private void BuildDupFilesList(List <DupFiles> dupsFiles)
{
verticalListView_results.Controls.Clear();
label_dupResTitle.Text = "Found " + dupsFiles.Count + " duplications:";
foreach (DupFiles df in dupsFiles)
{
DupResult dupResult = new DupResult();
dupResult.NumOfDups = df.DuplicationsFiles.Count;
dupResult.AddOnHandleBtClickListener((s, e) =>
{
DupMatchsForm dmf = new DupMatchsForm(df);
dmf.ShowDialog();
});
verticalListView_results.AddControl(dupResult);
}
}
private static void Perf(IDupDetector dupDetector, int workers, string folder, int times)
{
var files = Directory.GetFiles(folder, "*.*", SearchOption.AllDirectories);
var timer = new Stopwatch();
DupResult result = default;
if (times <= 0)
{
times = 10;
}
timer.Start();
for (var i = 0; i < times; i++)
{
result = dupDetector.Find(files, workers);
}
timer.Stop();
Log(string.Format("Dup method: {0}, workers: {1}, groups: {2}, times: {3}, avg elapse: {4}", dupDetector, workers, result.Duplicates.Count, times, TimeSpan.FromMilliseconds(timer.ElapsedMilliseconds / times)), true);
}
public DupResult Find(IEnumerable <string> files, int workers, int quickHashSize, int bufferSize)
{
var result = new DupResult {
Duplicates = new List <Duplicate>(), FailedToProcessFiles = new List <string>(), TotalFiles = files.LongCount()
};
var totalComparedFiles = 0L;
var totalFileBytes = 0L;
var totalReadBytes = 0L;
_workers = workers;
if (_workers <= 0)
{
_workers = 5;
}
if (bufferSize <= 3)
{
bufferSize = DefaultBufferSize;
}
if (quickHashSize <= 0)
{
quickHashSize = DefaulfQuickHashSize;
}
//groups with same file size
var sameSizeGroups = files.Select(f =>
{
try
{
return(GetDupFileItem(f));
}
catch (Exception)
{
result.FailedToProcessFiles.Add(f);
return(null);
}
}).Where(f => f != null).GroupBy(f => f.Size).Where(g => g.Count() > 1);
var mappedSameSizeGroupList = new ConcurrentBag <IGrouping <string, DupItem> >();
Parallel.ForEach(MapFileSizeGroups(sameSizeGroups), mappedSameSizeGroups =>
{
foreach (var group in mappedSameSizeGroups)
{
foreach (var file in group)
{
Interlocked.Increment(ref totalComparedFiles);
try
{
//fast random bytes checking
QuickHashFile(file, quickHashSize, ref totalFileBytes, ref totalReadBytes);
}
catch (Exception)
{
file.Status = CompareStatus.Failed;
result.FailedToProcessFiles.Add(file.FileName);
}
}
//groups with same quick hash value
var sameQuickHashGroups = group.Where(f => f.Status != CompareStatus.Failed).GroupBy(f => f.QuickHash).Where(g => g.Count() > 1);
foreach (var sameQuickHashGroup in sameQuickHashGroups)
{
mappedSameSizeGroupList.Add(sameQuickHashGroup);
}
}
});
Parallel.ForEach(MapFileHashGroups(mappedSameSizeGroupList), mappedSameSizehGroups =>
{
foreach (var quickHashGroup in mappedSameSizehGroups)
{
ProgressiveHash(quickHashGroup, bufferSize, ref totalReadBytes);
result.FailedToProcessFiles.AddRange(quickHashGroup.Where(f => f.Status == CompareStatus.Failed).Select(f => f.FileName));
//phew, finally.....
//group by same file hash
var sameFullHashGroups = quickHashGroup.Where(g => g.Status != CompareStatus.Failed).GroupBy(g => g.FullHash).Where(g => g.Count() > 1);
result.Duplicates.AddRange(sameFullHashGroups.Select(fullHashGroup => new Duplicate {
Items = fullHashGroup.Select(f => new FileItem {
FileName = f.FileName, ModifiedTime = f.ModifiedTime, Size = f.Size
})
}));
}
});
result.TotalComparedFiles = totalComparedFiles;
result.TotalBytesInComparedFiles = totalFileBytes;
result.TotalReadBytes = totalReadBytes;
return(result);
}
public DupResult Find(IEnumerable <string> files, int workers, int quickHashSize = 3, int bufferSize = 0)
{
var result = new DupResult {
Duplicates = new List <Duplicate>(), FailedToProcessFiles = new List <string>()
};
_workers = workers;
if (_workers <= 0)
{
_workers = 5;
}
if (bufferSize <= 3)
{
bufferSize = DefaultBufferSize;
}
//groups with same file size
var sameSizeGroups = files.Select(f =>
{
var info = new FileInfo(f);
return(new DupItem {
FileName = f, ModifiedTime = info.LastWriteTime, Size = info.Length
});
}).GroupBy(f => f.Size).Where(g => g.Count() > 1);
var mappedSameSizeGroupList = new ConcurrentBag <IGrouping <string, DupItem> >();
Parallel.ForEach(MapFileSizeGroups(sameSizeGroups), mappedSameSizeGroups =>
{
foreach (var group in mappedSameSizeGroups)
{
foreach (var file in group)
{
if (file.Size > 0)
{
//fast random byte checking
try
{
using (var stream = File.Open(file.FileName, FileMode.Open, FileAccess.Read, FileShare.Read))
{
var length = stream.Length;
file.Tags = new byte[3];
//first byte
stream.Seek(0, SeekOrigin.Begin);
file.Tags[0] = (byte)stream.ReadByte();
//middle byte, we need it especially for xml like files
if (length > 1)
{
stream.Seek(stream.Length / 2, SeekOrigin.Begin);
file.Tags[1] = (byte)stream.ReadByte();
}
//last byte
if (length > 2)
{
stream.Seek(0, SeekOrigin.End);
file.Tags[2] = (byte)stream.ReadByte();
}
file.QuickHash = HashTool.GetHashText(file.Tags);
}
}
catch (Exception)
{
file.Status = CompareStatus.Failed;
result.FailedToProcessFiles.Add(file.FileName);
}
}
}
//groups with same quick hash value
var sameQuickHashGroups = group.Where(f => f.Status != CompareStatus.Failed).GroupBy(f => f.QuickHash).Where(g => g.Count() > 1);
foreach (var sameQuickHashGroup in sameQuickHashGroups)
{
mappedSameSizeGroupList.Add(sameQuickHashGroup);
}
}
});
Parallel.ForEach(MapFileHashGroups(mappedSameSizeGroupList), mappedSameHashGroups =>
{
foreach (var quickHashGroup in mappedSameHashGroups)
{
foreach (var groupFile in quickHashGroup)
{
try
{
groupFile.FullHash = HashTool.HashFile(groupFile.FileName, bufferSize);
}
catch (Exception)
{
result.FailedToProcessFiles.Add(groupFile.FileName);
}
}
//phew, finally.....
//group by same file hash
var sameFullHashGroups = quickHashGroup.GroupBy(g => g.FullHash).Where(g => g.Count() > 1);
result.Duplicates.AddRange(sameFullHashGroups.Select(fullHashGroup => new Duplicate {
Items = fullHashGroup.Select(f => new FileItem {
FileName = f.FileName, ModifiedTime = f.ModifiedTime, Size = f.Size
})
}));
}
});
return(result);
}
