WSH: Определение кодировки файла

FILE_TO_READ = 'test.txt';
BYTES_FOR_READ = 1;

fso = new ActiveXObject("Scripting.FileSystemObject");
stream = new ActiveXObject("ADODB.stream");

stream.Type = 1;
stream.Open();
stream.LoadFromFile(FILE_TO_READ);
DATA = stream.Read(BYTES_FOR_READ);
stream.Close();

if (DATA == '0xFF' ) {
    WScript.Echo("OK")
}

stream.Type = 2;
stream.Charset = 'ISO-8859-1';
stream.Open();
stream.LoadFromFile(FILE_TO_READ);
DATA = stream.ReadText(1);
stream.Close();
if (DATA == "я" ) {
    WScript.Echo(DATA)
}

var GlobalObject = this;
var FSO = fso = new ActiveXObject("Scripting.FileSystemObject");
var WshShell = new ActiveXObject("WScript.Shell");

Function.prototype.GetResource = function(ResourceName) {
    if (!this.Resources) {
        var UnNamedResourceIndex = 0,
            _this = this;
        this.Resources = {};

        function f(match, resType, Content) {
            _this.Resources[(resType == "[[") ? UnNamedResourceIndex++:resType.slice(1, -1)] = Content;
        }
        this.toString().replace(/\/\*(\[(?:[^\[]+)?\[)((?:[\r\n]|.)*?)\]\]\*\//gi, f);
    }
    return this.Resources[ResourceName];
}

function Resourses() {
    /*[vbs[
Function [_MidB](v, s, l):[_MidB]=MidB(v, s+1, l):End Function
Function [_Asc](v):[_Asc]=Asc(v):End Function
Function [_Chr](v):[_Chr]=Chr(v):End Function
Function [_VarType](v):[_VarType]=VarType(v):End Function
Function [_TypeName](v):[_TypeName]=TypeName(v):End Function
Function [_ChrB](v):[_ChrB]=ChrB(v):End Function
Function [_AscB](v):[_AscB]=AscB(v):End Function
Function [_LeftB](v, l):[_LeftB]=LeftB(v, l):End Function
Function [_RightB](v, l):[_RightB]=RightB(v, l):End Function
Function [_InstrB](start, stringtosearch, stringtofind, comparemode):[_InstrB]=InstrB(start, stringtosearch, stringtofind, comparemode):End Function

Sub ConvertBinaryToJSArray(arrByteArray, js_arr)
Dim i
For i = 1 To LenB(arrByteArray)
    js_arr.push(AscB(MidB(arrByteArray, i, 1)))
Next
End Sub
]]*/
}

var VBSEngine = new ActiveXObject('ScriptControl');
VBSEngine.Language = 'VBScript';
VBSEngine.AddCode(Resourses.GetResource("vbs"));

WshShell.CurrentDirectory = GetScriptDir();

Array.prototype.map = function(mapper, that) {
    var other = new Array(this.length);
    for (var i = 0, n = this.length; i < n; i++)
    other[i] = mapper.call(that, this[i], i, this);
    return other;
}

function BinaryToArray(binary) {
    var ar = [];
    VBSEngine.CodeObject.ConvertBinaryToJSArray(binary, ar);
    return ar;
}

function TypeOf(o) {
    var type = typeof o;
    if (type != "object" && type != "unknown") return type;
    if (o === null || o === undefined) return "empty";
    if (Object.prototype.toString.call(o) === '[object Array]') return "array";
    if (Object.prototype.toString.call(o) === '[object Number]') return "number";
    if (Object.prototype.toString.call(o) === '[object String]') return "string";
    if (VBSEngine.CodeObject._VarType(o) == (8192 + 17)) return "bytearray";
    return type;
}

function Exit() {
    if (this.WScript) WScript.Quit();
    if (this.window) window.close();
}

function MsgBox(strText, strTitle, nType) {
    return WshShell.Popup(strText, 0, strTitle, nType);
}

function GetScriptDir() {
    return GetScriptFullPath().replace(/\\+[^\\]+$/g, "");
}

function GetScriptFullPath() {
    if (this.location && this.location.href) return location.href.replace(/^.+\/\/\//, "").replace(/\//g, "\\");
    if (this.WScript) return WScript.ScriptFullName;
    return "";
}

function StringToASCIIArray(string, start, end) {
    start = start || 0;
    end = end || string.length;
    string = string.slice(start, end);
    return string.split("").map(function a(s) {
        return Asc(s);
    })
}

function Asc(str) {
    return VBSEngine.CodeObject._Asc(str);
}

function Hex(expression, length) {
    var res = expression.toString(16);
    length = Math.max(length || 0, res.length);
    return new Array(length + 1).join("0").replace(new RegExp(".{" + res.length + "}$"), res).toUpperCase();
}

function Bin(expression) {
    var res = expression.toString(2);
    var length = Math.floor(res.length / 8) + ((res.length % 8 > 0) ? 1 : 0);
    length *= 8;
    return new Array(length + 1).join("0").replace(new RegExp(".{" + res.length + "}$"), res).toUpperCase();
}

function FileGetSize(path) {
    try {
        return FSO.GetFile(path).Size;
    } catch(e) {
        return 0;
    }
}

function FileGetEncoding(path) {
    var f = FSO.OpenTextFile(WshShell.ExpandEnvironmentStrings(path), 1, 0, 0);
    var n = FSO.GetFile(path).Size;
    n = n <= 60 ? n : 60;
    var val = StringToASCIIArray(f.Read(n));
    f.Close();
    return EncodingDetect(val);
}

function EncodingDetect(str) {
    var ar, type = TypeOf(str);
    if (type == "string") ar = StringToASCIIArray(str);
    if (type == "bytearray") ar = BinaryToArray(str);
    if (type == "array") ar = (str);
    if (!ar) return 0;
    var head = ar.map(function(x) {
        return Hex(x, 2)
    }).join("");
    if (/^EFBBBF/.test(head)) return "UTF-8";
    if (/^0000FEFF/.test(head)) return "UTF-32BE";
    if (/^FFFE0000/.test(head)) return "UTF-32LE";
    if (/^2B2F76(38|39|2B|2F)/.test(head)) return "UTF-7";
    if (/^F7644C/.test(head)) return "ISO-10646-UTF-1";
    if (/^DD736673/.test(head)) return "UTF-EBCDIC";
    if (/^0EFEFF/.test(head)) return "SCSU";
    if (/^FBEE28/.test(head)) return "BOCU-1";
    if (/^84319533/.test(head)) return "GB-18030";
    if (/^FEFF0000/.test(head)) return "ISO-10646-UCS-4";
    if (/^0000FFFE/.test(head)) return "ISO-10646-UCS-4";
    if (/^FEFF/.test(head)) return "UTF-16BE";
    if (/^FFFE/.test(head)) return "UTF-16LE";
    var head2 = ar.map(function(x) {
        return Bin(x)
    }).join("");
    
    if (/^(0.{7})+$/.test(head2)) return "ASCII"; 
    
    if (/^(0.{7}|110.{5}10.{6}|1110.{4}10.{6}10.{6}|11110.{3}10.{6}10.{6}10.{6}|111110.{2}10.{6}10.{6}10.{6}10.{6}|1111110.110.10.{6}10.{6}10.{6}10.{6}10.{6})+$/.test(head2)) return "UTF-8 without BOM";
    return "ASCII";
}

MsgBox(FileGetEncoding("123.txt"));
Exit();

var s = new ActiveXObject("ADODB.Stream");
s.type = 1;
WScript.Echo("Before load");
s.open();
s.loadFromFile("D:\\file.wmv");
WScript.Echo("After load");

var GlobalObject = this;
var FSO = fso = new ActiveXObject("Scripting.FileSystemObject");
var WshShell = new ActiveXObject("WScript.Shell");

Function.prototype.GetResource = function(ResourceName) {
    if (!this.Resources) {
        var UnNamedResourceIndex = 0,
            _this = this;
        this.Resources = {};

        function f(match, resType, Content) {
            _this.Resources[(resType == "[[") ? UnNamedResourceIndex++:resType.slice(1, -1)] = Content;
        }
        this.toString().replace(/\/\*(\[(?:[^\[]+)?\[)((?:[\r\n]|.)*?)\]\]\*\//gi, f);
    }
    return this.Resources[ResourceName];
}

Array.prototype.forEach = function (action, that /*opt*/ ) {
    for (var i = 0, n = this.length; i < n; i++)
    //if (i in this)
        action.call(that, this[i], i, this);
};

var VBSEngine = new ActiveXObject('ScriptControl');
VBSEngine.Language = 'VBScript';
VBSEngine.AddCode(Resourses.GetResource("vbs"));

function Resourses() {
    /*[vbs[
Function [_MidB](v, s, l):[_MidB]=MidB(v, s+1, l):End Function
Function [_Asc](v):[_Asc]=Asc(v):End Function
Function [_Chr](v):[_Chr]=Chr(v):End Function
Function [_LenB](v):[_LenB]=LenB(v):End Function
Function [_VarType](v):[_VarType]=VarType(v):End Function
Function [_TypeName](v):[_TypeName]=TypeName(v):End Function
Function [_ChrB](v):[_ChrB]=ChrB(v):End Function
Function [_AscB](v):[_AscB]=AscB(v):End Function
Function [_LeftB](v, l):[_LeftB]=LeftB(v, l):End Function
Function [_RightB](v, l):[_RightB]=RightB(v, l):End Function
Function [_InstrB](start, stringtosearch, stringtofind, comparemode):[_InstrB]=InstrB(start, stringtosearch, stringtofind, comparemode):End Function

Sub ConvertBinaryToJSArray(arrByteArray, js_arr)
Dim i
For i = 1 To LenB(arrByteArray)
    js_arr.push(AscB(MidB(arrByteArray, i, 1)))
Next
End Sub

Function MultiByteToBinary(MultiByte)
  ' 2000 Antonin Foller, http://www.motobit.com
  ' MultiByteToBinary converts multibyte string To real binary data (VT_UI1 | VT_ARRAY)
  ' Using recordset
  Dim RS, LMultiByte, Binary
  Const adLongVarBinary = 205
  Set RS = CreateObject("ADODB.Recordset")
  LMultiByte = LenB(MultiByte)
  If LMultiByte>0 Then
    RS.Fields.Append "mBinary", adLongVarBinary, LMultiByte
    RS.Open
    RS.AddNew
      RS("mBinary").AppendChunk MultiByte & ChrB(0)
    RS.Update
    Binary = RS("mBinary").GetChunk(LMultiByte)
  End If
  MultiByteToBinary = Binary
End Function

Function RSBinaryToString(xBinary)
'1999 Antonin Foller, Motobit Software
'This function converts binary data (VT_UI1 | VT_ARRAY or MultiByte string)
'to string (BSTR) using ADO recordset
'The fastest way - requires ADODB.Recordset
'Use this function instead of MBBinaryToString if you have ADODB.Recordset installed
'to eliminate problem with PureASP performance

Dim Binary
'MultiByte data must be converted to VT_UI1 | VT_ARRAY first.
if vartype(xBinary) = 8 then Binary = MultiByteToBinary(xBinary) else Binary = xBinary

Dim RS, LBinary
Const adLongVarChar = 201
Set RS = CreateObject("ADODB.Recordset")
LBinary = LenB(Binary)

if LBinary>0 then
RS.Fields.Append "mBinary", adLongVarChar, LBinary
RS.Open
RS.AddNew
RS("mBinary").AppendChunk Binary
RS.Update
RSBinaryToString = RS("mBinary")
Else
RSBinaryToString = ""
End If
End Function
]]*/
}
WshShell.CurrentDirectory = GetScriptDir();;

function BinaryMid(s, l, c) {
    return VBSEngine.CodeObject._MidB(s, l, c);
}

function BinaryLen(s) {
    return VBSEngine.CodeObject._LenB(s);
}

function ArrayToBinary(ar) {
    var ms = new ActiveXObject("SAPI.spMemoryStream");
    for (var i = 0, l = ar.length; i < l; i++) {
        ms.Write(ar[i]);
        ms.Seek(i + 1);
    }
    s = ms.GetData();
    var l = BinaryLen(s);
    s = BinaryMid(s, 0, l - 3)
    return MultiByteToBinary(s);
}

function Binary(s) {
    if (!s) s = 0;
    var value;
    var type = TypeOf(s);
    if (type == "array") value = ArrayToBinary(s);
    if (type == "bytearray") value = s;
    if (type == "function" || type == "object" || type == "empty") return null;
    try {
        var ms = new ActiveXObject("SAPI.spMemoryStream");
        ms.SetData(s.valueOf());
        value = ms.GetData();
    } catch(e) {}
    if (arguments.callee != this.constructor) return value;
    this.value = value;
}

function MultiByteToBinary(x) {
    return VBSEngine.CodeObject.MultiByteToBinary(x);
}

function BinaryToHex(binary) {
    var xml = new ActiveXObject('Microsoft.XMLDOM').createElement('Binary');
    xml.dataType = 'bin.hex';
    xml.nodeTypedValue = binary;
    return xml.text.toUpperCase();
}

function HexToBinary(text)
{
    var xml = new ActiveXObject('Microsoft.XMLDOM').createElement('Binary');
    xml.dataType = 'bin.hex';
    xml.text = text;
    
    return xml.nodeTypedValue;
}

function StringToBinary(str, dstCharset) { ! dstCharset ? dstCharset = "UTF-8" : 0;
    var stream = new ActiveXObject("ADODB.Stream");
    stream.type = 2;
    stream.open();
    stream.charset = dstCharset;
    stream.writeText(str);
    stream.position = 0;
    stream.type = 1;
    var result = stream.read();
    stream.close();
    return result;
}

//BinaryToString Converts a binary variant into a string.
function BinaryToString(binary, destCharset) {
    !destCharset ? destCharset = "UTF-8" : 0;
    
    var stream = new ActiveXObject("ADODB.Stream");
    stream.type = 1;
    stream.open();
    stream.write(binary);
    stream.position = 0;
    stream.type = 2;
    stream.charset = destCharset;
    var result = stream.readText();
    stream.close();
    return result;
}

function Asc(str) {
    return VBSEngine.CodeObject._Asc(str);
}

function TypeOf(o) {
    var type = typeof o;
    if (type != "object" && type != "unknown") return type;
    if (o === null || o === undefined) return "empty";
    if (Object.prototype.toString.call(o) === '[object Array]') return "array";
    if (Object.prototype.toString.call(o) === '[object Number]') return "number";
    if (Object.prototype.toString.call(o) === '[object String]') return "string";
    if (VBSEngine.CodeObject._VarType(o) == (8192 + 17)) return "bytearray";
    return type;
}

function Exit() {
    if (this.WScript) WScript.Quit();
    if (this.window) window.close();
}

function MsgBox(strText, strTitle, nType) {
    return WshShell.Popup(strText, 0, strTitle, nType);
}

function GetScriptDir() {
    return GetScriptFullPath().replace(/[^\\]+$/g, "");
}

function GetScriptFullPath() {
    if (this.location && this.location.href) return location.href.replace(/^.+\/\/\//, "").replace(/\//g, "\\");
    if (this.WScript) return WScript.ScriptFullName;
    return "";
}

function FileGetEncoding(path) {
    var f = FSO.OpenTextFile(WshShell.ExpandEnvironmentStrings(path), 1, 0, 0);
    var n = FSO.GetFile(path).Size;
    n = n <= 60 ? n : 60;
    var val = MultiByteToBinary(f.Read(n));
    f.Close();
    return EncodingDetect(val);
}

function EncodingDetect(bin) {
    var head = BinaryToHex(bin);
    if (/^EFBBBF/.test(head)) return "UTF-8";
    if (/^0000FEFF/.test(head)) return "UTF-32BE";
    if (/^FFFE0000/.test(head)) return "UTF-32LE";
    if (/^2B2F76(38|39|2B|2F)/.test(head)) return "UTF-7";
    if (/^F7644C/.test(head)) return "ISO-10646-UTF-1";
    if (/^DD736673/.test(head)) return "UTF-EBCDIC";
    if (/^0EFEFF/.test(head)) return "SCSU";
    if (/^FBEE28/.test(head)) return "BOCU-1";
    if (/^84319533/.test(head)) return "GB-18030";
    if (/^FEFF0000/.test(head)) return "ISO-10646-UCS-4";
    if (/^0000FFFE/.test(head)) return "ISO-10646-UCS-4";
    if (/^FEFF/.test(head)) return "UTF-16BE";
    if (/^FFFE/.test(head)) return "UTF-16LE";
    var o = {
        "UTF-32": 0,
        "UTF-16": 0,
        "UTF-8": 0,
        "ASCII": 0
    },
        re = /[-A-Za-z0-9А-Яа-я\?!=<>"\\\/\s\r\n]/,
        max = 0,
        maxP = "ASCII";

    function _tmp(p) {
        try {
            BinaryToString(bin, p).split("").forEach(function(s) {
                if (Asc(s) == 0) o[p] -= 0.1;
                if (re.test(s)) o[p]++
            });
            if (max <= o[p]) {
                maxP = p;
                max = o[p]
            }
        } catch(e) {}
    }
    for (i in o) _tmp(i);
    
    //MsgBox([o["UTF-32"], o["UTF-16"], o["UTF-8"], o["ASCII"]]);
    return maxP;
}
//*
MsgBox(EncodingDetect(Binary("привет")));
MsgBox(EncodingDetect(StringToBinary("привет", "UTF-8")));
MsgBox(EncodingDetect(StringToBinary("привет", "ascii")));
MsgBox(EncodingDetect(Binary("hello")));
MsgBox(EncodingDetect(StringToBinary("hello", "UTF-8")));
MsgBox(EncodingDetect(StringToBinary("hello", "ascii")));

MsgBox(FileGetEncoding("12345.txt"));
Exit();