On Sun, Jun 11, 2006 at 01:47:16AM +0200, Steinar H. Gunderson wrote:
> On Sat, Apr 08, 2006 at 02:30:12PM +0200, Martin Michlmayr wrote:
> > But then I get the following, which I'm not sure is a bug in wxwidgets-2.6
> > or
> > your package (I thought I had a patches wxwidgets installed so either it
> > needs
> > more patching or your package needs another fix):
> >
> > [...]
> >
> > Config.cpp:112: error: ambiguous overload for 'operator+' in
> > 'CConfig::GetGlobalDir()() + "BEID.conf"'
> > [...]
> > Config.cpp:117: error: cannot convert 'const wxChar*' to 'const char*' for
> > argument '2' to 'int NCONF_load(CONF*, const char*, long int*)'
>
> This looks like a Unicode-vs-non-Unicode mismatch; NCONF_load should probably
> take a const wxChar* (which probably maps to wchar_t, as I'd guess you
> compile against the Unicode build of wxWidgets) instead of const char *, and
> then move from there. Similiarily CONFFILE_BEID should probably read
> _T("foo") instead of "foo", or line 112 should include the _T() macro.
>
> Optionally, the entire thing could be compiled against the non-Unicode build
> of wxWidgets... (Or perhaps upstream solved these bugs as well? I see the bug
> was tagged fixed-upstream, but I'm not sure if this includes the wxWidgets
> bugs.)
The new version compiles just fine with G++ 4.*. I've got a package
mostly ready, I'm only waiting for upstream to release the official
version before I upload the Debian package (as they've asked me);
hopefully that won't take too long anymore. This is a government body
though :-)
I haven't checked whether the new version has unicode-related warning or
errors, but I do know that the unicode corruption which occurs in the
previous version of the eidviewer seems gone in the new version.
Whether this is because they fixed the code, or because QT (which
they've decided to include...) is more flexible regarding unicode, I
don't know.
--
Fun will now commence
-- Seven Of Nine, "Ashes to Ashes", stardate 53679.4
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
