dev-cpp-users

RE: [Dev-C++] unicode

[Jaap de W. <wo...@Ne...>]

at this moment there is no std::string<wchar_t> support in the Mingw
compiler.
This is a known problem, it has to do with missing functions in MSVCRT.DLL.

Jaap

> -----Original Message-----
> From: Torsten Curdt [mailto:tc...@df...]
> Sent: Monday, February 17, 2003 03:41 PM
> To: dev...@li...
> Subject: [Dev-C++] unicode
> 
> 
> I just switched my application to unicode by specifying the
> "-D UNICODE" and "-D _UNICODE" compiler options.
> 
> But the STL string class refused to accept that. Though
> it uses the "charT" type.
> 
> Any pointers how to switch it to unicode?
> --
> Torsten
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dev-cpp-users mailing list
> Dev...@li...
> TO UNSUBSCRIBE: http://www.noicys.cjb.net/devcpp/ub.htm
> https://lists.sourceforge.net/lists/listinfo/dev-cpp-users
> 

Re: [Dev-C++] unicode

[Torsten C. <tc...@df...>]

> Torsten,
> 
> This is off list, but would/can you share your changes.  I am interested in this.

Sure! I wish I could send a patch but I have no CVS checkout available.
But here are the changes "by hand" ;)


{INSTALLDIR}/include/g++-3/std/bastring.h line 338:
     ...
     { if (length () == 0) return TEXT(""); terminate (); return data ();
     ...                          ^^^^^^^^^

{INSTALLDIR}/include/g++-3/string:
...
extern "C++" {
typedef basic_string <char> string;
typedef basic_string <TCHAR> tstring;
typedef basic_string <WCHAR> wstring;
} // extern "C++"
....

I also noticed that the strstream.h needs to be changed to use
TCHAR/charT instead of char to work properly. But haven't done
that so far.

Would be cool if this could be fixed in the next Dev-Cpp release!
(I cc this mail to the dev-cpp-users group)

cheers
--
Torsten

Re: [Dev-C++] unicode

[Torsten C. <tc...@df...>]

Jaap de Wolff wrote:
> at this moment there is no std::string<wchar_t> support in the Mingw
> compiler.
> This is a known problem, it has to do with missing functions in MSVCRT.DLL.

Ah... thanks for the feedback!

But I did some changes in the include files and now have
unicode working :)

Do you know what functions are supposed to be missing?
At least the basic string template works fine for me now.
--
Torsten

Re: [Dev-C++] unicode

[Ioannis V. <no...@ho...>]

a-------Original Message-------=0D
 =0D
From: Torsten Curdt=0D
Date: =D4=F1=DF=F4=E7, 18 =D6=E5=E2=F1=EF=F5=DC=F1=E9=EF=F2 2003 1:02:29 =
=EC=EC=0D
To: dev...@li...=0D
Subject: Re: [Dev-C++] unicode=0D
 =0D
Jaap de Wolff wrote:=0D
> at this moment there is no std::string<wchar_t> support in the Mingw=0D
> compiler.=0D
> This is a known problem, it has to do with missing functions in MSVCRT.=
DLL
=0D
=0D
Ah... thanks for the feedback!=0D
=0D
But I did some changes in the include files and now have=0D
unicode working :)=0D
=0D
Do you know what functions are supposed to be missing?=0D
At least the basic string template works fine for me now.=0D
=0D
=0D
=3D=3D> At first i must say that there is no such thing as string<wchar_t=
> since
string is=0D
=0D
typedef basic_string<char> string;=0D
=0D
In some implementation char can be the same as wchar_t (switch or not it
doesn't matter). But since you can use wchar_t explicitly why use switche=
s?
Keep in mind that the type wstring exists!=0D
=0D
And as the standard nicely says:=0D
=0D
"[lib.string.classes] 21.2 String classes=0D
1 The header <string> defines a basic string class template and its trait=
s
that can handle all char-like (clause 21) template arguments with several
function signatures for manipulating varying length sequences of char-lik=
e
objects.=0D
2 The header <string> also defines two specific template classes string a=
nd
wstring and their special traits."=0D
=0D
=0D
=0D
Ioannis Vranos=0D
=0D
* Programming pages: http://www.noicys.freeurl.com=0D
* Alternative URL: http://run.to/noicys=20

Re: [Dev-C++] unicode

[Torsten C. <tc...@df...>]

<snip/>

> ==> At first i must say that there is no such thing as string<wchar_t> 
> since string is
>  
> typedef basic_string<char> string;
>  
> In some implementation char can be the same as wchar_t (switch or not it 
> doesn't matter). But since you can use wchar_t explicitly why use 
> switches? Keep in mind that the type wstring exists!

Well, IIRC it was commented out in my copy (strange)

Anyway - the problem is that I need *ALL* types in my application.

So now I have:

string - ANSI
tstring - ANSI/UNICODE depending on the switch
wstring - UNICODE
--
Torsten