/**
* get the first match from the many delimiters give as input
*
* @param source
* @param delimiters
* @return String
*/
public static String getFirstMatch(String source, String[] delimiters) {
Vector<Delimiter> indices = new Vector<Delimiter>();
for (int i = 0; i < delimiters.length; i++) {
// we added the indices of all matches to the vector
indices.add(new Delimiter(i, source.indexOf(delimiters[i]), delimiters[i]));
}
Delimiter delimiter = getFirstIndex(indices);
if (delimiter.getValue() == null) {
return null;
}
return source.substring(0, delimiter.getFoundAt() + 1);
}
@Override
@SuppressWarnings("unchecked")
protected void setup(Reducer.Context context) throws IOException, InterruptedException {
this.context = context;
String settingsStr = context.getConfiguration().get(ParameterProcessing.SETTINGS_STR);
Settings settings = Settings.loadFromString(settingsStr);
Settings.setSettings(settings);
String projectStr = context.getConfiguration().get(ParameterProcessing.PROJECT);
Project project = Project.loadFromString(projectStr);
if (project.isEnvHadoop()) {
String metadataFileContents =
context.getConfiguration().get(ParameterProcessing.METADATA_FILE);
new File(ColumnMetadata.metadataNamesFile).getParentFile().mkdirs();
Files.write(metadataFileContents.getBytes(), new File(ColumnMetadata.metadataNamesFile));
}
columnMetadata = new ColumnMetadata();
String fileSeparatorStr = project.getFieldSeparator();
char fieldSeparatorChar = Delimiter.getDelim(fileSeparatorStr);
columnMetadata.setFieldSeparator(String.valueOf(fieldSeparatorChar));
columnMetadata.setAllMetadata(project.getMetadataCollect());
// write standard metadata fields
context.write(null, new Text(columnMetadata.delimiterSeparatedHeaders()));
zipFileWriter.setup();
zipFileWriter.openZipForWriting();
luceneIndex = new LuceneIndex(settings.getLuceneIndexDir(), project.getProjectCode(), null);
luceneIndex.init();
}
/**
* Gets the first index of a match from the many delimiters given as input Takes many delimiters
* returns the delimiter that occured first
*
* @param collection
* @return instance of Delimiter class
*/
private static Delimiter getFirstIndex(Collection<Delimiter> collection) {
Delimiter delimiter = new Delimiter(-1, 999, null);
Iterator it = collection.iterator();
// comparing the matches to see which match occured first
while (it.hasNext()) {
Delimiter temp = (Delimiter) it.next();
int indexOf = temp.foundAt;
if (indexOf < delimiter.getFoundAt() && indexOf > -1) {
delimiter = temp;
}
}
if (delimiter == null) {
return null;
}
return delimiter;
}
public void run() {
boolean lookingForSduLength = true;
int length = 0;
byte[] sduLengthCandidate = new byte[MAX_SDU_LENGTH_SIZE];
int sduLengthPosition = 0;
byte nextByte = 0;
int value = 0;
byte[] pdu = null;
int bytesRead = 0;
int bytesToRead = 0;
int offset = 0;
BufferedInputStream bufInStream = null;
// byte[] nextBytes = new byte[MAX_SDU_SIZE];
// int numberOfBytesRead = 0;
log.debug(
"Reading socket from remote interface: "
+ socket.getInetAddress().getHostAddress()
+ "\n"
+ "Local port_id: "
+ socket.getLocalPort()
+ "\n"
+ "Remote port_id: "
+ socket.getPort());
try {
bufInStream = new BufferedInputStream(socket.getInputStream());
} catch (Exception ex) {
log.error("Socket reader stopping. " + ex.getMessage());
return;
}
while (!end) {
// Delimit the byte array that contains a serialized CDAP message
try {
if (lookingForSduLength) {
value = bufInStream.read();
if (value == -1) {
break;
}
nextByte = (byte) value;
sduLengthCandidate[sduLengthPosition] = nextByte;
length = delimiter.readVarint32(sduLengthCandidate, sduLengthPosition + 1);
if (length == -2) {
sduLengthPosition++;
} else if (length == -1) {
log.error("Detected SDU with SDU length encoded in too many bytes");
sduLengthPosition = 0;
} else if (length == 0) {
log.debug("Detected 0 length SDU");
this.process0LengthSDU();
} else {
sduLengthPosition = 0;
if (length > 0) {
lookingForSduLength = false;
}
}
} else {
if (pdu == null) {
pdu = new byte[length];
bytesToRead = length;
offset = 0;
}
bytesRead = bufInStream.read(pdu, offset, bytesToRead);
if (bytesRead == -1) {
break;
}
offset = offset + bytesRead;
bytesToRead = bytesToRead - bytesRead;
if (bytesToRead == 0) {
try {
processPDU(pdu);
} catch (Exception ex) {
log.error(
"Problems when processing PDU: "
+ ex.getMessage()
+ "\n.PDU: "
+ printBytes(pdu));
}
lookingForSduLength = true;
pdu = null;
}
}
} catch (Exception ex) {
end = true;
}
}
try {
log.debug("EOF detected, closing the socket");
socket.close();
log.debug("Socket closed");
} catch (IOException ex) {
log.error("Exception closing the socket: " + ex.getMessage());
}
log.info("The remote endpoint of socket " + socket.getPort() + " has disconnected");
socketDisconnected();
}