC.getdtablesize();
for (int i = 3; i < sz; i++) {
int flags = LIBC.fcntl(i, F_GETFD);
if (flags < 0)
continue;
LIBC.fcntl(i, F_SETFD, flags | FD_CLOEXEC);
}
// re-exec with the system property to indicate where to migrate the data to.
// the 2nd phase is implemented in the migrate method.
JavaVMArguments args = JavaVMArguments.current();
args.setSystemProperty(ZFSInstaller.class.getName() + ".migrate", datasetName);
Daemon.selfExec(args);
} catch (InterruptedException e) {
LOGGER.log(Level.SEVERE, "Restart failed", e);
} catch (IOException e) {
LOGGER.log(Level.SEVERE, "Restart failed", e);
}
}
}.start();
}
use of org.apache.commons.io.output.ByteArrayOutputStream in project hive by apache.
the class TestLazyAccumuloMap method toBytes.
protected byte[] toBytes(int i) throws IOException {
ByteArrayOutputStream baos = new ByteArrayOutputStream();
DataOutputStream out = new DataOutputStream(baos);
out.writeInt(i);
out.close();
return baos.toByteArray();
}
Aggregations