I have to MD5 hash files/folders on both a client(C#) and a server(PHP) file structure. (Server land is PHP and client land is c#.) The problem is while they work they do not match. Any ideas would be greatly appreciated
Here are my two algorithms
C#
using System;
using System.IO;
using System.Security.Cryptography;
using System.Text;
namespace nofolder
{
public class classHasher
{
/**********
* recursive folder MD5 hash of a dir
*/
MD5 hashAlgo = null;
StringBuilder sb;
public classHasher()
{
hashAlgo = new MD5CryptoServiceProvider();
}
public string getHash(String path)
{
// get the file attributes for file or directory
if (File.Exists(path)) return getHashOverFile(path);
if (Directory.Exists(path)) return getHashOverFolder(path);
return "";
}
public string getHashOverFolder(String path)
{
sb = new StringBuilder();
getFolderContents(path);
return sb.ToString().GetHashCode().ToString();
}
public string getHashOverFile(String filename)
{
sb = new StringBuilder();
getFileHash(filename);
return sb.ToString().GetHashCode().ToString();
}
private void getFolderContents(string fold)
{
foreach (var d in Directory.GetDirectories(fold))
{
getFolderContents(d);
}
foreach (var f in Directory.GetFiles(fold))
{
getFileHash(f);
}
}
private void getFileHash(String f)
{
using (FileStream file = new FileStream(f, FileMode.Open, FileAccess.Read))
{
byte[] retVal = hashAlgo.ComputeHash(file);
file.Close();
foreach (var y in retVal)
{
sb.Append(y.ToString());
}
}
}
}
}
PHP
function include__md5_dir($dir){
/**********
* recursive folder MD5 hash of a dir
*/
if (!is_dir($dir)){
return md5_file($dir);
}
$filemd5s = array();
$d = dir($dir);
while (false !== ($entry = $d->read())){
if ($entry != '.' && $entry != '..'){
if (is_dir($dir.'/'.$entry)){
$filemd5s[] = include__md5_dir($dir.'/'.$entry);
}
else{
$filemd5s[] = md5_file($dir.'/'.$entry);
}
}
}
$d->close();
return md5(implode('', $filemd5s));
}
EDIT.
I have decided the c# must change. the PHP is fine as it is. The first code that works 100% gets the bounty
I eventually fixed this myself and I include the answer for future posterity - the key to this solution was irradicating the different default dir ORDERING that linux and windows use. This was only tested on the linux server (Cent OS6.3) and Windows 7 Client.
C#
public class classHasher
{
/**********
* recursive folder MD5 hash of a dir
*/
MD5 hashAlgo = null;
StringBuilder sb;
public classHasher()
{
hashAlgo = new MD5CryptoServiceProvider();
}
public string UltraHasher(String path)
{
/**********
* recursive folder MD5 hash of a dir
*/
if (!Directory.Exists(path))
{
return getHashOverFile(path);
}
List<string> filemd5s = new List<string>();
List<string> dir = new List<string>();
if (Directory.GetDirectories(path) != null) foreach (var d in Directory.GetDirectories(path))
{
dir.Add(d);
}
if (Directory.GetFiles(path) != null) foreach (var f in Directory.GetFiles(path))
{
dir.Add(f);
}
dir.Sort();
foreach (string entry in dir)
{
if (Directory.Exists(entry))
{
string rtn = UltraHasher(entry.ToString());
//Debug.WriteLine(" ULTRRAAHASHER:! " + entry.ToString() + ":" + rtn);
filemd5s.Add(rtn);
}
if (File.Exists(entry))
{
string rtn = getHashOverFile(entry.ToString());
//Debug.WriteLine(" FILEEEEHASHER:! " + entry.ToString() + ":" + rtn);
filemd5s.Add(rtn);
}
}
//Debug.WriteLine(" ULTRRAASUMMMM:! " + String.Join("", filemd5s.ToArray()));
string tosend = CalculateMD5Hash(String.Join("", filemd5s.ToArray()));
//Debug.WriteLine(" YEAHHAHHAHHAH:! " + tosend);
return tosend;
}
public string getHashOverFile(String filename)
{
sb = new StringBuilder();
getFileHash(filename);
return sb.ToString();
}
private void getFileHash(String f)
{
using (FileStream file = new FileStream(f, FileMode.Open, FileAccess.Read))
{
byte[] retVal = hashAlgo.ComputeHash(file);
file.Close();
foreach (var y in retVal)
{
sb.Append(y.ToString("x2"));
}
}
}
public string CalculateMD5Hash(string input)
{
byte[] inputBytes = System.Text.Encoding.ASCII.GetBytes(input);
byte[] hash = hashAlgo.ComputeHash(inputBytes);
StringBuilder sz = new StringBuilder();
for (int i = 0; i < hash.Length; i++)
{
sz.Append(hash[i].ToString("x2"));
}
return sz.ToString();
}
}
PHP
function md5_dir($dir){
/**********
* recursive folder MD5 hash of a dir
*/
if (!is_dir($dir)){
return md5_file($dir);
}
$filemd5s = array();
$bit = array();
$d = scandir($dir);
foreach($d as $entry){
if ($entry != '.' && $entry != '..'){
$bit[] = $entry;
}
}
asort($bit);
foreach($bit as $entry){
if (is_dir($dir.'/'.$entry)){
$sz = md5_dir($dir.'/'.$entry);
//echo "\n ULTRRAAHASHER:! ".$dir.'/'.$entry.":$sz";
$filemd5s[] = $sz;
}
else{
$sz = md5_file($dir.'/'.$entry);
$filemd5s[] = $sz;
//echo "\n FILEEEEHASHER:! ".$dir.'/'.$entry.":$sz";
}
}
//echo "\n ULTRRAASUMMMM:! ".implode('', $filemd5s)."";
//echo "\n YEAHHAHHAHHAH:! ".md5(implode('', $filemd5s))."";
return md5(implode('', $filemd5s));
}
these two will traverse either a C# Windows and or a PHP linux folder and return the SAME hashes for all dirs (recursive, so it includes sub dirs) inside Linuxland and all inside Windowsland.