/// <summary>
/// Detects the byte order mark of a file and returns
/// an appropriate encoding for the file.
/// </summary>
/// <param name="sourceFile"></param>
/// <returns>A TextEncodingMetadata object</returns>
public static TextEncodingMetadata GetFileEncoding(string sourceFile)
{
var metaData = new TextEncodingMetadata();
EncodingMetaInfo metaInfo = DetectBom(sourceFile);
metaData.HasBom = metaInfo.HasBom;
using (FileStream fs = File.OpenRead(sourceFile))
{
var cdet = new Ude.CharsetDetector();
cdet.Feed(fs);
cdet.DataEnd();
if (cdet.Charset != null)
{
metaData.CharacterSet = cdet.Charset;
metaData.DetectionConfidence = cdet.Confidence;
}
else
{
Console.WriteLine("Detection failed.");
}
}
using (var sr = new StreamReader(sourceFile))
{
Encoding encoding = sr.CurrentEncoding;
metaData.FileEncoding = encoding;
}
return(metaData);
}
/// <summary>
/// Detects the bom.
/// </summary>
/// <param name="filePath">The file path.</param>
/// <returns>An EncodingMetaInfo object</returns>
public static EncodingMetaInfo DetectBom(string filePath)
{
var metaInfo = new EncodingMetaInfo();
// *** Use Default of Encoding.Default (Ansi CodePage)
Encoding enc = Encoding.Default;
// *** Detect byte order mark if any - otherwise assume default
var buffer = new byte[5];
var file = new FileStream(filePath, FileMode.Open);
file.Read(buffer, 0, 5);
file.Close();
if (buffer[0] == 0xef && buffer[1] == 0xbb && buffer[2] == 0xbf)
{
metaInfo.HasBom = true;
metaInfo.TextEncoding = Encoding.UTF8;
}
else if (buffer[0] == 0xff && buffer[1] == 0xfe)
{
metaInfo.HasBom = true;
metaInfo.TextEncoding = Encoding.Unicode; // utf-16le
}
else if (buffer[0] == 0xfe && buffer[1] == 0xff)
{
metaInfo.HasBom = true;
metaInfo.TextEncoding = Encoding.BigEndianUnicode; // utf-16be
}
else if (buffer[0] == 0 && buffer[1] == 0 && buffer[2] == 0xfe && buffer[3] == 0xff)
{
metaInfo.HasBom = true;
metaInfo.TextEncoding = Encoding.UTF32;
}
else if (buffer[0] == 0x2b && buffer[1] == 0x2f && buffer[2] == 0x76)
{
metaInfo.HasBom = true;
metaInfo.TextEncoding = Encoding.UTF7;
}
return(metaInfo);
}