SevenZipExtractor(
IngestJobContext context,
FileTypeDetector fileTypeDetector,
String moduleDirRelative,
String moduleDirAbsolute)
throws IngestModuleException {
if (!SevenZip.isInitializedSuccessfully()
&& (SevenZip.getLastInitializationException() == null)) {
try {
SevenZip.initSevenZipFromPlatformJAR();
String platform = SevenZip.getUsedPlatform();
logger.log(
Level.INFO,
"7-Zip-JBinding library was initialized on supported platform: {0}",
platform); // NON-NLS
} catch (SevenZipNativeInitializationException e) {
logger.log(Level.SEVERE, "Error initializing 7-Zip-JBinding library", e); // NON-NLS
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.init.errInitModule.msg",
EmbeddedFileExtractorModuleFactory.getModuleName());
String details =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.init.errCantInitLib",
e.getMessage());
services.postMessage(
IngestMessage.createErrorMessage(
EmbeddedFileExtractorModuleFactory.getModuleName(), msg, details));
throw new IngestModuleException(e.getMessage());
}
}
this.context = context;
this.fileTypeDetector = fileTypeDetector;
this.moduleDirRelative = moduleDirRelative;
this.moduleDirAbsolute = moduleDirAbsolute;
this.archiveDepthCountTree = new ArchiveDepthCountTree();
}
/**
* Unpack the file to local folder and return a list of derived files
*
* @param pipelineContext current ingest context
* @param archiveFile file to unpack
* @return list of unpacked derived files
*/
void unpack(AbstractFile archiveFile) {
String archiveFilePath;
try {
archiveFilePath = archiveFile.getUniquePath();
} catch (TskCoreException ex) {
archiveFilePath = archiveFile.getParentPath() + archiveFile.getName();
}
// check if already has derived files, skip
try {
if (archiveFile.hasChildren()) {
// check if local unpacked dir exists
if (new File(EmbeddedFileExtractorIngestModule.getUniqueName(archiveFile)).exists()) {
logger.log(
Level.INFO,
"File already has been processed as it has children and local unpacked file, skipping: {0}",
archiveFilePath); // NON-NLS
return;
}
}
} catch (TskCoreException e) {
logger.log(
Level.INFO,
"Error checking if file already has been processed, skipping: {0}",
archiveFilePath); // NON-NLS
return;
}
List<AbstractFile> unpackedFiles = Collections.<AbstractFile>emptyList();
// recursion depth check for zip bomb
final long archiveId = archiveFile.getId();
SevenZipExtractor.ArchiveDepthCountTree.Archive parentAr =
archiveDepthCountTree.findArchive(archiveId);
if (parentAr == null) {
parentAr = archiveDepthCountTree.addArchive(null, archiveId);
} else if (parentAr.getDepth() == MAX_DEPTH) {
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.warnMsg.zipBomb",
archiveFile.getName());
String details =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.warnDetails.zipBomb",
parentAr.getDepth(),
archiveFilePath);
// MessageNotifyUtil.Notify.error(msg, details);
services.postMessage(
IngestMessage.createWarningMessage(
EmbeddedFileExtractorModuleFactory.getModuleName(), msg, details));
return;
}
boolean hasEncrypted = false;
boolean fullEncryption = true;
ISevenZipInArchive inArchive = null;
SevenZipContentReadStream stream = null;
final ProgressHandle progress =
ProgressHandleFactory.createHandle(
NbBundle.getMessage(
this.getClass(), "EmbeddedFileExtractorIngestModule.ArchiveExtractor.moduleName"));
int processedItems = 0;
boolean progressStarted = false;
try {
stream = new SevenZipContentReadStream(new ReadContentInputStream(archiveFile));
// for RAR files we need to open them explicitly as RAR. Otherwise, if there is a ZIP archive
// inside RAR archive
// it will be opened incorrectly when using 7zip's built-in auto-detect functionality.
// All other archive formats are still opened using 7zip built-in auto-detect functionality.
ArchiveFormat options = get7ZipOptions(archiveFile);
inArchive = SevenZip.openInArchive(options, stream);
int numItems = inArchive.getNumberOfItems();
logger.log(
Level.INFO,
"Count of items in archive: {0}: {1}",
new Object[] {archiveFilePath, numItems}); // NON-NLS
progress.start(numItems);
progressStarted = true;
final ISimpleInArchive simpleInArchive = inArchive.getSimpleInterface();
// setup the archive local root folder
final String uniqueArchiveFileName =
EmbeddedFileExtractorIngestModule.getUniqueName(archiveFile);
final String localRootAbsPath = getLocalRootAbsPath(uniqueArchiveFileName);
final File localRoot = new File(localRootAbsPath);
if (!localRoot.exists()) {
try {
localRoot.mkdirs();
} catch (SecurityException e) {
logger.log(
Level.SEVERE,
"Error setting up output path for archive root: {0}",
localRootAbsPath); // NON-NLS
// bail
return;
}
}
// initialize tree hierarchy to keep track of unpacked file structure
SevenZipExtractor.UnpackedTree unpackedTree =
new SevenZipExtractor.UnpackedTree(
moduleDirRelative + "/" + uniqueArchiveFileName, archiveFile);
long freeDiskSpace = services.getFreeDiskSpace();
// unpack and process every item in archive
int itemNumber = 0;
for (ISimpleInArchiveItem item : simpleInArchive.getArchiveItems()) {
String pathInArchive = item.getPath();
if (pathInArchive == null || pathInArchive.isEmpty()) {
// some formats (.tar.gz) may not be handled correctly -- file in archive has no name/path
// handle this for .tar.gz and tgz but assuming the child is tar,
// otherwise, unpack using itemNumber as name
// TODO this should really be signature based, not extension based
String archName = archiveFile.getName();
int dotI = archName.lastIndexOf(".");
String useName = null;
if (dotI != -1) {
String base = archName.substring(0, dotI);
String ext = archName.substring(dotI);
switch (ext) {
case ".gz": // NON-NLS
useName = base;
break;
case ".tgz": // NON-NLS
useName = base + ".tar"; // NON-NLS
break;
}
}
if (useName == null) {
pathInArchive = "/" + archName + "/" + Integer.toString(itemNumber);
} else {
pathInArchive = "/" + useName;
}
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.unknownPath.msg",
archiveFilePath,
pathInArchive);
logger.log(Level.WARNING, msg);
}
++itemNumber;
logger.log(Level.INFO, "Extracted item path: {0}", pathInArchive); // NON-NLS
// check if possible zip bomb
if (isZipBombArchiveItemCheck(archiveFile, item)) {
continue; // skip the item
}
// find this node in the hierarchy, create if needed
SevenZipExtractor.UnpackedTree.UnpackedNode unpackedNode =
unpackedTree.addNode(pathInArchive);
String fileName = unpackedNode.getFileName();
// update progress bar
progress.progress(archiveFile.getName() + ": " + fileName, processedItems);
final boolean isEncrypted = item.isEncrypted();
final boolean isDir = item.isFolder();
if (isEncrypted) {
logger.log(
Level.WARNING, "Skipping encrypted file in archive: {0}", pathInArchive); // NON-NLS
hasEncrypted = true;
continue;
} else {
fullEncryption = false;
}
final Long size = item.getSize();
if (size == null) {
// If the size property cannot be determined, out-of-disk-space
// situations cannot be ascertained.
// Hence skip this file.
logger.log(
Level.WARNING,
"Size cannot be determined. Skipping file in archive: {0}",
pathInArchive); // NON-NLS
continue;
}
// check if unpacking this file will result in out of disk space
// this is additional to zip bomb prevention mechanism
if (freeDiskSpace != IngestMonitor.DISK_FREE_SPACE_UNKNOWN
&& size > 0) { // if known free space and file not empty
long newDiskSpace = freeDiskSpace - size;
if (newDiskSpace < MIN_FREE_DISK_SPACE) {
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.notEnoughDiskSpace.msg",
archiveFilePath,
fileName);
String details =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.notEnoughDiskSpace.details");
// MessageNotifyUtil.Notify.error(msg, details);
services.postMessage(
IngestMessage.createErrorMessage(
EmbeddedFileExtractorModuleFactory.getModuleName(), msg, details));
logger.log(
Level.INFO,
"Skipping archive item due to insufficient disk space: {0}, {1}",
new Object[] {archiveFilePath, fileName}); // NON-NLS
logger.log(
Level.INFO, "Available disk space: {0}", new Object[] {freeDiskSpace}); // NON-NLS
continue; // skip this file
} else {
// update est. disk space during this archive, so we don't need to poll for every file
// extracted
freeDiskSpace = newDiskSpace;
}
}
final String uniqueExtractedName =
uniqueArchiveFileName
+ File.separator
+ (item.getItemIndex() / 1000)
+ File.separator
+ item.getItemIndex()
+ new File(pathInArchive).getName();
// final String localRelPath = unpackDir + File.separator + localFileRelPath;
final String localRelPath = moduleDirRelative + File.separator + uniqueExtractedName;
final String localAbsPath = moduleDirAbsolute + File.separator + uniqueExtractedName;
// create local dirs and empty files before extracted
File localFile = new java.io.File(localAbsPath);
// cannot rely on files in top-bottom order
if (!localFile.exists()) {
try {
if (isDir) {
localFile.mkdirs();
} else {
localFile.getParentFile().mkdirs();
try {
localFile.createNewFile();
} catch (IOException ex) {
logger.log(
Level.SEVERE,
"Error creating extracted file: " + localFile.getAbsolutePath(),
ex); // NON-NLS
}
}
} catch (SecurityException e) {
logger.log(
Level.SEVERE,
"Error setting up output path for unpacked file: {0}",
pathInArchive); // NON-NLS
// TODO consider bail out / msg to the user
}
}
// skip the rest of this loop if we couldn't create the file
if (localFile.exists() == false) {
continue;
}
final Date createTime = item.getCreationTime();
final Date accessTime = item.getLastAccessTime();
final Date writeTime = item.getLastWriteTime();
final long createtime = createTime == null ? 0L : createTime.getTime() / 1000;
final long modtime = writeTime == null ? 0L : writeTime.getTime() / 1000;
final long accesstime = accessTime == null ? 0L : accessTime.getTime() / 1000;
// record derived data in unode, to be traversed later after unpacking the archive
unpackedNode.addDerivedInfo(
size, !isDir, 0L, createtime, accesstime, modtime, localRelPath);
// unpack locally if a file
if (!isDir) {
SevenZipExtractor.UnpackStream unpackStream = null;
try {
unpackStream = new SevenZipExtractor.UnpackStream(localAbsPath);
item.extractSlow(unpackStream);
} catch (Exception e) {
// could be something unexpected with this file, move on
logger.log(
Level.WARNING,
"Could not extract file from archive: " + localAbsPath,
e); // NON-NLS
} finally {
if (unpackStream != null) {
unpackStream.close();
}
}
}
// update units for progress bar
++processedItems;
}
// add them to the DB. We wait until the end so that we have the metadata on all of the
// intermediate nodes since the order is not guaranteed
try {
unpackedTree.addDerivedFilesToCase();
unpackedFiles = unpackedTree.getAllFileObjects();
// check if children are archives, update archive depth tracking
for (AbstractFile unpackedFile : unpackedFiles) {
if (isSevenZipExtractionSupported(unpackedFile)) {
archiveDepthCountTree.addArchive(parentAr, unpackedFile.getId());
}
}
} catch (TskCoreException e) {
logger.log(
Level.SEVERE,
"Error populating complete derived file hierarchy from the unpacked dir structure"); // NON-NLS
// TODO decide if anything to cleanup, for now bailing
}
} catch (SevenZipException ex) {
logger.log(Level.SEVERE, "Error unpacking file: " + archiveFile, ex); // NON-NLS
// inbox message
// print a message if the file is allocated
if (archiveFile.isMetaFlagSet(TskData.TSK_FS_META_FLAG_ENUM.ALLOC)) {
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.errUnpacking.msg",
archiveFile.getName());
String details =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.errUnpacking.details",
archiveFilePath,
ex.getMessage());
services.postMessage(
IngestMessage.createErrorMessage(
EmbeddedFileExtractorModuleFactory.getModuleName(), msg, details));
}
} finally {
if (inArchive != null) {
try {
inArchive.close();
} catch (SevenZipException e) {
logger.log(Level.SEVERE, "Error closing archive: " + archiveFile, e); // NON-NLS
}
}
if (stream != null) {
try {
stream.close();
} catch (IOException ex) {
logger.log(
Level.SEVERE,
"Error closing stream after unpacking archive: " + archiveFile,
ex); // NON-NLS
}
}
// close progress bar
if (progressStarted) {
progress.finish();
}
}
// create artifact and send user message
if (hasEncrypted) {
String encryptionType = fullEncryption ? ENCRYPTION_FULL : ENCRYPTION_FILE_LEVEL;
try {
BlackboardArtifact artifact =
archiveFile.newArtifact(BlackboardArtifact.ARTIFACT_TYPE.TSK_ENCRYPTION_DETECTED);
artifact.addAttribute(
new BlackboardAttribute(
BlackboardAttribute.ATTRIBUTE_TYPE.TSK_NAME.getTypeID(),
EmbeddedFileExtractorModuleFactory.getModuleName(),
encryptionType));
services.fireModuleDataEvent(
new ModuleDataEvent(
EmbeddedFileExtractorModuleFactory.getModuleName(),
BlackboardArtifact.ARTIFACT_TYPE.TSK_ENCRYPTION_DETECTED));
} catch (TskCoreException ex) {
logger.log(
Level.SEVERE,
"Error creating blackboard artifact for encryption detected for file: "
+ archiveFilePath,
ex); // NON-NLS
}
String msg =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.encrFileDetected.msg");
String details =
NbBundle.getMessage(
this.getClass(),
"EmbeddedFileExtractorIngestModule.ArchiveExtractor.unpack.encrFileDetected.details",
archiveFile.getName(),
EmbeddedFileExtractorModuleFactory.getModuleName());
services.postMessage(
IngestMessage.createWarningMessage(
EmbeddedFileExtractorModuleFactory.getModuleName(), msg, details));
}
// adding unpacked extracted derived files to the job after closing relevant resources.
if (!unpackedFiles.isEmpty()) {
// currently sending a single event for all new files
services.fireModuleContentEvent(new ModuleContentEvent(archiveFile));
context.addFilesToJob(unpackedFiles);
}
}