Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- type
- //UTF8String = string;
- UTF8String = AnsiString;
- function Utf8Decode(const S: UTF8String): WideString;
- var
- L: Integer;
- Temp: WideString;
- begin
- Result := '';
- if S = '' then Exit;
- SetLength(Temp, Length(S));
- L := Utf8ToUnicode(PWideChar(Temp), Length(Temp)+1, PChar(S), Length(S));
- if L > 0 then
- SetLength(Temp, L-1)
- else
- Temp := '';
- Result := Temp;
- end;
- function Utf8ToUnicode(Dest: PWideChar; Source: PChar; MaxChars: Integer): Integer;
- var
- len: Cardinal;
- begin
- len := 0;
- if Source <> nil then
- while Source[len] <> #0 do
- Inc(len);
- Result := Utf8ToUnicode4(Dest, MaxChars, Source, len);
- end;
- function Utf8ToUnicode4(Dest: PWideChar; MaxDestChars: Cardinal; Source: PChar; SourceBytes: Cardinal): Cardinal;
- var
- i, count: Cardinal;
- c: Byte;
- wc: Cardinal;
- begin
- if Source = nil then
- begin
- Result := 0;
- Exit;
- end;
- Result := Cardinal(-1);
- count := 0;
- i := 0;
- if Dest <> nil then
- begin
- while (i < SourceBytes) and (count < MaxDestChars) do
- begin
- wc := Cardinal(Source[i]);
- Inc(i);
- if (wc and $80) <> 0 then
- begin
- if i >= SourceBytes then Exit; // incomplete multibyte char
- wc := wc and $3F;
- if (wc and $20) <> 0 then
- begin
- c := Byte(Source[i]);
- Inc(i);
- if (c and $C0) <> $80 then Exit; // malformed trail byte or out of range char
- if i >= SourceBytes then Exit; // incomplete multibyte char
- wc := (wc shl 6) or (c and $3F);
- end;
- c := Byte(Source[i]);
- Inc(i);
- if (c and $C0) <> $80 then Exit; // malformed trail byte
- Dest[count] := WideChar((wc shl 6) or (c and $3F));
- end
- else
- Dest[count] := WideChar(wc);
- Inc(count);
- end;
- if count >= MaxDestChars then count := MaxDestChars-1;
- Dest[count] := #0;
- end
- else
- begin
- while (i < SourceBytes) do
- begin
- c := Byte(Source[i]);
- Inc(i);
- if (c and $80) <> 0 then
- begin
- if i >= SourceBytes then Exit; // incomplete multibyte char
- c := c and $3F;
- if (c and $20) <> 0 then
- begin
- c := Byte(Source[i]);
- Inc(i);
- if (c and $C0) <> $80 then Exit; // malformed trail byte or out of range char
- if i >= SourceBytes then Exit; // incomplete multibyte char
- end;
- c := Byte(Source[i]);
- Inc(i);
- if (c and $C0) <> $80 then Exit; // malformed trail byte
- end;
- Inc(count);
- end;
- end;
- Result := count+1;
- end;
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement