Sample Code

Compressing and Decompressing Files with Stream Compression

Perform compression or the appropriate kind of decompression to a file based on its path extension.



This sample code project uses the Compression framework to encode (compress) and decode (decompress) files by dragging them to the app window. The app decompresses files with an extension that matches a supported compression algorithm (that is, .lz4, .zlib, .lzma, or .lzfse), and compresses all other files. The app writes the encoded or decoded result to the temporary directory returned by the NSTemporaryDirectory function.

This sample implements stream compression, where the app calls the encode or decode functions repeatedly to compress or decompress data from a source buffer to a destination buffer. Between calls, the app removes processed data from the source buffer and loads the new data into the destination buffer:

Graphic illustrating the stream compression and decompression workflow

The advantage of this approach is that by moving the encoding or decoding to a background thread, you’re able to keep your app interactive and update the user with progress of the operation (for example, with an NSProgressIndicator). Stream compression also enables tasks such as:

  • Decoding a compressed stream into a buffer with the ability to grow the buffer and resume decoding if the expanded stream is too large to fit, without repeating any work.

  • Encoding a stream as pieces of it become available, without ever needing to create a buffer large enough to hold all the uncompressed data at one time.

Accept Dropped Files

Register the app’s view, DragDropCompressView, as a destination for file URLs by adding registerForDraggedTypes: to the viewDidMoveToSuperview method:

override func viewDidMoveToSuperview() {

After the user drops a conforming file onto the app, the view calls the performDragOperation: method. Iterate over the dragged items inside performDragOperation, using guard to ensure each item is a URL instance:

sender.enumerateDraggingItems(options: [],
                              for: nil,
                              classes: [ NSURL.self ],
                              searchOptions: [ .urlReadingFileURLsOnly: true ]) { (draggingItem, _, _) in
        let url = draggingItem.item as? URL else {

Select a Compression Algorithm

If speed and compression ratio are important, use COMPRESSION_LZFSE:

let encodeAlgorithm = COMPRESSION_LZFSE

If you require interoperability with non-Apple devices, use COMPRESSION_ZLIB instead. For more information about other compression algorithms, see compression_algorithm.

Distinguish Between Compressed and Uncompressed Files

Use the path extension to infer whether a file is already compressed, or if the file needs compressing. To simplify this process, create a failable initializer in an extension to the Compression framework’s compression_algorithm structure:

init?(name: String) {
    switch name.lowercased() {
    case "lz4":
        self = COMPRESSION_LZ4
    case "zlib":
        self = COMPRESSION_ZLIB
    case "lzma":
        self = COMPRESSION_LZMA
    case "lzfse":
        self = COMPRESSION_LZFSE
        return nil

Use this new initializer to define the algorithm and operation constants:

let algorithm: compression_algorithm
let operation: compression_stream_operation
if let decodeAlgorithm = compression_algorithm(name: url.pathExtension) {
    algorithm = decodeAlgorithm
} else {
    algorithm = self.encodeAlgorithm

Define the Source and Destination File Handles

The sample uses NSFileHandle instances to read from the source file and write to the destination file. Use optional binding to define the required file handles:

    let sourceFileHandle = try? FileHandle(forReadingFrom: url),
    let sourceLength = FileHelper.fileSize(atURL: url),
    let fileName = url.pathComponents.last,
    let fileNameDeletingPathExtension = url.deletingPathExtension().pathComponents.last,
    let destinationFileHandle = FileHandle.makeFileHandle(forWritingToFileNameInTempDirectory:
        operation == COMPRESSION_STREAM_ENCODE
            ? fileName + self.encodeAlgorithm.pathExtension
            : fileNameDeletingPathExtension)

If the optional binding succeeded, the destination file handle points to the source filename with the compression algorithm appended as an extension (for compression), or the source filename with the compression name extension removed (for decopression). For example, the source file MyCompressedFile.PDF.lzfse would have a destination of MyCompressedFile.PDF, and MyRawFile.PDF would have a destination of MyRawFile.PDF.lzfse.

Pass the source and destination file handles, with the operation and algorithm values to the helper function streamingCompression:

Compressor.streamingCompression(operation: operation,
                                sourceFileHandle: sourceFileHandle,
                                destinationFileHandle: destinationFileHandle,
                                algorithm: algorithm) {
                                    self.progress.completedUnitCount = $0

Create a Destination Buffer

The streamingCompression method iterates over the source data, encoding or decoding data in blocks based on the length defined by bufferSize. The method writes the result into the destination buffer, and writes the destination buffer data to the destination file handle. The following declares the buffer size and allocates the destination buffer:

let bufferSize = 32_768
let destinationBufferPointer = UnsafeMutablePointer<UInt8>.allocate(capacity: bufferSize)
defer {

Create a Compression Stream

The compression_stream structure defines the source and destination pointers and sizes. To simplify the instantiation of a compression_stream structure, create an extension that allocates the required memory:

extension compression_stream {
    init() {
        self = UnsafeMutablePointer<compression_stream>.allocate(capacity: 1).pointee

Use the following extension to declare and initialize the compression stream:

var stream = compression_stream()
var status = compression_stream_init(&stream, operation, algorithm)
guard status != COMPRESSION_STATUS_ERROR else {
    fatalError("Unable to initialize the compression stream.")

After you’re finished with the compression_stream structure, it’s important that you free the memory allocated to it with compression_stream_destroy. Use a defer block to ensure the memory is freed even if the streamingCompression method exits early:

defer {

With the compression stream inialized, set up the stream by defining its source and destination sizes and destination pointer:

stream.src_size = 0
stream.dst_ptr = destinationBufferPointer
stream.dst_size = bufferSize

Read the Source File Data Iteratively

Use a repeat-while loop to manage the read-encode/decode-write process. If the stream’s source size is zero, read a block of data from the source file handle and point the stream’s source pointer to that data. If the read data is shorter than the buffer size, you can infer that you’re reading the last block of the source file and set the stream’s status to COMPRESSION_STREAM_FINALIZE:

var sourceData: Data?
repeat {
    var flags = Int32(0)
    // If this iteration has consumed all of the source data,
    // read a new tempData buffer from the input file.
    if stream.src_size == 0 {
        sourceData = sourceFileHandle.readData(ofLength: bufferSize)
        stream.src_size = sourceData!.count
        if sourceData!.count < bufferSize {
            flags = Int32(COMPRESSION_STREAM_FINALIZE.rawValue)

Compress or Decompress the Dragged File

Use compression_stream_process to encode or decode the current block:

if let sourceData = sourceData {
    let count = sourceData.count
    sourceData.withUnsafeBytes {
        let baseAddress = $0.bindMemory(to: UInt8.self).baseAddress!
        stream.src_ptr = baseAddress.advanced(by: count - stream.src_size)
        status = compression_stream_process(&stream, flags)

On return, destinationBufferPointer points to the encoded or decoded data.

Write Encoded or Decoded Data to a Destination File

Check the status returned by compression_stream_process, if it’s either COMPRESSION_STATUS_OK or COMPRESSION_STATUS_END, you can write the destination data to the destination file handler:

switch status {
    // Get the number of bytes put in the destination buffer. This is the difference between
    // stream.dst_size before the call (here bufferSize), and stream.dst_size after the call.
    let count = bufferSize - stream.dst_size
    let outputData = Data(bytesNoCopy: destinationBufferPointer,
                          count: count,
                          deallocator: .none)
    // Write all produced bytes to the output file.
    // Reset the stream to receive the next batch of output.
    stream.dst_ptr = destinationBufferPointer
    stream.dst_size = bufferSize

This read-encode/decode-write loop continues while status equals COMPRESSION_STATUS_OK.

Close the Source and Destination Files

After you’re finished with the source and destination file handles, close them with the closeFile method:


See Also

Multiple-Step Compression


A structure representing a compression stream.


Initializes a compression stream for either compression or decompression.


Performs compression or decompression using an initialized compression stream structure.


Frees any memory allocated by stream initialization function.


A set of values used to represent the status of stream compression.


A set of values used to represent stream compression flags.


A set of values used to represent a stream compression operation.