Skip to content

Embed URL

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
GZip support for PCL HttpClient. Create HttpClient using: HttpClient client = new HttpClient(new HttpGZipClientHandler());
using System.IO;
using System.Net.Http;
using System.Net.Http.Headers;
using System.Threading;
using System.Threading.Tasks;
namespace SharpGIS.Http
{
public class HttpGZipClientHandler : HttpClientHandler
{
protected override async Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, CancellationToken cancellationToken)
{
request.Headers.AcceptEncoding.Add(new StringWithQualityHeaderValue("gzip"));
var response = await base.SendAsync(request, cancellationToken);
if (response.Content.Headers.ContentEncoding.Contains("gzip"))
{
await response.Content.LoadIntoBufferAsync();
response.Content = new HttpGZipContent(await response.Content.ReadAsStreamAsync());
}
return response;
}
public override bool SupportsAutomaticDecompression
{
get
{
return true;
}
}
}
internal sealed class HttpGZipContent : HttpContent
{
private readonly GZipInflateStream m_stream;
public HttpGZipContent(Stream stream)
{
m_stream = new GZipInflateStream(stream);
}
protected override Task SerializeToStreamAsync(System.IO.Stream stream, System.Net.TransportContext context)
{
return m_stream.CopyToAsync(stream);
}
protected override bool TryComputeLength(out long length)
{
length = m_stream.Length;
return true;
}
protected override void Dispose(bool disposing)
{
m_stream.Dispose();
base.Dispose(disposing);
}
}
}
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.IO;
using System.Windows;
using System.Windows.Resources;
namespace SharpGIS.Http
{
internal sealed class GZipInflateStream : Stream
{
private readonly Stream _deflatedStream;
private Stream _inflatedStream;
public GZipInflateStream(System.IO.Stream deflatedStream)
{
_deflatedStream = deflatedStream;
ProcessStream();
}
private void ProcessStream()
{
if ((0x1f != _deflatedStream.ReadByte()) || // ID1
(0x8b != _deflatedStream.ReadByte()) || // ID2
(8 != _deflatedStream.ReadByte())) // CM (8 == deflate)
{
throw new NotSupportedException("Compressed data not in the expected format.");
}
// Read flags
var flg = _deflatedStream.ReadByte(); // FLG
var fhcrc = 0 != (0x2 & flg); // CRC16 present before compressed data
var fextra = 0 != (0x4 & flg); // extra fields present
var fname = 0 != (0x8 & flg); // original file name present
var fcomment = 0 != (0x10 & flg); // file comment present
// Skip unsupported fields
if (_deflatedStream.CanSeek)
_deflatedStream.Seek(6, SeekOrigin.Current);
else
{
_deflatedStream.ReadByte();
_deflatedStream.ReadByte();
_deflatedStream.ReadByte();
_deflatedStream.ReadByte(); // MTIME
_deflatedStream.ReadByte(); // XFL
_deflatedStream.ReadByte(); // OS
}
if (fextra)
{
// Skip XLEN bytes of data
var xlen = _deflatedStream.ReadByte() | (_deflatedStream.ReadByte() << 8);
while (0 < xlen)
{
_deflatedStream.ReadByte();
xlen--;
}
}
if (fname)
{
// Skip 0-terminated file name
while (0 != _deflatedStream.ReadByte())
{
}
}
if (fcomment)
{
// Skip 0-terminated file comment
while (0 != _deflatedStream.ReadByte())
{
}
}
if (fhcrc)
{
_deflatedStream.ReadByte(); _deflatedStream.ReadByte(); // CRC16
}
// Read compressed data
const int zipHeaderSize = 30 + 1; // 30 bytes + 1 character for file name
const int zipFooterSize = 68 + 1; // 68 bytes + 1 character for file name
// Download unknown amount of compressed data efficiently (note: Content-Length header is not always reliable)
var buffers = new List<byte[]>();
var buffer = new byte[4096];
var bytesInBuffer = 0;
var totalBytes = 0;
var bytesRead = 0;
do
{
if (buffer.Length == bytesInBuffer)
{
// Full, allocate another
buffers.Add(buffer);
buffer = new byte[buffer.Length];
bytesInBuffer = 0;
}
Debug.Assert(bytesInBuffer < buffer.Length);
bytesRead = _deflatedStream.Read(buffer, bytesInBuffer, buffer.Length - bytesInBuffer);
bytesInBuffer += bytesRead;
totalBytes += bytesRead;
} while (0 < bytesRead);
buffers.Add(buffer);
// "Trim" crc32 and isize fields off the end
var compressedSize = totalBytes - 4 - 4;
if (compressedSize < 0)
{
throw new NotSupportedException("Compressed data not in the expected format.");
}
// Create contiguous buffer
var compressedBytes = new byte[zipHeaderSize + compressedSize + zipFooterSize];
var offset = zipHeaderSize;
var remainingBytes = totalBytes;
foreach (var b in buffers)
{
var length = Math.Min(b.Length, remainingBytes);
Array.Copy(b, 0, compressedBytes, offset, length);
offset += length;
remainingBytes -= length;
}
Debug.Assert(0 == remainingBytes);
// Read footer from end of compressed bytes (note: footer is within zipFooterSize; will be overwritten below)
Debug.Assert(totalBytes <= compressedSize + zipFooterSize);
offset = zipHeaderSize + compressedSize;
var crc32 = compressedBytes[offset + 0] | (compressedBytes[offset + 1] << 8) | (compressedBytes[offset + 2] << 16) | (compressedBytes[offset + 3] << 24);
var isize = compressedBytes[offset + 4] | (compressedBytes[offset + 5] << 8) | (compressedBytes[offset + 6] << 16) | (compressedBytes[offset + 7] << 24);
if (0 == isize) // HACK to handle compressed 0-byte streams without figuring out what's really going wrong
{
_inflatedStream = new MemoryStream();
return;
}
// Create ZIP file stream
const string fileName = "f"; // MUST be 1 character (offsets below assume this)
Debug.Assert(1 == fileName.Length);
var zipFileMemoryStream = new MemoryStream(compressedBytes);
var writer = new BinaryWriter(zipFileMemoryStream);
// Local file header
writer.Write((uint)0x04034b50); // local file header signature
writer.Write((ushort)20); // version needed to extract (2.0 == compressed using deflate)
writer.Write((ushort)0); // general purpose bit flag
writer.Write((ushort)8); // compression method (8: deflate)
writer.Write((ushort)0); // last mod file time
writer.Write((ushort)0); // last mod file date
writer.Write(crc32); // crc-32
writer.Write(compressedSize); // compressed size
writer.Write(isize); // uncompressed size
writer.Write((ushort)1); // file name length
writer.Write((ushort)0); // extra field length
writer.Write((byte)fileName[0]); // file name
// File data (already present)
zipFileMemoryStream.Seek(compressedSize, SeekOrigin.Current);
// Central directory structure
writer.Write((uint)0x02014b50); // central file header signature
writer.Write((ushort)20); // version made by
writer.Write((ushort)20); // version needed to extract (2.0 == compressed using deflate)
writer.Write((ushort)0); // general purpose bit flag
writer.Write((ushort)8); // compression method
writer.Write((ushort)0); // last mod file time
writer.Write((ushort)0); // last mod file date
writer.Write(crc32); // crc-32
writer.Write(compressedSize); // compressed size
writer.Write(isize); // uncompressed size
writer.Write((ushort)1); // file name length
writer.Write((ushort)0); // extra field length
writer.Write((ushort)0); // file comment length
writer.Write((ushort)0); // disk number start
writer.Write((ushort)0); // internal file attributes
writer.Write((uint)0); // external file attributes
writer.Write((uint)0); // relative offset of local header
writer.Write((byte)fileName[0]); // file name
// End of central directory record
writer.Write((uint)0x06054b50); // end of central dir signature
writer.Write((ushort)0); // number of this disk
writer.Write((ushort)0); // number of the disk with the start of the central directory
writer.Write((ushort)1); // total number of entries in the central directory on this disk
writer.Write((ushort)1); // total number of entries in the central directory
writer.Write((uint)(46 + 1)); // size of the central directory (46 bytes + 1 character for file name)
writer.Write((uint)(zipHeaderSize + compressedSize)); // offset of start of central directory with respect to the starting disk number
writer.Write((ushort)0); // .ZIP file comment length
// Reset ZIP file stream to beginning
zipFileMemoryStream.Seek(0, SeekOrigin.Begin);
// Return the decompressed stream
_inflatedStream = Application.GetResourceStream(
new StreamResourceInfo(zipFileMemoryStream, null),
new Uri(fileName, UriKind.Relative))
.Stream;
}
public override bool CanRead
{
get { return _inflatedStream.CanRead; }
}
public override bool CanSeek
{
get { return _inflatedStream.CanSeek; }
}
public override bool CanWrite
{
get { return _inflatedStream.CanWrite; }
}
public override void Flush()
{
_inflatedStream.Flush();
}
public override long Length
{
get { return _inflatedStream.Length; }
}
public override long Position
{
get { return _inflatedStream.Position; }
set { _inflatedStream.Position = value; }
}
public override int Read(byte[] buffer, int offset, int count)
{
return _inflatedStream.Read(buffer, offset, count);
}
public override long Seek(long offset, SeekOrigin origin)
{
return _inflatedStream.Seek(offset, origin);
}
public override void SetLength(long value)
{
_inflatedStream.SetLength(value);
}
public override void Write(byte[] buffer, int offset, int count)
{
throw new NotSupportedException();
}
public override void Close()
{
_deflatedStream.Close();
_inflatedStream.Close();
}
protected override void Dispose(bool disposing)
{
_deflatedStream.Dispose();
_inflatedStream.Dispose();
}
public override IAsyncResult BeginRead(byte[] buffer, int offset, int count, AsyncCallback callback, object state)
{
return _inflatedStream.BeginRead(buffer, offset, count, callback, state);
}
public override int ReadByte()
{
return _inflatedStream.ReadByte();
}
public override int EndRead(IAsyncResult asyncResult)
{
return _inflatedStream.EndRead(asyncResult);
}
public override int ReadTimeout
{
get { return _inflatedStream.ReadTimeout; }
set { _inflatedStream.ReadTimeout = value; }
}
public override bool CanTimeout
{
get { return _inflatedStream.CanTimeout; }
}
}
}
@advancedrei

Line 193 relies on classes not available in PCLs... can you please provide a version that will compile to portable-net403+sl4+wp71+win8? Thanks!

@advancedrei

This works in Silverlight & Windows Phone, but if you are building a PCL that needs compression, this will do the job: https://github.com/advancedrei/HttpClient.Compression (check the readme for details on how to use from NuGet).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Something went wrong with that request. Please try again.