We Keep Coding

Unzip Byte Array in C#

I'm working on EBICS protocol and i want to read a data in an XML File to compare with another file.
I have successfull decode data from base64 using
Convert.FromBase64String(OrderData); but now i have a byte array.
To read the content i have to unzip it. I tried to unzip it using Gzip like this example :
static byte[] Decompress(byte[] data)
{
using (var compressedStream = new MemoryStream(data))
using (var zipStream = new GZipStream(compressedStream, CompressionMode.Decompress))
using (var resultStream = new MemoryStream())
{
zipStream.CopyTo(resultStream);
return resultStream.ToArray();
}
}
But it does not work i have an error message :
the magic number in gzip header is not correct. make sure you are passing in a gzip stream
Now i have no idea how i can unzip it, please help me !
Thanks !

The first four bytes provided by the OP in a comment to another answer: 0x78 0xda 0xe5 0x98 is the start of a zlib stream. It is neither gzip, nor zip, but zlib. You need a ZlibStream, which for some reason Microsoft does not provide. That's fine though, since what Microsoft does provide is buggy.
You should use DotNetZip, which provides ZlibStream, and it works.

Try using SharpZipLib. It copes with various compression formats and is free under the GPL license.
As others have pointed out, I suspect you have a zip stream and not gzip. If you check the first 4 bytes in a hex view, ZIP files always start with 0x04034b50 ZIP File Format Wiki whereas GZIP files start with 0x8b1f GZIP File Format Wiki

I think I finally got it - as usual the problem is not what is in the title. Luckily I've noticed the word EBICS in your post. So, according to EBICS spec the data is first compressed, then encrypted and finally base64 encoded. As you see, after decoding base64 you need first to decrypt the data and then try to unzip it.
UPDATE: If that's not the case, it turns out from the EBICS spec Chapter 16 Appendix: Standards and references that ZIP refers to zlib/deflate format, so all you need to do is to replace GZipStream with the DeflateStream

Decompressing a Zip file from a string

I'm fetching an object from couchbase where one of the fields has a file. The file is zipped and then encoded in base64.
How would I be able to take this string and decompress it back to the original file?
Then, if I'm using ASP.MVC 4 - How would I send it back to the browser as a downloadable file?
The original file is being created on a Linux system and decoded on a Windows system (C#).

You should use Convert.FromBase64String to get the bytes, then decompress, and then use Controller.File to have the client download the file. To decompress, you need to open the zip file using some sort of ZIP library. .NET 4.5's built-in ZipArchive class should work. Or you could use another library, both SharpZipLib and DotNetZip support reading from streams.
public ActionResult MyAction()
{
string base64String = // get from Linux system
byte[] zipBytes = Convert.FromBase64String(base64String);
using (var zipStream = new MemoryStream(zipBytes))
using (var zipArchive = new ZipArchive(zipStream))
{
var entry = zipArchive.Entries.Single();
string mimeType = MimeMapping.GetMimeMapping(entry.Name);
using (var decompressedStream = entry.Open())
return File(decompressedStream, mimeType);
}
}
You'll also need the MIME type of the file, you can use MimeMapping.GetMimeMapping to help you get that for most common types.

I've used SharpZipLib successfully for this type of task in the past.
For an example that's very close to what you need to do have a look here.
Basically, the steps should be something like this:
you get the compressed input as a string from the database
create a MemoryStream and write the string to it
seek back to the beginning of the memory stream
use the MemoryStream as an input to the SharpZipLib ZipFile class
follow the example provided above to unpack the contents of the ZipFile
Update
If the string contains only the zipped contents of the file (not a full Zip archive) then you can simply use the GZipStream class in .NET to unzip the contents. You can find a sample here. But the initial steps are the same as above (get string from db, write to memory stream, feed memory stream as input to the GZipStream to decompress).

Detect Stream or Byte Array Encoding on Windows Phone

I'm trying to read xmls that I downloaded with the WebClient.OpenReadAsync() in a Windows Phone application. The problem is that sometimes, the xml won't come with UTF8 Encoding, it might come with other encodings such as "ISO-8859-1", so the accents come messed up.
I was able to load one of the ISO-8858-1 xmls perfectly using the code:
var buff = e.Result.ReadFully(); //Gets byte array from the stream
var resultBuff = Encoding.Convert(Encoding.GetEncoding("ISO-8859-1"), Encoding.UTF8, buff);
var result = Encoding.UTF8.GetString(resultBuff, 0, resultBuff.Length);
It works beautifully with ISO-8859-1, the text came perfect after, but It messed up the UTF8 xmls.
So, the idea here is to detect the encoding of the byte array or the stream before doing this, then if it's not UTF8, it will convert the data using the method above with the detected encoding.
I am searching for some method that can detect the encoding on the internet but I cannot find any!
Does anybody know how I could do this kind of thing on Windows Phone?
Thanks!

You can look for the "Content-Type" value in the WebClient.ResponseHeaders property; If you are lucky the server is setting it to indicate the type of media plus its encoding (e.g. "text/html; charset=ISO-8859-4").

Playing wav with HTML5

I am trying to output a text-to-speech wav file and play it with the HTML5 <audio> tag. The text-to-speech method is outputting the bytes, but the html5 control isn't playing it.
If instead of streaming the bytes directly to the control, i save it as a file first, then convert the file to bytes with filestream and output them, it starts to play, but i don't want to have to save the file every time. I'm using MVC 4.
// in a class library
public byte[] GenerateAudio(string randomText)
{
MemoryStream wavAudioStream = new MemoryStream();
SpeechSynthesizer speechEngine = new SpeechSynthesizer();
speechEngine.SetOutputToWaveStream(wavAudioStream);
speechEngine.Speak(randomText);
wavAudioStream.Flush();
Byte[] wavBytes = wavAudioStream.GetBuffer();
return wavBytes;
}
// in my controller
public ActionResult Listen()
{
return new FileContentResult(c.GenerateAudio(Session["RandomText"].ToString()), "audio/wav");
}
// in my view
<audio controls autoplay>
<source src="#Url.Content("~/Captcha/Listen")" type="audio/wav" />
Your browser does not support the <audio> element.
</audio>

I am also playing back a wav file to an audio element, and your code has the same logic as mine. I just noticed that you are flushing your stream before you return the byte array, which will seem to be empty.
Also, you can use file as return type and pass the byte array to its constructor. The content type is just the same as in your code. I would like to mention (maybe it could be a help also) that I used 2 streams: an outside scope stream and the actual stream where the data will be saved. After I have populated the actual stream I copied its contents to the outside stream using stream.CopyTo() and the instance of that outside stream is the one I used in my return statement. This avoids the error "Cannot Access Closed stream" (not the exact the error statement).

Can PHP decompress a file compressed with the .NET GZipStream class?

I have a C# application that communicates with a PHP-based SOAP web service for updates and licensing.
I am now working on a feedback system for users to submit errors and tracelogs automatically through the software. Based on a previous question I posted, I felt that a web service would be the best way to do it (most likely to work properly with least configuration).
My current thought is to use .NET built-in gzip compression to compress the text file, convert to base64, send to the web-service, and have the PHP script convert to binary and uncompress the data.
Can PHP decompress data compressed with GZipStream, and if so, how?

I actually tried this. GZipStream doesn't work. On the other hand, compressing with DeflateStream on .NET side and decompressing with gzinflate on PHP side do work. Your mileage may vary...

If the http-level libraries implements it (Both client and server), http has support for gzip-compression, in which case there would be no reason to manually compress anything. You should check if this is already happening before you venture any further.

Since the server is accepting web requests you really should be checking the HTTP headers to determine if any client accepts GZIP encoding rather than just guessing and gzipping each and every time.
If the PHP client can do gzip itll set the header and your code will then react according and do the right thing. Assuming or guessing is a poor choice when the facility is provided for your code to learn the capabilities of the client.

I wrote an article I recently posted that shows how to compress/decompress in C#. I used it for almost the same scenario. I wanted to transfer log files from the client to the server and they were often quite large. However in my case my webservice was running in .NET so I could use the decompress method. But looks like PHP does support a method called gzdecode that would work.
http://coding.infoconex.com/post/2009/05/Compress-and-Decompress-using-net-framework-and-built-in-GZipStream.aspx

Yes, PHP can decompress GZIP compressed strings, with or without headers.
gzdecode for GZIP file format (ie, compatible with gzip)
gzinflate for "raw" DEFLATE format
gzuncompress for ZLIB format (GZIP format without some header info)
I don't know for sure which one you'd want as I'm unfamiliar with .NET GZipStream. It sounds a little like gzuncompress, as the ZLIB format is kind of a "streaming" format, but try all three.

I was able to demo this with Gzip on C# and PHP.
Gzip Compressing in C#:
using System;
using System.IO;
using System.IO.Compression;
using System.Text;
public class Program {
public static void Main() {
string s = "Hi!!";
byte[] byteArray = Encoding.UTF8.GetBytes(s);
byte[] b2 = Compress(byteArray);
Console.WriteLine(System.Convert.ToBase64String(b2));
}
public static byte[] Compress(byte[] bytes) {
using (var memoryStream = new MemoryStream()) {
using (var gzipStream = new GZipStream(memoryStream, CompressionLevel.Optimal)) {
gzipStream.Write(bytes, 0, bytes.Length);
}
return memoryStream.ToArray();
}
}
public static byte[] Decompress(byte[] bytes) {
using (var memoryStream = new MemoryStream(bytes)) {
using (var outputStream = new MemoryStream()) {
using (var decompressStream = new GZipStream(memoryStream, CompressionMode.Decompress)) {
decompressStream.CopyTo(outputStream);
}
return outputStream.ToArray();
}
}
}
}
the code above prints the base64 encoded compressed string which is H4sIAAAAAAAEAPPIVFQEANxaFPgEAAAA for the Hi!! input.
Here's the code to decompress in PHP:
echo gzdecode(base64_decode('H4sIAAAAAAAEAPPIVFQEANxaFPgEAAAA'));