ANSI to UTF-8 or #228 ---> 'ä'

[Bernd82]

I would like to display old Delphi ANSI strings (Windows) in a TMemo under Lazarus. The strings contain German and French special characters:

Code: [Select]

    VAR s:String;
    BEGIN
      s:=#228;
      Memo.Lines.Add(AnsiToUtf8(s));
    END;This works with Lazarus under/for Windows. Will this also work if I compiled it under/for Linux? Thanks for any hint!

Regards Bernd

[Blaazen]

No, it does not work on Linux. There appears "?" in Memo.

But you can do this:

Code: [Select]

s:='ä';
Memo1.Lines.Add(s); instead of using number, Lazarus is fully UTF8.

[typo]

Code: [Select]

ShowMessage(#195 + #164      // tiny a
              + ' - ' +
              #195  + #132); // capital A     


[Bernd82]

Thanks for your answers. So is there a chance to convert a string containing one-byte ANSI characters to a normal Lazarus string (UTF-8)?

[Michl]

There is a crosspost (the second in two days - thats realy not fine, as I had explained in German Lazarusforum):
http://www.lazarusforum.de/viewtopic.php?f=10&t=8663#p76727

[typo]

Code: [Select]

ShowMessage(#$C3#$A4
             + ' - ' +
             'A'#$CC#$84);


[wp]

This works under Linux:

Code: [Select]

uses
  lconvencoding;
...
var
  s: String;
begin
  s := CP1250ToUTF8(#228);
  // or: s := ISO_8859_1ToUTF8(#228);
  ShowMessage(s);
end;

[typo]

http://www.utf8-chartable.de/

[Bernd82]

Hello wp,

ah, unit lConvEncoding works with code pages. I understand. I estimate the CP1252 might be better in my case. It equals the ISO 8859-1 or Latin-1. I just fond this on Wikipedia:

http://en.wikipedia.org/wiki/Windows-1252

Thanks a lot for your competent information!

Best regards, Bernd

[Bernd82]

Hello typo,

thanks for your answers. Yes that is the right character list for my conversions.

Regards, Bernd

[Bernd82]

Hello Blaazen, thanks for your testing! Does this here work properly?

Code: [Select]

    USES LConvEncoding;
    VAR s:String;
    BEGIN
      s:=#228;
      Memo.Lines.Add(CP1252ToUTF8(s));
      Memo.Lines.Add(ISO_8859_1ToUTF8(s));
    END;Thanks in advance for your help. Unfortunately I have no Linux here at the moment...

Regards Bernd

[Bernd82]

I had a closer look under Windows at all converted characters and found that

CP1252ToUTF8( )

seems to be the better alternative.

ISO_8859_1ToUTF8( )

has an undefined gap from character #128 to #159. For example the € sign is missing then...

Bernd

[marcov]

In threads like this it is important to remember that "ansi" in freepascal/Delphi jargon means "default one byte codepage", not necessarily ansi/ascii.

The windows codepage for latin with euro is cp28591:

28591   iso-8859-1   ISO 8859-1 Latin 1; Western European (ISO)

and following:

28592   iso-8859-2   ISO 8859-2 Central European; Central European (ISO)

[Bernd82]

Hello marcov,

cp28591 like ISO_8859_1 has that gap between #128 and #159 or #$80 and #$9F! Look here:

https://msdn.microsoft.com/de-de/goglobal/cc305167.aspx

So the '€' and others are missing.

Regards, Bernd