Hello,

I'm trying to write "Atenção" in terminal, but it writes "At├º├úo" instead of "Atenção" (please see http://imagebin.org/217387 ). :(

On Linux it's OK.

My test code is (or zip in attached):

------------------------------------------------------------
program test;
{$mode objfpc}{$H+}
{$IFDEF MSWINDOWS}
uses
  windows;
{$ENDIF}
const
  C = 'Atenção';
begin
{$IFDEF MSWINDOWS}
  SetConsoleOutputCP(CP_UTF8);
  SetTextCodePage(Output, CP_UTF8);
{$ENDIF}
  WriteLn(C);
end.
------------------------------------------------------------

Lazarus 1.1 r37649 FPC 2.7.1 i386-win32-win32/win64 | Windows 7 - 64

Thanks.

silvioprog wrote

Hello, I'm trying to write "Atenção" in terminal, but it writes "At├º├úo" instead of "Atenção" (please see http://imagebin.org/217387 ). :(

open a cmd window and type "chcp"... if that still exists in win7, it should tell you the codepage being used... looking at your example, it appears that something is translating what you type to utf-8 and that your terminal cannot render utf-8... i fight the same battle in another world with a huge amount of CP437 oriented software that will never be updated...

i will try your code on my boxen to see what it gives... now to figure out how to generate those characters with my US keyboard... i /think/ ALT-135 is the "lowercase c with a tail" 'ç' and the other one is the "lowercase a with a wavy line over it" but i can't find it :? maybe i can paste it from this message into that window?

22.06.12 5:39, silvioprog пишет:

Hello, I'm trying to write "Atenção" in terminal, but it writes "At├º├úo" instead of "Atenção" (please see http://imagebin.org/217387 ). :(

...

program test; {$mode objfpc}{$H+}

try {$codepage utf8} or add an utf8 BOM because your constant is encoded in utf8.

Best regards,
Paul Ishenin

i said i'd try to test...

Options changed, recompiling clean with -B
Hint: Start of reading config file C:\freepascal\fpc\2.6.0\bin\i386-win32\fpc.cfg Hint: End of reading config file C:\freepascal\fpc\2.6.0\bin\i386-win32\fpc.cfg Free Pascal Compiler version 2.6.0 [2012/05/21] for i386 Copyright (c) 1993-2011 by Florian Klaempfl and others Target OS: Win32 for i386
Compiling test.lpr
test.lpr(15,18) Error: Identifier not found "SetTextCodePage" test.lpr(20) Fatal: There were 1 errors compiling module, stopping

lazarus 1.1 r37695 fpc 2.6.0 i386-win32 | w2k

maybe i'll try on my vista box? i need to updatelaz on it anyway ;)

ha! i should have caught that when i was playing with it in fp because it did show as 2character characters... i just didn't recognize for such... my eyes are crossed from too much codepage fighting in my other world :?

2012/6/21 Paul Ishenin :

Oh yeah! Worked fine with this define: http://imagebin.org/217405.

Thank you! ^^

What version of FPC introduced this directive? I've got a couple of places I could do with documenting it as a prerequisite.

2.6.0 has it already, AFAIK at least also 2.4.x.

Regards,
sven

Sven Barth wrote on Fri, 22 Jun 2012:

Support for this feature predates svn revision 1, so it was probably already in 2.0.x.

Jonas

already in 2.0.x.

Even better then :)

Regards,
Sven

Thanks Jonas, noted. So in practical terms it predates the point at which the LCL moved from favouring GTK to GTK2.

I've been using non-ASCII characters for a year or so when displaying e.g. Cyrillic RTTY output, but in general have set up the translation tables using numeric constants.

In our previous episode, silvioprog said:

Windows has two sets of console routines. One set that is universal over win9x and WinNT, one that is WinNT only.

FPC uses the universal ones (still), the above code is WinNT only.

Another reason to deprecate 9x? ;)

Regards,
Sven

In our previous episode, Sven Barth said:

Unicode? Actually, it was the main one.

No, the point that we currently use the universal APIs while there is a NT only set of APIs which might solve some things.

Regards,
Sven