use of org.apache.nifi.stream.io.BufferedInputStream in project nifi by apache.
the class ExtractTNEFAttachments method onTrigger.
@Override
public void onTrigger(final ProcessContext context, final ProcessSession session) {
final ComponentLog logger = getLogger();
final FlowFile originalFlowFile = session.get();
if (originalFlowFile == null) {
return;
}
final List<FlowFile> attachmentsList = new ArrayList<>();
final List<FlowFile> invalidFlowFilesList = new ArrayList<>();
final List<FlowFile> originalFlowFilesList = new ArrayList<>();
session.read(originalFlowFile, new InputStreamCallback() {
@Override
public void process(final InputStream rawIn) throws IOException {
try (final InputStream in = new BufferedInputStream(rawIn)) {
Properties props = new Properties();
HMEFMessage hmefMessage = null;
// This will trigger an exception in case content is not a TNEF.
hmefMessage = new HMEFMessage(in);
// Add otiginal flowfile (may revert later on in case of errors) //
originalFlowFilesList.add(originalFlowFile);
if (hmefMessage != null) {
// Attachments isn empty, proceeding.
if (!hmefMessage.getAttachments().isEmpty()) {
final String originalFlowFileName = originalFlowFile.getAttribute(CoreAttributes.FILENAME.key());
try {
for (final Attachment attachment : hmefMessage.getAttachments()) {
FlowFile split = session.create(originalFlowFile);
final Map<String, String> attributes = new HashMap<>();
if (StringUtils.isNotBlank(attachment.getLongFilename())) {
attributes.put(CoreAttributes.FILENAME.key(), attachment.getFilename());
}
String parentUuid = originalFlowFile.getAttribute(CoreAttributes.UUID.key());
attributes.put(ATTACHMENT_ORIGINAL_UUID, parentUuid);
attributes.put(ATTACHMENT_ORIGINAL_FILENAME, originalFlowFileName);
// TODO: Extract Mime Type (HMEF doesn't seem to be able to get this info.
split = session.append(split, new OutputStreamCallback() {
@Override
public void process(OutputStream out) throws IOException {
out.write(attachment.getContents());
}
});
split = session.putAllAttributes(split, attributes);
attachmentsList.add(split);
}
} catch (FlowFileHandlingException e) {
// Something went wrong
// Removing splits that may have been created
session.remove(attachmentsList);
// Removing the original flow from its list
originalFlowFilesList.remove(originalFlowFile);
logger.error("Flowfile {} triggered error {} while processing message removing generated FlowFiles from sessions", new Object[] { originalFlowFile, e });
invalidFlowFilesList.add(originalFlowFile);
}
}
}
} catch (Exception e) {
// Another error hit...
// Removing the original flow from its list
originalFlowFilesList.remove(originalFlowFile);
logger.error("Could not parse the flowfile {} as an email, treating as failure", new Object[] { originalFlowFile, e });
// Message is invalid or triggered an error during parsing
invalidFlowFilesList.add(originalFlowFile);
}
}
});
session.transfer(attachmentsList, REL_ATTACHMENTS);
// As per above code, originalFlowfile may be routed to invalid or
// original depending on RFC2822 compliance.
session.transfer(invalidFlowFilesList, REL_FAILURE);
session.transfer(originalFlowFilesList, REL_ORIGINAL);
// check if attachments have been extracted
if (attachmentsList.size() != 0) {
if (attachmentsList.size() > 10) {
// If more than 10, summarise log
logger.info("Split {} into {} files", new Object[] { originalFlowFile, attachmentsList.size() });
} else {
// Otherwise be more verbose and list each individual split
logger.info("Split {} into {} files: {}", new Object[] { originalFlowFile, attachmentsList.size(), attachmentsList });
}
}
}
use of org.apache.nifi.stream.io.BufferedInputStream in project nifi by apache.
the class FlowFileEvent method getBody.
@Override
public byte[] getBody() {
synchronized (bodyLock) {
if (!bodyLoaded) {
if (flowFile.getSize() > Integer.MAX_VALUE) {
throw new RuntimeException("Can't get body of Event because the backing FlowFile is too large (" + flowFile.getSize() + " bytes)");
}
final ByteArrayOutputStream baos = new ByteArrayOutputStream((int) flowFile.getSize());
session.read(flowFile, new InputStreamCallback() {
@Override
public void process(InputStream in) throws IOException {
try (BufferedInputStream input = new BufferedInputStream(in)) {
StreamUtils.copy(input, baos);
}
baos.close();
}
});
body = baos.toByteArray();
bodyLoaded = true;
}
}
return body;
}
use of org.apache.nifi.stream.io.BufferedInputStream in project nifi by apache.
the class AbstractJsonPathProcessor method validateAndEstablishJsonContext.
static DocumentContext validateAndEstablishJsonContext(ProcessSession processSession, FlowFile flowFile) {
// Parse the document once into an associated context to support multiple path evaluations if specified
final AtomicReference<DocumentContext> contextHolder = new AtomicReference<>(null);
processSession.read(flowFile, new InputStreamCallback() {
@Override
public void process(InputStream in) throws IOException {
try (BufferedInputStream bufferedInputStream = new BufferedInputStream(in)) {
DocumentContext ctx = JsonPath.using(STRICT_PROVIDER_CONFIGURATION).parse(bufferedInputStream);
contextHolder.set(ctx);
}
}
});
return contextHolder.get();
}
use of org.apache.nifi.stream.io.BufferedInputStream in project nifi by apache.
the class CompressContent method onTrigger.
@Override
public void onTrigger(final ProcessContext context, final ProcessSession session) {
FlowFile flowFile = session.get();
if (flowFile == null) {
return;
}
final ComponentLog logger = getLogger();
final long sizeBeforeCompression = flowFile.getSize();
final String compressionMode = context.getProperty(MODE).getValue();
String compressionFormatValue = context.getProperty(COMPRESSION_FORMAT).getValue();
if (compressionFormatValue.equals(COMPRESSION_FORMAT_ATTRIBUTE)) {
final String mimeType = flowFile.getAttribute(CoreAttributes.MIME_TYPE.key());
if (mimeType == null) {
logger.error("No {} attribute exists for {}; routing to failure", new Object[] { CoreAttributes.MIME_TYPE.key(), flowFile });
session.transfer(flowFile, REL_FAILURE);
return;
}
compressionFormatValue = compressionFormatMimeTypeMap.get(mimeType);
if (compressionFormatValue == null) {
logger.info("Mime Type of {} is '{}', which does not indicate a supported Compression Format; routing to success without decompressing", new Object[] { flowFile, mimeType });
session.transfer(flowFile, REL_SUCCESS);
return;
}
}
final String compressionFormat = compressionFormatValue;
final AtomicReference<String> mimeTypeRef = new AtomicReference<>(null);
final StopWatch stopWatch = new StopWatch(true);
final String fileExtension;
switch(compressionFormat.toLowerCase()) {
case COMPRESSION_FORMAT_GZIP:
fileExtension = ".gz";
break;
case COMPRESSION_FORMAT_LZMA:
fileExtension = ".lzma";
break;
case COMPRESSION_FORMAT_XZ_LZMA2:
fileExtension = ".xz";
break;
case COMPRESSION_FORMAT_BZIP2:
fileExtension = ".bz2";
break;
case COMPRESSION_FORMAT_SNAPPY:
fileExtension = ".snappy";
break;
case COMPRESSION_FORMAT_SNAPPY_FRAMED:
fileExtension = ".sz";
break;
default:
fileExtension = "";
break;
}
try {
flowFile = session.write(flowFile, new StreamCallback() {
@Override
public void process(final InputStream rawIn, final OutputStream rawOut) throws IOException {
final OutputStream compressionOut;
final InputStream compressionIn;
final OutputStream bufferedOut = new BufferedOutputStream(rawOut, 65536);
final InputStream bufferedIn = new BufferedInputStream(rawIn, 65536);
try {
if (MODE_COMPRESS.equalsIgnoreCase(compressionMode)) {
compressionIn = bufferedIn;
switch(compressionFormat.toLowerCase()) {
case COMPRESSION_FORMAT_GZIP:
final int compressionLevel = context.getProperty(COMPRESSION_LEVEL).asInteger();
compressionOut = new GZIPOutputStream(bufferedOut, compressionLevel);
mimeTypeRef.set("application/gzip");
break;
case COMPRESSION_FORMAT_LZMA:
compressionOut = new LzmaOutputStream.Builder(bufferedOut).build();
mimeTypeRef.set("application/x-lzma");
break;
case COMPRESSION_FORMAT_XZ_LZMA2:
compressionOut = new XZOutputStream(bufferedOut, new LZMA2Options());
mimeTypeRef.set("application/x-xz");
break;
case COMPRESSION_FORMAT_SNAPPY:
compressionOut = new SnappyOutputStream(bufferedOut);
mimeTypeRef.set("application/x-snappy");
break;
case COMPRESSION_FORMAT_SNAPPY_FRAMED:
compressionOut = new SnappyFramedOutputStream(bufferedOut);
mimeTypeRef.set("application/x-snappy-framed");
break;
case COMPRESSION_FORMAT_BZIP2:
default:
mimeTypeRef.set("application/x-bzip2");
compressionOut = new CompressorStreamFactory().createCompressorOutputStream(compressionFormat.toLowerCase(), bufferedOut);
break;
}
} else {
compressionOut = bufferedOut;
switch(compressionFormat.toLowerCase()) {
case COMPRESSION_FORMAT_LZMA:
compressionIn = new LzmaInputStream(bufferedIn, new Decoder());
break;
case COMPRESSION_FORMAT_XZ_LZMA2:
compressionIn = new XZInputStream(bufferedIn);
break;
case COMPRESSION_FORMAT_BZIP2:
// need this two-arg constructor to support concatenated streams
compressionIn = new BZip2CompressorInputStream(bufferedIn, true);
break;
case COMPRESSION_FORMAT_GZIP:
compressionIn = new GzipCompressorInputStream(bufferedIn, true);
break;
case COMPRESSION_FORMAT_SNAPPY:
compressionIn = new SnappyInputStream(bufferedIn);
break;
case COMPRESSION_FORMAT_SNAPPY_FRAMED:
compressionIn = new SnappyFramedInputStream(bufferedIn);
break;
default:
compressionIn = new CompressorStreamFactory().createCompressorInputStream(compressionFormat.toLowerCase(), bufferedIn);
}
}
} catch (final Exception e) {
closeQuietly(bufferedOut);
throw new IOException(e);
}
try (final InputStream in = compressionIn;
final OutputStream out = compressionOut) {
final byte[] buffer = new byte[8192];
int len;
while ((len = in.read(buffer)) > 0) {
out.write(buffer, 0, len);
}
out.flush();
}
}
});
stopWatch.stop();
final long sizeAfterCompression = flowFile.getSize();
if (MODE_DECOMPRESS.equalsIgnoreCase(compressionMode)) {
flowFile = session.removeAttribute(flowFile, CoreAttributes.MIME_TYPE.key());
if (context.getProperty(UPDATE_FILENAME).asBoolean()) {
final String filename = flowFile.getAttribute(CoreAttributes.FILENAME.key());
if (filename.toLowerCase().endsWith(fileExtension)) {
flowFile = session.putAttribute(flowFile, CoreAttributes.FILENAME.key(), filename.substring(0, filename.length() - fileExtension.length()));
}
}
} else {
flowFile = session.putAttribute(flowFile, CoreAttributes.MIME_TYPE.key(), mimeTypeRef.get());
if (context.getProperty(UPDATE_FILENAME).asBoolean()) {
final String filename = flowFile.getAttribute(CoreAttributes.FILENAME.key());
flowFile = session.putAttribute(flowFile, CoreAttributes.FILENAME.key(), filename + fileExtension);
}
}
logger.info("Successfully {}ed {} using {} compression format; size changed from {} to {} bytes", new Object[] { compressionMode.toLowerCase(), flowFile, compressionFormat, sizeBeforeCompression, sizeAfterCompression });
session.getProvenanceReporter().modifyContent(flowFile, stopWatch.getDuration(TimeUnit.MILLISECONDS));
session.transfer(flowFile, REL_SUCCESS);
} catch (final ProcessException e) {
logger.error("Unable to {} {} using {} compression format due to {}; routing to failure", new Object[] { compressionMode.toLowerCase(), flowFile, compressionFormat, e });
session.transfer(flowFile, REL_FAILURE);
}
}
use of org.apache.nifi.stream.io.BufferedInputStream in project nifi by apache.
the class EvaluateXPath method onTrigger.
@Override
@SuppressWarnings("unchecked")
public void onTrigger(final ProcessContext context, final ProcessSession session) {
final List<FlowFile> flowFiles = session.get(50);
if (flowFiles.isEmpty()) {
return;
}
final ComponentLog logger = getLogger();
final XMLReader xmlReader;
try {
xmlReader = XMLReaderFactory.createXMLReader();
} catch (SAXException e) {
logger.error("Error while constructing XMLReader {}", new Object[] { e });
throw new ProcessException(e.getMessage());
}
if (!context.getProperty(VALIDATE_DTD).asBoolean()) {
xmlReader.setEntityResolver(new EntityResolver() {
@Override
public InputSource resolveEntity(String publicId, String systemId) throws SAXException, IOException {
return new InputSource(new StringReader(""));
}
});
}
final XPathFactory factory = factoryRef.get();
final XPathEvaluator xpathEvaluator = (XPathEvaluator) factory.newXPath();
final Map<String, XPathExpression> attributeToXPathMap = new HashMap<>();
for (final Map.Entry<PropertyDescriptor, String> entry : context.getProperties().entrySet()) {
if (!entry.getKey().isDynamic()) {
continue;
}
final XPathExpression xpathExpression;
try {
xpathExpression = xpathEvaluator.compile(entry.getValue());
attributeToXPathMap.put(entry.getKey().getName(), xpathExpression);
} catch (XPathExpressionException e) {
// should not happen because we've already validated the XPath (in XPathValidator)
throw new ProcessException(e);
}
}
final XPathExpression slashExpression;
try {
slashExpression = xpathEvaluator.compile("/");
} catch (XPathExpressionException e) {
logger.error("unable to compile XPath expression due to {}", new Object[] { e });
session.transfer(flowFiles, REL_FAILURE);
return;
}
final String destination = context.getProperty(DESTINATION).getValue();
final QName returnType;
switch(context.getProperty(RETURN_TYPE).getValue()) {
case RETURN_TYPE_AUTO:
if (DESTINATION_ATTRIBUTE.equals(destination)) {
returnType = STRING;
} else if (DESTINATION_CONTENT.equals(destination)) {
returnType = NODESET;
} else {
throw new IllegalStateException("The only possible destinations should be CONTENT or ATTRIBUTE...");
}
break;
case RETURN_TYPE_NODESET:
returnType = NODESET;
break;
case RETURN_TYPE_STRING:
returnType = STRING;
break;
default:
throw new IllegalStateException("There are no other return types...");
}
flowFileLoop: for (FlowFile flowFile : flowFiles) {
final AtomicReference<Throwable> error = new AtomicReference<>(null);
final AtomicReference<Source> sourceRef = new AtomicReference<>(null);
session.read(flowFile, new InputStreamCallback() {
@Override
public void process(final InputStream rawIn) throws IOException {
try (final InputStream in = new BufferedInputStream(rawIn)) {
final List<Source> rootList = (List<Source>) slashExpression.evaluate(new SAXSource(xmlReader, new InputSource(in)), NODESET);
sourceRef.set(rootList.get(0));
} catch (final Exception e) {
error.set(e);
}
}
});
if (error.get() != null) {
logger.error("unable to evaluate XPath against {} due to {}; routing to 'failure'", new Object[] { flowFile, error.get() });
session.transfer(flowFile, REL_FAILURE);
continue;
}
final Map<String, String> xpathResults = new HashMap<>();
for (final Map.Entry<String, XPathExpression> entry : attributeToXPathMap.entrySet()) {
Object result = null;
try {
result = entry.getValue().evaluate(sourceRef.get(), returnType);
if (result == null) {
continue;
}
} catch (final XPathExpressionException e) {
logger.error("failed to evaluate XPath for {} for Property {} due to {}; routing to failure", new Object[] { flowFile, entry.getKey(), e });
session.transfer(flowFile, REL_FAILURE);
continue flowFileLoop;
}
if (returnType == NODESET) {
List<Source> nodeList = (List<Source>) result;
if (nodeList.isEmpty()) {
logger.info("Routing {} to 'unmatched'", new Object[] { flowFile });
session.transfer(flowFile, REL_NO_MATCH);
continue flowFileLoop;
} else if (nodeList.size() > 1) {
logger.error("Routing {} to 'failure' because the XPath evaluated to {} XML nodes", new Object[] { flowFile, nodeList.size() });
session.transfer(flowFile, REL_FAILURE);
continue flowFileLoop;
}
final Source sourceNode = nodeList.get(0);
if (DESTINATION_ATTRIBUTE.equals(destination)) {
try {
ByteArrayOutputStream baos = new ByteArrayOutputStream();
doTransform(sourceNode, baos);
xpathResults.put(entry.getKey(), baos.toString("UTF-8"));
} catch (UnsupportedEncodingException e) {
throw new ProcessException(e);
} catch (TransformerException e) {
error.set(e);
}
} else if (DESTINATION_CONTENT.equals(destination)) {
flowFile = session.write(flowFile, new OutputStreamCallback() {
@Override
public void process(final OutputStream rawOut) throws IOException {
try (final OutputStream out = new BufferedOutputStream(rawOut)) {
doTransform(sourceNode, out);
} catch (TransformerException e) {
error.set(e);
}
}
});
}
} else if (returnType == STRING) {
final String resultString = (String) result;
if (DESTINATION_ATTRIBUTE.equals(destination)) {
xpathResults.put(entry.getKey(), resultString);
} else if (DESTINATION_CONTENT.equals(destination)) {
flowFile = session.write(flowFile, new OutputStreamCallback() {
@Override
public void process(final OutputStream rawOut) throws IOException {
try (final OutputStream out = new BufferedOutputStream(rawOut)) {
out.write(resultString.getBytes("UTF-8"));
}
}
});
}
}
}
if (error.get() == null) {
if (DESTINATION_ATTRIBUTE.equals(destination)) {
flowFile = session.putAllAttributes(flowFile, xpathResults);
final Relationship destRel = xpathResults.isEmpty() ? REL_NO_MATCH : REL_MATCH;
logger.info("Successfully evaluated XPaths against {} and found {} matches; routing to {}", new Object[] { flowFile, xpathResults.size(), destRel.getName() });
session.transfer(flowFile, destRel);
session.getProvenanceReporter().modifyAttributes(flowFile);
} else if (DESTINATION_CONTENT.equals(destination)) {
logger.info("Successfully updated content for {}; routing to 'matched'", new Object[] { flowFile });
session.transfer(flowFile, REL_MATCH);
session.getProvenanceReporter().modifyContent(flowFile);
}
} else {
logger.error("Failed to write XPath result for {} due to {}; routing original to 'failure'", new Object[] { flowFile, error.get() });
session.transfer(flowFile, REL_FAILURE);
}
}
}
Aggregations