If you only want to iterate through all the files (using {@link #entries()}, you should * consider {@link ZipInputStream}, which provides stream-like read access to a zip file and * has a lower up-front cost because you don't pay to build an in-memory index. * *
If you want to create a zip file, use {@link ZipOutputStream}. There is no API for updating
* an existing zip file.
*/
public class ZipFile implements Closeable, ZipConstants {
/**
* General Purpose Bit Flags, Bit 0.
* If set, indicates that the file is encrypted.
*/
static final int GPBF_ENCRYPTED_FLAG = 1 << 0;
/**
* General Purpose Bit Flags, Bit 3.
* If this bit is set, the fields crc-32, compressed
* size and uncompressed size are set to zero in the
* local header. The correct values are put in the
* data descriptor immediately following the compressed
* data. (Note: PKZIP version 2.04g for DOS only
* recognizes this bit for method 8 compression, newer
* versions of PKZIP recognize this bit for any
* compression method.)
*/
static final int GPBF_DATA_DESCRIPTOR_FLAG = 1 << 3;
/**
* General Purpose Bit Flags, Bit 11.
* Language encoding flag (EFS). If this bit is set,
* the filename and comment fields for this file
* must be encoded using UTF-8.
*/
static final int GPBF_UTF8_FLAG = 1 << 11;
/**
* Supported General Purpose Bit Flags Mask.
* Bit mask of bits not supported.
* Note: The only bit that we will enforce at this time
* is the encrypted bit. Although other bits are not supported,
* we must not enforce them as this could break some legitimate
* use cases (See http://b/8617715).
*/
static final int GPBF_UNSUPPORTED_MASK = GPBF_ENCRYPTED_FLAG;
/**
* Open zip file for reading.
*/
public static final int OPEN_READ = 1;
/**
* Delete zip file when closed.
*/
public static final int OPEN_DELETE = 4;
private final String filename;
private File fileToDeleteOnClose;
private RandomAccessFile raf;
private final LinkedHashMap UTF-8 is used to decode all comments and entry names in the file.
*
* @throws ZipException if a zip error occurs.
* @throws IOException if an {@code IOException} occurs.
*/
public ZipFile(File file) throws ZipException, IOException {
this(file, OPEN_READ);
}
/**
* Constructs a new {@code ZipFile} allowing read access to the contents of the given file.
*
* UTF-8 is used to decode all comments and entry names in the file.
*
* @throws IOException if an IOException occurs.
*/
public ZipFile(String name) throws IOException {
this(new File(name), OPEN_READ);
}
/**
* Constructs a new {@code ZipFile} allowing access to the given file.
*
* UTF-8 is used to decode all comments and entry names in the file.
*
* The {@code mode} must be either {@code OPEN_READ} or {@code OPEN_READ|OPEN_DELETE}.
* If the {@code OPEN_DELETE} flag is supplied, the file will be deleted at or before the
* time that the {@code ZipFile} is closed (the contents will remain accessible until
* this {@code ZipFile} is closed); it also calls {@code File.deleteOnExit}.
*
* @throws IOException if an {@code IOException} occurs.
*/
public ZipFile(File file, int mode) throws IOException {
filename = file.getPath();
if (mode != OPEN_READ && mode != (OPEN_READ | OPEN_DELETE)) {
throw new IllegalArgumentException("Bad mode: " + mode);
}
if ((mode & OPEN_DELETE) != 0) {
fileToDeleteOnClose = file;
fileToDeleteOnClose.deleteOnExit();
} else {
fileToDeleteOnClose = null;
}
raf = new RandomAccessFile(filename, "r");
// Make sure to close the RandomAccessFile if reading the central directory fails.
boolean mustCloseFile = true;
try {
readCentralDir();
// Read succeeded so do not close the underlying RandomAccessFile.
mustCloseFile = false;
} finally {
if (mustCloseFile) {
IoUtils.closeQuietly(raf);
}
}
guard.open("close");
}
@Override protected void finalize() throws IOException {
try {
if (guard != null) {
guard.warnIfOpen();
}
} finally {
try {
super.finalize();
} catch (Throwable t) {
throw new AssertionError(t);
}
}
}
/**
* Closes this zip file. This method is idempotent. This method may cause I/O if the
* zip file needs to be deleted.
*
* @throws IOException
* if an IOException occurs.
*/
public void close() throws IOException {
guard.close();
RandomAccessFile localRaf = raf;
if (localRaf != null) { // Only close initialized instances
synchronized (localRaf) {
raf = null;
localRaf.close();
}
if (fileToDeleteOnClose != null) {
fileToDeleteOnClose.delete();
fileToDeleteOnClose = null;
}
}
}
private void checkNotClosed() {
if (raf == null) {
throw new IllegalStateException("Zip file closed");
}
}
/**
* Returns an enumeration of the entries. The entries are listed in the
* order in which they appear in the zip file.
*
* If you only need to iterate over the entries in a zip file, and don't
* need random-access entry lookup by name, you should probably use {@link ZipInputStream}
* instead, to avoid paying to construct the in-memory index.
*
* @throws IllegalStateException if this zip file has been closed.
*/
public Enumeration entries() {
checkNotClosed();
final Iterator The central directory can be followed by a variable-length comment
* field, so we have to scan through it backwards. The comment is at
* most 64K, plus we have 18 bytes for the end-of-central-dir stuff
* itself, plus apparently sometimes people throw random junk on the end
* just for the fun of it.
*
* This is all a little wobbly. If the wrong value ends up in the EOCD
* area, we're hosed. This appears to be the way that everybody handles
* it though, so we're in good company if this fails.
*/
private void readCentralDir() throws IOException {
// Scan back, looking for the End Of Central Directory field. If the zip file doesn't
// have an overall comment (unrelated to any per-entry comments), we'll hit the EOCD
// on the first try.
// No need to synchronize raf here -- we only do this when we first open the zip file.
long scanOffset = raf.length() - ENDHDR;
if (scanOffset < 0) {
throw new ZipException("File too short to be a zip file: " + raf.length());
}
raf.seek(0);
final int headerMagic = Integer.reverseBytes(raf.readInt());
if (headerMagic == ENDSIG) {
throw new ZipException("Empty zip archive not supported");
}
if (headerMagic != LOCSIG) {
throw new ZipException("Not a zip archive");
}
long stopOffset = scanOffset - 65536;
if (stopOffset < 0) {
stopOffset = 0;
}
long eocdOffset;
while (true) {
raf.seek(scanOffset);
if (Integer.reverseBytes(raf.readInt()) == ENDSIG) {
eocdOffset = scanOffset;
break;
}
scanOffset--;
if (scanOffset < stopOffset) {
throw new ZipException("End Of Central Directory signature not found");
}
}
final long zip64EocdRecordOffset = Zip64.parseZip64EocdRecordLocator(raf, eocdOffset);
// Seek back past the eocd signature so that we can continue with our search.
// Note that we add 4 bytes to the offset to skip past the signature.
EocdRecord record = parseEocdRecord(raf, eocdOffset + 4, (zip64EocdRecordOffset != -1) /* isZip64 */);
// Read the comment now to avoid an additional seek. We also know the commentLength
// won't change because that information isn't present in the zip64 eocd record.
if (record.commentLength > 0) {
byte[] commentBytes = new byte[record.commentLength];
raf.readFully(commentBytes);
comment = new String(commentBytes, 0, commentBytes.length, StandardCharsets.UTF_8);
}
// We have a zip64 eocd record : use that for getting the information we need.
if (zip64EocdRecordOffset != -1) {
record = Zip64.parseZip64EocdRecord(raf, zip64EocdRecordOffset, record.commentLength);
}
// Seek to the first CDE and read all entries.
// We have to do this now (from the constructor) rather than lazily because the
// public API doesn't allow us to throw IOException except from the constructor
// or from getInputStream.
RAFStream rafStream = new RAFStream(raf, record.centralDirOffset);
BufferedInputStream bufferedStream = new BufferedInputStream(rafStream, 4096);
byte[] hdrBuf = new byte[CENHDR]; // Reuse the same buffer for each entry.
for (long i = 0; i < record.numEntries; ++i) {
ZipEntry newEntry = new ZipEntry(hdrBuf, bufferedStream, StandardCharsets.UTF_8,
(zip64EocdRecordOffset != -1) /* isZip64 */);
if (newEntry.localHeaderRelOffset >= record.centralDirOffset) {
throw new ZipException("Local file header offset is after central directory");
}
String entryName = newEntry.getName();
if (entries.put(entryName, newEntry) != null) {
throw new ZipException("Duplicate entry name: " + entryName);
}
}
}
private static EocdRecord parseEocdRecord(RandomAccessFile raf, long offset, boolean isZip64) throws IOException {
raf.seek(offset);
// Read the End Of Central Directory. ENDHDR includes the signature bytes,
// which we've already read.
byte[] eocd = new byte[ENDHDR - 4];
raf.readFully(eocd);
BufferIterator it = HeapBufferIterator.iterator(eocd, 0, eocd.length, ByteOrder.LITTLE_ENDIAN);
final long numEntries;
final long centralDirOffset;
if (isZip64) {
numEntries = -1;
centralDirOffset = -1;
// If we have a zip64 end of central directory record, we skip through the regular
// end of central directory record and use the information from the zip64 eocd record.
// We're still forced to read the comment length (below) since it isn't present in the
// zip64 eocd record.
it.skip(16);
} else {
// If we don't have a zip64 eocd record, we read values from the "regular"
// eocd record.
int diskNumber = it.readShort() & 0xffff;
int diskWithCentralDir = it.readShort() & 0xffff;
numEntries = it.readShort() & 0xffff;
int totalNumEntries = it.readShort() & 0xffff;
it.skip(4); // Ignore centralDirSize.
centralDirOffset = ((long) it.readInt()) & 0xffffffffL;
if (numEntries != totalNumEntries || diskNumber != 0 || diskWithCentralDir != 0) {
throw new ZipException("Spanned archives not supported");
}
}
final int commentLength = it.readShort() & 0xffff;
return new EocdRecord(numEntries, centralDirOffset, commentLength);
}
static void throwZipException(String msg, int magic) throws ZipException {
final String hexString = IntegralToString.intToHexString(magic, true, 8);
throw new ZipException(msg + " signature not found; was " + hexString);
}
/**
* Wrap a stream around a RandomAccessFile. The RandomAccessFile is shared
* among all streams returned by getInputStream(), so we have to synchronize
* access to it. (We can optimize this by adding buffering here to reduce
* collisions.)
*
* We could support mark/reset, but we don't currently need them.
*
* @hide
*/
public static class RAFStream extends InputStream {
private final RandomAccessFile sharedRaf;
private long endOffset;
private long offset;
public RAFStream(RandomAccessFile raf, long initialOffset, long endOffset) {
sharedRaf = raf;
offset = initialOffset;
this.endOffset = endOffset;
}
public RAFStream(RandomAccessFile raf, long initialOffset) throws IOException {
this(raf, initialOffset, raf.length());
}
@Override public int available() throws IOException {
return (offset < endOffset ? 1 : 0);
}
@Override public int read() throws IOException {
return Streams.readSingleByte(this);
}
@Override public int read(byte[] buffer, int byteOffset, int byteCount) throws IOException {
synchronized (sharedRaf) {
final long length = endOffset - offset;
if (byteCount > length) {
byteCount = (int) length;
}
sharedRaf.seek(offset);
int count = sharedRaf.read(buffer, byteOffset, byteCount);
if (count > 0) {
offset += count;
return count;
} else {
return -1;
}
}
}
@Override public long skip(long byteCount) throws IOException {
if (byteCount > endOffset - offset) {
byteCount = endOffset - offset;
}
offset += byteCount;
return byteCount;
}
public int fill(Inflater inflater, int nativeEndBufSize) throws IOException {
synchronized (sharedRaf) {
int len = Math.min((int) (endOffset - offset), nativeEndBufSize);
int cnt = inflater.setFileInput(sharedRaf.getFD(), offset, nativeEndBufSize);
// setFileInput read from the file, so we need to get the OS and RAFStream back
// in sync...
skip(cnt);
return len;
}
}
}
/** @hide */
public static class ZipInflaterInputStream extends InflaterInputStream {
private final ZipEntry entry;
private long bytesRead = 0;
public ZipInflaterInputStream(InputStream is, Inflater inf, int bsize, ZipEntry entry) {
super(is, inf, bsize);
this.entry = entry;
}
@Override public int read(byte[] buffer, int byteOffset, int byteCount) throws IOException {
final int i;
try {
i = super.read(buffer, byteOffset, byteCount);
} catch (IOException e) {
throw new IOException("Error reading data for " + entry.getName() + " near offset "
+ bytesRead, e);
}
if (i == -1) {
if (entry.size != bytesRead) {
throw new IOException("Size mismatch on inflated file: " + bytesRead + " vs "
+ entry.size);
}
} else {
bytesRead += i;
}
return i;
}
@Override public int available() throws IOException {
if (closed) {
// Our superclass will throw an exception, but there's a jtreg test that
// explicitly checks that the InputStream returned from ZipFile.getInputStream
// returns 0 even when closed.
return 0;
}
return super.available() == 0 ? 0 : (int) (entry.getSize() - bytesRead);
}
}
}