use of org.apache.cassandra.io.util.TrackedDataInputPlus in project cassandra by apache.
the class CassandraCompressedStreamReader method read.
/**
* @return SSTable transferred
* @throws java.io.IOException if reading the remote sstable fails. Will throw an RTE if local write fails.
*/
@Override
// input needs to remain open, streams on top of it can't be closed
@SuppressWarnings("resource")
public SSTableMultiWriter read(DataInputPlus inputPlus) throws Throwable {
long totalSize = totalSize();
ColumnFamilyStore cfs = ColumnFamilyStore.getIfExists(tableId);
if (cfs == null) {
// schema was dropped during streaming
throw new IOException("CF " + tableId + " was dropped during streaming");
}
logger.debug("[Stream #{}] Start receiving file #{} from {}, repairedAt = {}, size = {}, ks = '{}', pendingRepair = '{}', table = '{}'.", session.planId(), fileSeqNum, session.peer, repairedAt, totalSize, cfs.keyspace.getName(), pendingRepair, cfs.getTableName());
StreamDeserializer deserializer = null;
SSTableMultiWriter writer = null;
try (CompressedInputStream cis = new CompressedInputStream(inputPlus, compressionInfo, ChecksumType.CRC32, cfs::getCrcCheckChance)) {
TrackedDataInputPlus in = new TrackedDataInputPlus(cis);
deserializer = new StreamDeserializer(cfs.metadata(), in, inputVersion, getHeader(cfs.metadata()));
writer = createWriter(cfs, totalSize, repairedAt, pendingRepair, format);
String filename = writer.getFilename();
int sectionIdx = 0;
for (SSTableReader.PartitionPositionBounds section : sections) {
assert cis.chunkBytesRead() <= totalSize;
long sectionLength = section.upperPosition - section.lowerPosition;
logger.trace("[Stream #{}] Reading section {} with length {} from stream.", session.planId(), sectionIdx++, sectionLength);
// skip to beginning of section inside chunk
cis.position(section.lowerPosition);
in.reset(0);
while (in.getBytesRead() < sectionLength) {
writePartition(deserializer, writer);
// when compressed, report total bytes of compressed chunks read since remoteFile.size is the sum of chunks transferred
session.progress(filename + '-' + fileSeqNum, ProgressInfo.Direction.IN, cis.chunkBytesRead(), totalSize);
}
assert in.getBytesRead() == sectionLength;
}
logger.info("[Stream #{}] Finished receiving file #{} from {} readBytes = {}, totalSize = {}", session.planId(), fileSeqNum, session.peer, FBUtilities.prettyPrintMemory(cis.chunkBytesRead()), FBUtilities.prettyPrintMemory(totalSize));
return writer;
} catch (Throwable e) {
Object partitionKey = deserializer != null ? deserializer.partitionKey() : "";
logger.warn("[Stream {}] Error while reading partition {} from stream on ks='{}' and table='{}'.", session.planId(), partitionKey, cfs.keyspace.getName(), cfs.getTableName());
if (writer != null)
e = writer.abort(e);
throw e;
}
}
use of org.apache.cassandra.io.util.TrackedDataInputPlus in project cassandra by apache.
the class CassandraStreamReader method read.
/**
* @param inputPlus where this reads data from
* @return SSTable transferred
* @throws IOException if reading the remote sstable fails. Will throw an RTE if local write fails.
*/
// input needs to remain open, streams on top of it can't be closed
@SuppressWarnings("resource")
@Override
public SSTableMultiWriter read(DataInputPlus inputPlus) throws Throwable {
long totalSize = totalSize();
ColumnFamilyStore cfs = ColumnFamilyStore.getIfExists(tableId);
if (cfs == null)
// schema was dropped during streaming
throw new IllegalStateException("Table " + tableId + " was dropped during streaming");
logger.debug("[Stream #{}] Start receiving file #{} from {}, repairedAt = {}, size = {}, ks = '{}', table = '{}', pendingRepair = '{}'.", session.planId(), fileSeqNum, session.peer, repairedAt, totalSize, cfs.keyspace.getName(), cfs.getTableName(), pendingRepair);
StreamDeserializer deserializer = null;
SSTableMultiWriter writer = null;
try (StreamCompressionInputStream streamCompressionInputStream = new StreamCompressionInputStream(inputPlus, current_version)) {
TrackedDataInputPlus in = new TrackedDataInputPlus(streamCompressionInputStream);
deserializer = new StreamDeserializer(cfs.metadata(), in, inputVersion, getHeader(cfs.metadata()));
writer = createWriter(cfs, totalSize, repairedAt, pendingRepair, format);
while (in.getBytesRead() < totalSize) {
writePartition(deserializer, writer);
// TODO move this to BytesReadTracker
session.progress(writer.getFilename() + '-' + fileSeqNum, ProgressInfo.Direction.IN, in.getBytesRead(), totalSize);
}
logger.debug("[Stream #{}] Finished receiving file #{} from {} readBytes = {}, totalSize = {}", session.planId(), fileSeqNum, session.peer, FBUtilities.prettyPrintMemory(in.getBytesRead()), FBUtilities.prettyPrintMemory(totalSize));
return writer;
} catch (Throwable e) {
Object partitionKey = deserializer != null ? deserializer.partitionKey() : "";
logger.warn("[Stream {}] Error while reading partition {} from stream on ks='{}' and table='{}'.", session.planId(), partitionKey, cfs.keyspace.getName(), cfs.getTableName(), e);
if (writer != null)
e = writer.abort(e);
throw e;
}
}
use of org.apache.cassandra.io.util.TrackedDataInputPlus in project cassandra by apache.
the class BytesReadTrackerTest method internalTestSkipBytesAndReadFully.
public void internalTestSkipBytesAndReadFully(boolean inputStream) throws Exception {
String testStr = "1234567890";
byte[] testData = testStr.getBytes();
DataInputPlus.DataInputStreamPlus in = new DataInputPlus.DataInputStreamPlus(new ByteArrayInputStream(testData));
BytesReadTracker tracker = inputStream ? new TrackedInputStream(in) : new TrackedDataInputPlus(in);
DataInputPlus reader = inputStream ? new DataInputPlus.DataInputStreamPlus((TrackedInputStream) tracker) : (DataInputPlus) tracker;
try {
// read first 5 bytes
byte[] out = new byte[5];
reader.readFully(out, 0, 5);
assertEquals("12345", new String(out));
assertEquals(5, tracker.getBytesRead());
// then skip 2 bytes
reader.skipBytes(2);
assertEquals(7, tracker.getBytesRead());
// and read the rest
out = new byte[3];
reader.readFully(out);
assertEquals("890", new String(out));
assertEquals(10, tracker.getBytesRead());
assertEquals(testData.length, tracker.getBytesRead());
} finally {
in.close();
}
}
use of org.apache.cassandra.io.util.TrackedDataInputPlus in project cassandra by apache.
the class BytesReadTrackerTest method internalTestUnsignedRead.
public void internalTestUnsignedRead(boolean inputStream) throws Exception {
byte[] testData;
ByteArrayOutputStream baos = new ByteArrayOutputStream();
DataOutputStream out = new DataOutputStream(baos);
try {
// byte
out.writeByte(0x1);
// short
out.writeShort(1);
testData = baos.toByteArray();
} finally {
out.close();
}
DataInputPlus.DataInputStreamPlus in = new DataInputPlus.DataInputStreamPlus(new ByteArrayInputStream(testData));
BytesReadTracker tracker = inputStream ? new TrackedInputStream(in) : new TrackedDataInputPlus(in);
DataInputPlus reader = inputStream ? new DataInputPlus.DataInputStreamPlus((TrackedInputStream) tracker) : (DataInputPlus) tracker;
try {
// byte = 1byte
int b = reader.readUnsignedByte();
assertEquals(b, 1);
assertEquals(1, tracker.getBytesRead());
// short = 2bytes
int s = reader.readUnsignedShort();
assertEquals(1, s);
assertEquals(3, tracker.getBytesRead());
assertEquals(testData.length, tracker.getBytesRead());
} finally {
in.close();
}
}
use of org.apache.cassandra.io.util.TrackedDataInputPlus in project cassandra by apache.
the class DTestSerializer method deserialize.
public HintMessage deserialize(DataInputPlus in, int version) throws IOException {
UUID hostId = UUIDSerializer.serializer.deserialize(in, version);
long hintSize = in.readUnsignedVInt();
TrackedDataInputPlus countingIn = new TrackedDataInputPlus(in);
if (hintSize == 0)
return new HintMessage(hostId, TableId.deserialize(countingIn));
try {
return new HintMessage(hostId, Hint.serializer.deserialize(countingIn, version));
} catch (UnknownTableException e) {
in.skipBytes(Ints.checkedCast(hintSize - countingIn.getBytesRead()));
return new HintMessage(hostId, e.id);
}
}
Aggregations