utf 8 - Create a UTF8 file without BOM with Inno Setup (Unicode version)

Question

I have to read and modify some JSON files. The file encoding must be UTF8 without BOM or the JSON file will be not accepted.

I tried the following Code:

const
    Utf8Bom     = #$EF#$BB#$BF;
    Utf16BomLE  = #$FF#$FE;                 // little endian // 
    Utf16BomBE  = #$FE#$FF;                 // big endian // 
    Utf16Bom    = Utf16BomBE;
    CP_UTF16    = 1200;
    CP_UTF8     = 65001;

function WideStringToString (const wStr: string; codePage: Word): string;
var
    len: Integer;
begin
    len := WideCharToMultiByte (codePage, 0, wStr, -1, '', 0, 0, 0);
    if len > 0 then
    begin
        SetLength (Result, len-1);
        WideCharToMultiByte (codePage, 0, wStr, -1, Result, Length (Result), 0,  0);
    end;
end;

function ClearBom(const s, sig: string): string;
var
    i, n, len: Integer;
begin
    Result := s;
    len := Length (sig);
    n := 0;
    if (len> 0) and (Length (Result)> len) then
        repeat
            for i := 1 to len do
                if Result [1] = sig [i] then
                begin
                    Delete (Result, 1, 1);
                    Break;
                end;
            n := n + 1;
        until (n = len) or (Result = '');
end;

function ConvertUtf16(const SourceStr: string; codePage: Word): string;
var
    wStr: string;
begin
    try
        wStr := ClearBom(SourceStr, Utf16Bom);
        Result := WideStringToString(wStr, codePage);
    finally
        SetLength(wStr, 0);
    end;
end;

function Utf16ToUtf8(const SourceStr: string): string;
begin
    Result := ConvertUtf16(SourceStr, CP_UTF8);
end;

function JSONSaveFile(const Filename: String; s: String): Boolean;
var
    fs: TFileStream;
    i, len : Integer;
begin
    i := 1;
    len := Length(s)
    If len > 0 then
    begin
        try
            try
                fs := TFileStream.Create(Filename, fmCreate or f mShareExclusive);
                fs.Seek(0, 0);
                while (s[i] <> #0) and (i < len) do
                begin
                    fs.WriteBuffer(s[i],CharLength(s,i));
                    i := i + CharLength(s,i);
                end;
                Result := True;
            except
                Log('EXCEPTION RAISED in JSONSaveFile: '+Filename);
            end;
        finally
            fs.free;
        end;
    end;
end;

I get only a ANSI coded files. Build-in functions like SaveStringsToUTF8File() function will not work, cause the BOM is added by default.

Or is the a better way to save/create this text file with SaveStringToFile()?

How to solve it?

Answer 1

Use the WideCharToMultiByte function to convert the string to UTF-8 and just save it:

const
  CP_UTF8 = 65001;

function WideCharToMultiByte(CodePage: UINT; dwFlags: DWORD;
  lpWideCharStr: string; cchWideChar: Integer; lpMultiByteStr: AnsiString;
  cchMultiByte: Integer; lpDefaultCharFake: Integer;
  lpUsedDefaultCharFake: Integer): Integer;
  external '[email protected] stdcall';

function GetStringAsUtf8(S: string): AnsiString;
var
  Len: Integer;
begin
  Len := WideCharToMultiByte(CP_UTF8, 0, S, Length(S), Result, 0, 0, 0);
  SetLength(Result, Len);
  WideCharToMultiByte(CP_UTF8, 0, S, Length(S), Result, Len, 0, 0);
end;

function SaveStringToUTF8FileWithoutBOM(FileName: string; S: string): Boolean;
var
  Utf8: AnsiString;
begin
  Utf8 := GetStringAsUtf8(S);
  Result := SaveStringToFile(FileName, Utf8, False);
end;

You have to use Unicode version of Inno Setup (the only version as of Inno Setup 6).

---

See also:

• LoadStringFromFileInCP and LoadStringsFromFileInCP functions in:
  Inno Setup - Convert array of string to Unicode and back to ANSI
• Inno Setup replace a string in a UTF-8 file without BOM