public boolean accept(Path file) {
try {
FileSystem fs = file.getFileSystem(conf);
boolean unpack = conf.getBoolean(unpackParamName, true);
if (defaultIgnores.accept(file) && fs.getFileStatus(file).isDir() == false) {
String URI = file.toUri().toString();
// detect whether a file is likely to be an archive
// TODO extend to other known types
if (unpack && URI.toLowerCase().endsWith(".zip")) {
FSDataInputStream fis = null;
try {
fis = fs.open(file);
ArchiveInputStream input =
new ArchiveStreamFactory().createArchiveInputStream(new BufferedInputStream(fis));
ArchiveEntry entry = null;
while ((entry = input.getNextEntry()) != null) {
String name = entry.getName();
long size = entry.getSize();
byte[] content = new byte[(int) size];
input.read(content);
key.set(name);
// fill the values for the content object
value.setUrl(name);
value.setContent(content);
writer.append(key, value);
counter++;
if (reporter != null) {
reporter.incrCounter(Counters.DOC_COUNT, 1);
}
}
} catch (ArchiveException e) {
// TODO Auto-generated catch block
e.printStackTrace();
} finally {
fis.close();
}
} else {
// Hmm, kind of dangerous to do this
byte[] fileBArray = new byte[(int) fs.getFileStatus(file).getLen()];
FSDataInputStream fis = null;
try {
fis = fs.open(file);
fis.readFully(0, fileBArray);
fis.close();
key.set(URI);
// fill the values for the content object
value.setUrl(URI);
value.setContent(fileBArray);
writer.append(key, value);
counter++;
if (reporter != null) {
reporter.incrCounter(Counters.DOC_COUNT, 1);
}
} catch (FileNotFoundException e) {
throw new RuntimeException(e);
} catch (IOException e) {
throw new RuntimeException(e);
}
}
}
// if it is a directory, accept it so we can possibly recurse on
// it,
// otherwise we don't care about actually accepting the file,
// since
// all the work is done in the accept method here.
return fs.getFileStatus(file).isDir();
} catch (IOException e) {
log.error("Exception", e);
}
return false;
}
private void handleTARArchive(
ArchiveStoreContext ctx,
FreenetURI key,
InputStream data,
String element,
ArchiveExtractCallback callback,
MutableBoolean gotElement,
boolean throwAtExit,
ClientContext context)
throws ArchiveFailureException, ArchiveRestartException {
if (logMINOR) Logger.minor(this, "Handling a TAR Archive");
TarArchiveInputStream tarIS = null;
try {
tarIS = new TarArchiveInputStream(data);
// MINOR: Assumes the first entry in the tarball is a directory.
ArchiveEntry entry;
byte[] buf = new byte[32768];
HashSet<String> names = new HashSet<String>();
boolean gotMetadata = false;
outerTAR:
while (true) {
try {
entry = tarIS.getNextEntry();
} catch (IllegalArgumentException e) {
// Annoyingly, it can throw this on some corruptions...
throw new ArchiveFailureException("Error reading archive: " + e.getMessage(), e);
}
if (entry == null) break;
if (entry.isDirectory()) continue;
String name = stripLeadingSlashes(entry.getName());
if (names.contains(name)) {
Logger.error(this, "Duplicate key " + name + " in archive " + key);
continue;
}
long size = entry.getSize();
if (name.equals(".metadata")) gotMetadata = true;
if (size > maxArchivedFileSize && !name.equals(element)) {
addErrorElement(
ctx,
key,
name,
"File too big: "
+ size
+ " greater than current archived file size limit "
+ maxArchivedFileSize,
true);
} else {
// Read the element
long realLen = 0;
Bucket output = tempBucketFactory.makeBucket(size);
OutputStream out = output.getOutputStream();
try {
int readBytes;
while ((readBytes = tarIS.read(buf)) > 0) {
out.write(buf, 0, readBytes);
readBytes += realLen;
if (readBytes > maxArchivedFileSize) {
addErrorElement(
ctx,
key,
name,
"File too big: "
+ maxArchivedFileSize
+ " greater than current archived file size limit "
+ maxArchivedFileSize,
true);
out.close();
out = null;
output.free();
continue outerTAR;
}
}
} finally {
if (out != null) out.close();
}
if (size <= maxArchivedFileSize) {
addStoreElement(ctx, key, name, output, gotElement, element, callback, context);
names.add(name);
trimStoredData();
} else {
// We are here because they asked for this file.
callback.gotBucket(output, context);
gotElement.value = true;
addErrorElement(
ctx,
key,
name,
"File too big: "
+ size
+ " greater than current archived file size limit "
+ maxArchivedFileSize,
true);
}
}
}
// If no metadata, generate some
if (!gotMetadata) {
generateMetadata(ctx, key, names, gotElement, element, callback, context);
trimStoredData();
}
if (throwAtExit) throw new ArchiveRestartException("Archive changed on re-fetch");
if ((!gotElement.value) && element != null) callback.notInArchive(context);
} catch (IOException e) {
throw new ArchiveFailureException("Error reading archive: " + e.getMessage(), e);
} finally {
Closer.close(tarIS);
}
}
public static int unpackFileToFolder(
@Nonnull final Log logger,
@Nullable final String folder,
@Nonnull final File archiveFile,
@Nonnull final File destinationFolder,
final boolean makeAllExecutable)
throws IOException {
final String normalizedName = archiveFile.getName().toLowerCase(Locale.ENGLISH);
final ArchEntryGetter entryGetter;
boolean modeZipFile = false;
final ZipFile theZipFile;
final ArchiveInputStream archInputStream;
if (normalizedName.endsWith(".zip")) {
logger.debug("Detected ZIP archive");
modeZipFile = true;
theZipFile = new ZipFile(archiveFile);
archInputStream = null;
entryGetter =
new ArchEntryGetter() {
private final Enumeration<ZipArchiveEntry> iterator = theZipFile.getEntries();
@Override
@Nullable
public ArchiveEntry getNextEntry() throws IOException {
ArchiveEntry result = null;
if (this.iterator.hasMoreElements()) {
result = this.iterator.nextElement();
}
return result;
}
};
} else {
theZipFile = null;
final InputStream in = new BufferedInputStream(new FileInputStream(archiveFile));
try {
if (normalizedName.endsWith(".tar.gz")) {
logger.debug("Detected TAR.GZ archive");
archInputStream = new TarArchiveInputStream(new GZIPInputStream(in));
entryGetter =
new ArchEntryGetter() {
@Override
@Nullable
public ArchiveEntry getNextEntry() throws IOException {
return ((TarArchiveInputStream) archInputStream).getNextTarEntry();
}
};
} else {
logger.debug("Detected OTHER archive");
archInputStream = ARCHIVE_STREAM_FACTORY.createArchiveInputStream(in);
logger.debug("Created archive stream : " + archInputStream.getClass().getName());
entryGetter =
new ArchEntryGetter() {
@Override
@Nullable
public ArchiveEntry getNextEntry() throws IOException {
return archInputStream.getNextEntry();
}
};
}
} catch (ArchiveException ex) {
IOUtils.closeQuietly(in);
throw new IOException("Can't recognize or read archive file : " + archiveFile, ex);
} catch (CantReadArchiveEntryException ex) {
IOUtils.closeQuietly(in);
throw new IOException("Can't read entry from archive file : " + archiveFile, ex);
}
}
try {
final String normalizedFolder =
folder == null ? null : FilenameUtils.normalize(folder, true) + '/';
int unpackedFilesCounter = 0;
while (true) {
final ArchiveEntry entry = entryGetter.getNextEntry();
if (entry == null) {
break;
}
final String normalizedPath = FilenameUtils.normalize(entry.getName(), true);
logger.debug("Detected archive entry : " + normalizedPath);
if (normalizedFolder == null || normalizedPath.startsWith(normalizedFolder)) {
final File targetFile =
new File(
destinationFolder,
normalizedFolder == null
? normalizedPath
: normalizedPath.substring(normalizedFolder.length()));
if (entry.isDirectory()) {
logger.debug("Folder : " + normalizedPath);
if (!targetFile.exists() && !targetFile.mkdirs()) {
throw new IOException("Can't create folder " + targetFile);
}
} else {
final File parent = targetFile.getParentFile();
if (parent != null && !parent.isDirectory() && !parent.mkdirs()) {
throw new IOException("Can't create folder : " + parent);
}
final FileOutputStream fos = new FileOutputStream(targetFile);
try {
if (modeZipFile) {
logger.debug("Unpacking ZIP entry : " + normalizedPath);
final InputStream zipEntryInStream =
theZipFile.getInputStream((ZipArchiveEntry) entry);
try {
if (IOUtils.copy(zipEntryInStream, fos) != entry.getSize()) {
throw new IOException(
"Can't unpack file, illegal unpacked length : " + entry.getName());
}
} finally {
IOUtils.closeQuietly(zipEntryInStream);
}
} else {
logger.debug("Unpacking archive entry : " + normalizedPath);
if (!archInputStream.canReadEntryData(entry)) {
throw new IOException("Can't read archive entry data : " + normalizedPath);
}
if (IOUtils.copy(archInputStream, fos) != entry.getSize()) {
throw new IOException(
"Can't unpack file, illegal unpacked length : " + entry.getName());
}
}
} finally {
fos.close();
}
if (makeAllExecutable) {
try {
targetFile.setExecutable(true, true);
} catch (SecurityException ex) {
throw new IOException("Can't make file executable : " + targetFile, ex);
}
}
unpackedFilesCounter++;
}
} else {
logger.debug("Archive entry " + normalizedPath + " ignored");
}
}
return unpackedFilesCounter;
} finally {
IOUtils.closeQuietly(theZipFile);
IOUtils.closeQuietly(archInputStream);
}
}
