{*******************************************************************************   UTF-8就是以8位为单元对UCS进行编码。从UCS-2到UTF-8的编码方式如下：   UCS-2编码(16进制) UTF-8 字节流(二进制)   0000 - 007F 0xxxxxxx  0080 - 07FF 110xxxxx 10xxxxxx  0800 - FFFF 1110xxxx 10xxxxxx 10xxxxxx   例如“汉”字的Unicode编码是6C49。6C49在0800-FFFF之间，所以肯定要用3字节模板了：1110xxxx 10xxxxxx 10xxxxxx。将6C49写成二进制是：0110 110001 001001， 用这个比特流依次代替模板中的x，得到：11100110 10110001 10001001，即E6 B1 89。 *******************************************************************************} unit uUnicodeUtils; interfaceuses SysUtils, windows; function Gb2Utf8(const GbStr:String):String;function Utf82Gb(const Utf8Str:String):String; implementation function Gb2Utf8(const GbStr:String):String;var  wideStr:array[0..2048] of WideChar;  SourceLength:integer;  DoneLength:integer;  AscNo:integer;  Byte1,Byte2,Byte3:integer;begin  Result := '';  //不做变化  //Result:=GbStr;  //exit;   SourceLength := Length(GbStr);  if SourceLength = 0 then exit;   SourceLength:=MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED, PAnsiChar(GbStr), SourceLength, wideStr, SourceLength);   if SourceLength = 0 then exit;   DoneLength := 0;    repeat    AscNo := Integer(wideStr[DoneLength]);    case AscNo of      0..$007F: begin        Result := Result + Chr(AscNo);      end;      $0080..$07FF: begin        Byte1 := ((AscNo shr 6) and $001F) + $00C0;        Byte2 := (AscNo and $003F) + $0080;        Result := Result + Chr(Byte1) + Chr(Byte2);      end;      $0800..$FFFF: begin        Byte1 := ((AscNo shr 12) and $000F) + $00E0;        Byte2 := ((AscNo shr 6) and $003F) + $0080;        Byte3 := (AscNo and $003F) + $0080;        Result := Result + Chr(Byte1) + Chr(Byte2) + Chr(Byte3);      end;    end;    Inc(DoneLength);  until DoneLength = SourceLength;end; function Utf82Gb(const Utf8Str:String):String;var  SourceLength:integer;  DoneLength:integer;  AscNo:integer;  Byte1,Byte2,Byte3:integer;begin  Byte1:=0;  Byte2:=0;  Byte3:=0;  Result:='';   //不做变化  //Result:=Utf8Str;  //exit;    if Trim(Utf8Str)='' then exit;   SourceLength:=Length(Utf8Str);  DoneLength:=1;  repeat  AscNo:=ord(Utf8Str[DoneLength]);  case (AscNo and $E0) of  $E0:begin    Byte1:=(AscNo and $0f) shl 12;    Inc(DoneLength);    if DoneLength>SourceLength then break;    AscNo:=ord(Utf8Str[DoneLength]);    Byte2:=(AscNo and $3f) shl 6;    Inc(DoneLength);    if DoneLength>SourceLength then break;    AscNo:=ord(Utf8Str[DoneLength]);    Byte3:=AscNo and $3f;  end;  $C0:begin    Byte1:=(AscNo and $1f) shl 6;    Inc(DoneLength);    if DoneLength>SourceLength then break;    AscNo:=ord(Utf8Str[DoneLength]);    Byte2:=(AscNo and $3f);    Byte3:=0;  end;  0..$60:begin    Byte1:=AscNo;    Byte2:=0;    Byte3:=0;    end;  end;  //case;    Result:=Result+widechar(Byte1+Byte2+Byte3);    Inc(DoneLength);    if DoneLength>SourceLength then break;  until DoneLength > SourceLength;end;end.

评论