Lazarus 1.8 Unicode 字符与UTF8字符的转换

// UCS2格式文本转换为UTF8
// UCS2格式以两个字节UNICODE编码字符，所以只有保存码值为0XFFFF 之前的字符
// UCS2格式保存时，是低字先存，高字后存，如严字UNICODE编码为4E25,保存时先存25,再存4E
// C1,C2为UCS2文件中连续的两个字节

function UCS2ToUTF8(C1,C2:byte):UTF8string;
 var
   SH,SM,SL:Byte;
   VL,VH:Byte;
   V:Word;
 begin
   result:='';

   VL:=C1;  //25
   VH:=C2;  //4E
   V:=VH;
   V:=(V SHL 8)+VL;

   if (V>=$0000) AND (V<=$007F) then
      begin
         result:=Chr(VL);
      end
   else if (v>=$0080) and (v<=$07ff) then
      begin
         SM:=VL SHR 6;
         VL:=%10000000 +(%00111111 AND VL);
         VH:=%11000000 +(VH SHL 2)+SM;
         Result:=chr(VH)+CHR(VL);
      end
   else if (v>=$0800) and (v<=$FFFF) then
      begin
                 SL:=%10000000+(%00111111 AND VL);
                 SM:=%10000000+(VL SHR 6);
                 SM:=SM+((VH SHL 2) AND %00111100);
                 SH:=%11100000+(VH SHR 4);
                 Result:=chr(SH)+CHR(SM)+CHR(SL);
      end;
 end;     

下面是检测文本文件格式的函数：
//检测文本文件编码格式

function  TextEncode(FileName:String):string;
var
  fs:TFileStream;
  flag:array[0..1] of byte;
begin
  flag[0]:=0;
  flag[1]:=0;
  fs:=TFileStream.Create(FileName,fmOpenRead);
  ALenth:=fs.Size;
  fs.read(flag,2);
  fs.Free;
  if (flag[0]=$EF) AND (flag[1]=$BB) then
     Result:='UTF-8'
  else if (flag[0]=$FE) AND (flag[1]=$FF) then
     Result:='BigEndianUnicode'
  else if (flag[0]=$FF) AND (flag[1]=$FE) then
     Result:='Unicode'
  else
    Result:='ANSI';
end;

转码原理参考文档：字符编码笔记 阮一峰的博客文章