Reputation: 1900
I am writing an API that allows users to upload files (image, video, etc). I use a SHA-1 hash to make sure the same file isn't uploaded multiple times. Previously we only allowed smaller files so I was reading them into a byte array and hashing that but now we allow larger files so I am using a file stream. The problem is the SHA-1 algorithm returns a different hash. I need to figure out how to get the same hash regardless of the method, even if I have to turn the byte array into a file stream or something. However, I've tried writing the byte array to a temp file and reading it in and it returns the same hash as the byte array. Here is an example console app that shows what I am doing:
static void Main(string[] args)
{
string file = "C:\\CUWCDFileStorage\\temp\\test.png";
var bytes = File.ReadAllBytes(file);
using (var stream = File.Open(file, FileMode.Open))
{
Console.WriteLine(Sha1HashFile(bytes)); // Returns B7F6D90C30233F91FCEFE05FB49679F8B26C9D80
Console.WriteLine(Sha1HashFile(stream)); // Returns DA39A3EE5E6B4B0D3255BFEF95601890AFD80709
Console.WriteLine(Sha1HashFile2(bytes)); // Returns B7F6D90C30233F91FCEFE05FB49679F8B26C9D80
}
Console.Read();
}
public static string Sha1HashFile(byte[] file)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(file)).Replace("-", "");
}
}
public static string Sha1HashFile(Stream stream)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(stream)).Replace("-", "");
}
}
public static string Sha1HashFile2(byte[] bytes)
{
string file = "C:\\CUWCDFileStorage\\temp\\test2.png";
File.WriteAllBytes(file, bytes);
return Sha1HashFile(File.OpenRead(file));
}
I've even tried to just put the byte array into a MemoryStream
with new MemoryStream(bytes)
but that didn't work either. It seems like once I have the file in a byte array it can't be put back.
EDIT:
I removed some code from my example because I thought MD5 was working. Here is the original code I was using to test:
static void Main(string[] args)
{
string file = "C:\\CUWCDFileStorage\\temp\\test.png";
var bytes = File.ReadAllBytes(file);
using (var stream = File.Open(file, FileMode.Open))
{
Console.WriteLine(Md5HashFile(bytes));
Console.WriteLine(Md5HashFile(stream));
Console.WriteLine(Sha1HashFile(bytes));
Console.WriteLine(Sha1HashFile(stream));
Console.WriteLine(Sha1HashFile2(bytes));
}
Console.Read();
}
public static string Md5HashFile(byte[] file)
{
using (MD5 md5 = MD5.Create())
{
return BitConverter.ToString(md5.ComputeHash(file)).Replace("-", "");
}
}
public static string Sha1HashFile(byte[] file)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(file)).Replace("-", "");
}
}
public static string Md5HashFile(Stream stream)
{
using (MD5 md5 = MD5.Create())
{
return BitConverter.ToString(md5.ComputeHash(stream)).Replace("-", "");
}
}
public static string Sha1HashFile(Stream stream)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(stream)).Replace("-", "");
}
}
public static string Sha1HashFile2(byte[] bytes)
{
string file = "C:\\CUWCDFileStorage\\temp\\test2.png";
File.WriteAllBytes(file, bytes);
return Sha1HashFile(File.OpenRead(file));
}
See answer below for explanation of the problem.
Upvotes: 4
Views: 2702
Reputation: 1900
The problem is that the stream is being read to the end when hashing the first way. That is causing the second hash to be wrong. Because of that, I need to either reopen a stream for the second hash or rewind the stream to the beginning before hashing the second way. Here is the solution:
static void Main(string[] args)
{
string file = "C:\\CUWCDFileStorage\\temp\\test.png";
var bytes = File.ReadAllBytes(file);
using (var stream = File.Open(file, FileMode.Open))
{
Console.WriteLine(Md5HashFile(bytes));
Console.WriteLine(Md5HashFile(stream));
}
using (var stream = File.Open(file, FileMode.Open))
{
Console.WriteLine(Sha1HashFile(bytes));
Console.WriteLine(Sha1HashFile(stream));
Console.WriteLine(Sha1HashFile2(bytes));
}
Console.Read();
}
public static string Md5HashFile(byte[] file)
{
using (MD5 md5 = MD5.Create())
{
return BitConverter.ToString(md5.ComputeHash(file)).Replace("-", "");
}
}
public static string Sha1HashFile(byte[] file)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(file)).Replace("-", "");
}
}
public static string Md5HashFile(Stream stream)
{
using (MD5 md5 = MD5.Create())
{
return BitConverter.ToString(md5.ComputeHash(stream)).Replace("-", "");
}
}
public static string Sha1HashFile(Stream stream)
{
using (SHA1Managed sha1 = new SHA1Managed())
{
return BitConverter.ToString(sha1.ComputeHash(stream)).Replace("-", "");
}
}
public static string Sha1HashFile2(byte[] bytes)
{
string file = "C:\\CUWCDFileStorage\\temp\\test2.png";
File.WriteAllBytes(file, bytes);
return Sha1HashFile(File.OpenRead(file));
}
Upvotes: 7