MikeThomsen commented on a change in pull request #4336:
URL: https://github.com/apache/nifi/pull/4336#discussion_r444433577
##########
File path:
nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/UnpackContent.java
##########
@@ -394,6 +412,65 @@ public void process(final OutputStream out) throws
IOException {
}
}
+ private static class SevenZipUnpacker extends Unpacker {
+ public SevenZipUnpacker(Pattern fileFilter) {
+ super(fileFilter);
+ }
+
+ @Override
+ public void unpack(final ProcessSession session, final FlowFile
source, final List<FlowFile> unpacked) {
+ final String fragmentId = UUID.randomUUID().toString();
+ session.read(source, new InputStreamCallback() {
+ @Override
+ public void process(final InputStream in) throws IOException {
+
+ int fragmentCount = 0;
+ byte[] inputData = IOUtils.toByteArray(in);
Review comment:
This could easily cause the JVM to run out of memory. It looks like what
you might need to do is create a temporary `java.io.File` and export the
contents of the flowfile to that because the SevenZFile class doesn't allow you
to natively use an InputStream. That could be dangerous in its own right, but
at least users could mitigate it by resizing the partition if there is
sufficient documentation warning them.
##########
File path:
nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/UnpackContent.java
##########
@@ -394,6 +412,65 @@ public void process(final OutputStream out) throws
IOException {
}
}
+ private static class SevenZipUnpacker extends Unpacker {
+ public SevenZipUnpacker(Pattern fileFilter) {
+ super(fileFilter);
+ }
+
+ @Override
+ public void unpack(final ProcessSession session, final FlowFile
source, final List<FlowFile> unpacked) {
+ final String fragmentId = UUID.randomUUID().toString();
+ session.read(source, new InputStreamCallback() {
+ @Override
+ public void process(final InputStream in) throws IOException {
+
+ int fragmentCount = 0;
+ byte[] inputData = IOUtils.toByteArray(in);
+
+ SeekableInMemoryByteChannel inMemoryByteChannel = new
SeekableInMemoryByteChannel(inputData);
+ SevenZFile sevenZFile = new
SevenZFile(inMemoryByteChannel);
+ SevenZArchiveEntry sevenZArchiveEntry;
+
+ while ((sevenZArchiveEntry = sevenZFile.getNextEntry()) !=
null) {
+ if (sevenZArchiveEntry.isDirectory() ||
!fileMatches(sevenZArchiveEntry)) {
+ continue;
+ }
+
+ final File file = new
File(sevenZArchiveEntry.getName());
+ final String parentDirectory = (file.getParent() ==
null) ? "/" : file.getParent();
+ final Path absPath = file.toPath().toAbsolutePath();
+ final String absPathString =
absPath.getParent().toString() + "/";
+
+ FlowFile unpackedFile = session.create(source);
+ try {
+ final Map<String, String> attributes = new
HashMap<>();
+ attributes.put(CoreAttributes.FILENAME.key(),
file.getName());
+ attributes.put(CoreAttributes.PATH.key(),
parentDirectory);
+ attributes.put(CoreAttributes.ABSOLUTE_PATH.key(),
absPathString);
+ attributes.put(CoreAttributes.MIME_TYPE.key(),
OCTET_STREAM);
+
+ attributes.put(FRAGMENT_ID, fragmentId);
+ attributes.put(FRAGMENT_INDEX,
String.valueOf(++fragmentCount));
+
+ unpackedFile =
session.putAllAttributes(unpackedFile, attributes);
+ byte[] content = new
byte[(int)sevenZArchiveEntry.getSize()];
Review comment:
You should be reading the file progressively into a buffer using a loop
because you could easily blow up the JVM here.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
