. Net (C #): Detect encoding from the file

Today, I encountered a problem. I need to try to obtain the text encoding from a file. It's not difficult. It's nothing more than determining the Character Sequence mark (BOM) based on the first few bytes. If the file does not contain Bom, a default value is returned.

 

Code:

/// <Summary>

/// Returns the character encoding Based on the object Attempt

/// </Summary>

/// <Param name = "file"> file path </param>

/// <Param name = "defenc"> No Default Code returned by BOM </param>

/// <Returns> If the object cannot be read, null is returned. Otherwise, the encoding or default encoding Based on BOM is returned (no Bom is available ). </Returns>

Static encoding getencoding (string file, encoding defenc)

{

Using (VAR stream = file. openread (File ))

{

// Judge whether the stream is readable?

If (! Stream. Canread)

Return NULL;

// Byte array storage BOM

VaR BOM = new byte [4];

// Actual read Length

Int readc;

 

Readc = stream. Read (Bom, 0, 4 );

 

If (readc> = 2)

{

If (readc> = 4)

{

// UTF32, big-Endian

If (checkbytes (Bom, 4, 0x00, 0x00, 0xfe, 0xff ))

Return new utf32encoding (True, true );

// UTF32, little-Endian

If (checkbytes (Bom, 4, 0xff, 0xfe, 0x00, 0x00 ))

Return new utf32encoding (false, true );

}

// Utf8

If (readc> = 3 & checkbytes (Bom, 3, 0xef, 0xbb, 0xbf ))

Return new utf8encoding (true );

 

// UTF16, big-Endian

If (checkbytes (Bom, 2, 0xfe, 0xff ))

Return new unicodeencoding (True, true );

// UTF16, little-Endian

If (checkbytes (Bom, 2, 0xff, 0xfe ))

Return new unicodeencoding (false, true );

}

 

Return defenc;

}

}

 

// Auxiliary function to determine the value in the byte

Static bool checkbytes (byte [] bytes, int count, Params int [] values)

{

For (INT I = 0; I <count; I ++)

If (Bytes [I]! = Values [I])

Return false;

Return true;

}

 

For example, use the above Code to check the encoding information of all TXT files in the "My Documents" directory:

Foreach (var file in directory. getfiles (environment. getfolderpath (environment. specialfolder. mydocuments), "*. txt "))

Console. writeline ("{0} \ n {1} \ n", file, getencoding (file, encoding. ASCII). encodingname );