use of co.cask.cdap.data.dataset.SystemDatasetInstantiator in project cdap by caskdata.
the class ExistingEntitySystemMetadataWriter method writeSystemMetadataForDatasets.
private void writeSystemMetadataForDatasets(NamespaceId namespace, DatasetFramework dsFramework) throws DatasetManagementException, IOException, NamespaceNotFoundException {
SystemDatasetInstantiatorFactory systemDatasetInstantiatorFactory = new SystemDatasetInstantiatorFactory(locationFactory, dsFramework, cConf);
try (SystemDatasetInstantiator systemDatasetInstantiator = systemDatasetInstantiatorFactory.create()) {
for (DatasetSpecificationSummary summary : dsFramework.getInstances(namespace)) {
final DatasetId dsInstance = namespace.dataset(summary.getName());
DatasetProperties dsProperties = DatasetProperties.of(summary.getProperties());
String dsType = summary.getType();
Dataset dataset = null;
try {
try {
dataset = impersonator.doAs(dsInstance, new Callable<Dataset>() {
@Override
public Dataset call() throws Exception {
return systemDatasetInstantiator.getDataset(dsInstance);
}
});
} catch (Exception e) {
LOG.warn("Exception while instantiating dataset {}", dsInstance, e);
}
SystemMetadataWriter writer = new DatasetSystemMetadataWriter(metadataStore, dsInstance, dsProperties, dataset, dsType, summary.getDescription());
writer.write();
} finally {
if (dataset != null) {
dataset.close();
}
}
}
}
}
use of co.cask.cdap.data.dataset.SystemDatasetInstantiator in project cdap by caskdata.
the class ExploreExecutorHttpHandler method doAddPartition.
private void doAddPartition(HttpRequest request, HttpResponder responder, DatasetId datasetId) {
Dataset dataset;
try (SystemDatasetInstantiator datasetInstantiator = datasetInstantiatorFactory.create()) {
dataset = datasetInstantiator.getDataset(datasetId);
if (dataset == null) {
responder.sendString(HttpResponseStatus.NOT_FOUND, "Cannot load dataset " + datasetId);
return;
}
} catch (IOException e) {
String classNotFoundMessage = isClassNotFoundException(e);
if (classNotFoundMessage != null) {
JsonObject json = new JsonObject();
json.addProperty("handle", QueryHandle.NO_OP.getHandle());
responder.sendJson(HttpResponseStatus.OK, json);
return;
}
LOG.error("Exception instantiating dataset {}.", datasetId, e);
responder.sendString(HttpResponseStatus.INTERNAL_SERVER_ERROR, "Exception instantiating dataset " + datasetId.getDataset());
return;
}
try {
if (!(dataset instanceof PartitionedFileSet)) {
responder.sendString(HttpResponseStatus.BAD_REQUEST, "not a partitioned dataset.");
return;
}
Partitioning partitioning = ((PartitionedFileSet) dataset).getPartitioning();
Reader reader = new InputStreamReader(new ChannelBufferInputStream(request.getContent()));
Map<String, String> properties = GSON.fromJson(reader, new TypeToken<Map<String, String>>() {
}.getType());
String fsPath = properties.get("path");
if (fsPath == null) {
responder.sendString(HttpResponseStatus.BAD_REQUEST, "path was not specified.");
return;
}
PartitionKey partitionKey;
try {
partitionKey = PartitionedFileSetArguments.getOutputPartitionKey(properties, partitioning);
} catch (Exception e) {
responder.sendString(HttpResponseStatus.BAD_REQUEST, "invalid partition key: " + e.getMessage());
return;
}
if (partitionKey == null) {
responder.sendString(HttpResponseStatus.BAD_REQUEST, "no partition key was given.");
return;
}
QueryHandle handle = exploreTableManager.addPartition(datasetId, properties, partitionKey, fsPath);
JsonObject json = new JsonObject();
json.addProperty("handle", handle.getHandle());
responder.sendJson(HttpResponseStatus.OK, json);
} catch (Throwable e) {
LOG.error("Got exception:", e);
responder.sendString(HttpResponseStatus.INTERNAL_SERVER_ERROR, e.getMessage());
}
}
use of co.cask.cdap.data.dataset.SystemDatasetInstantiator in project cdap by caskdata.
the class HiveExploreStructuredRecordTestRun method start.
@BeforeClass
public static void start() throws Exception {
initialize(tmpFolder);
DatasetModuleId moduleId = NAMESPACE_ID.datasetModule("email");
datasetFramework.addModule(moduleId, new EmailTableDefinition.EmailTableModule());
datasetFramework.addInstance("email", MY_TABLE, DatasetProperties.EMPTY);
transactional = Transactions.createTransactional(new MultiThreadDatasetCache(new SystemDatasetInstantiator(datasetFramework), transactionSystemClient, NAMESPACE_ID, Collections.<String, String>emptyMap(), null, null));
transactional.execute(new TxRunnable() {
@Override
public void run(DatasetContext context) throws Exception {
// Accessing dataset instance to perform data operations
EmailTableDefinition.EmailTable table = context.getDataset(MY_TABLE.getDataset());
Assert.assertNotNull(table);
table.writeEmail("email1", "this is the subject", "this is the body", "sljackson@boss.com");
}
});
datasetFramework.addModule(NAMESPACE_ID.datasetModule("TableWrapper"), new TableWrapperDefinition.Module());
}
use of co.cask.cdap.data.dataset.SystemDatasetInstantiator in project cdap by caskdata.
the class ExploreTableManager method generateDisableStatement.
private String generateDisableStatement(DatasetId datasetId, DatasetSpecification spec) throws ExploreException {
String tableName = tableNaming.getTableName(datasetId, spec.getProperties());
String databaseName = ExploreProperties.getExploreDatabaseName(spec.getProperties());
// If table does not exist, nothing to be done
try {
exploreService.getTableInfo(datasetId.getNamespace(), databaseName, tableName);
} catch (TableNotFoundException e) {
// Ignore exception, since this means table was not found.
return null;
}
try (SystemDatasetInstantiator datasetInstantiator = datasetInstantiatorFactory.create()) {
Dataset dataset = datasetInstantiator.getDataset(datasetId);
try {
if (dataset instanceof FileSet || dataset instanceof PartitionedFileSet) {
// do not drop the explore table that dataset is reusing an existing table
if (FileSetProperties.isUseExisting(spec.getProperties())) {
return null;
}
}
return generateDeleteStatement(dataset, databaseName, tableName);
} finally {
Closeables.closeQuietly(dataset);
}
} catch (IOException e) {
LOG.error("Exception creating dataset classLoaderProvider for dataset {}.", datasetId, e);
throw new ExploreException("Exception instantiating dataset " + datasetId);
}
}
use of co.cask.cdap.data.dataset.SystemDatasetInstantiator in project cdap by caskdata.
the class FileMetadataCleanerTest method testFileMetadataWithCommonContextPrefix.
@Test
public void testFileMetadataWithCommonContextPrefix() throws Exception {
DatasetFramework datasetFramework = injector.getInstance(DatasetFramework.class);
DatasetManager datasetManager = new DefaultDatasetManager(datasetFramework, NamespaceId.SYSTEM, co.cask.cdap.common.service.RetryStrategies.noRetry(), null);
Transactional transactional = Transactions.createTransactionalWithRetry(Transactions.createTransactional(new MultiThreadDatasetCache(new SystemDatasetInstantiator(datasetFramework), injector.getInstance(TransactionSystemClient.class), NamespaceId.SYSTEM, ImmutableMap.<String, String>of(), null, null)), RetryStrategies.retryOnConflict(20, 100));
FileMetaDataWriter fileMetaDataWriter = new FileMetaDataWriter(datasetManager, transactional);
FileMetaDataReader fileMetadataReader = injector.getInstance(FileMetaDataReader.class);
FileMetadataCleaner fileMetadataCleaner = new FileMetadataCleaner(datasetManager, transactional);
try {
List<LogPathIdentifier> logPathIdentifiers = new ArrayList<>();
// this should be able to scan and delete common prefix programs like testFlow1, testFlow10 during clenaup.
for (int i = 1; i <= 20; i++) {
logPathIdentifiers.add(new LogPathIdentifier(NamespaceId.DEFAULT.getNamespace(), "testApp", String.format("testFlow%s", i)));
}
LocationFactory locationFactory = injector.getInstance(LocationFactory.class);
Location location = locationFactory.create(TMP_FOLDER.newFolder().getPath()).append("/logs");
long currentTime = System.currentTimeMillis();
long newCurrentTime = currentTime + 100;
for (int i = 1; i <= 20; i++) {
LogPathIdentifier identifier = logPathIdentifiers.get(i - 1);
for (int j = 0; j < 10; j++) {
fileMetaDataWriter.writeMetaData(identifier, newCurrentTime + j, newCurrentTime + j, location.append("testFileNew" + Integer.toString(j)));
}
}
List<LogLocation> locations;
for (int i = 1; i <= 20; i++) {
locations = fileMetadataReader.listFiles(logPathIdentifiers.get(i - 1), newCurrentTime, newCurrentTime + 10);
// should include files from currentTime (0..9)
Assert.assertEquals(10, locations.size());
}
long tillTime = newCurrentTime + 4;
List<FileMetadataCleaner.DeletedEntry> deleteEntries = fileMetadataCleaner.scanAndGetFilesToDelete(tillTime, TRANSACTION_TIMEOUT);
// 20 context, 5 entries each
Assert.assertEquals(100, deleteEntries.size());
for (int i = 1; i <= 20; i++) {
locations = fileMetadataReader.listFiles(logPathIdentifiers.get(i - 1), newCurrentTime, newCurrentTime + 10);
// should include files from time (5..9)
Assert.assertEquals(5, locations.size());
int startIndex = 5;
for (LogLocation logLocation : locations) {
Assert.assertEquals(String.format("testFileNew%s", startIndex), logLocation.getLocation().getName());
startIndex++;
}
}
} finally {
// cleanup meta
cleanupMetadata(transactional, datasetManager);
}
}
Aggregations