Re: MinGW: 'QUnixTerminalImpl' has not been declared

[Daniel J Sebald]

On 09/20/2012 03:37 PM, Daniel J Sebald wrote:
> On 09/20/2012 03:24 PM, Philip Nienhuis wrote:
> > Daniel J Sebald wrote:
> > > On 09/20/2012 03:07 PM, Philip Nienhuis wrote:
[snip]
> > > > FYI, I did hg clone ....... on my Linux laptop, ran ./autogen.sh there,
> > > > tar.gz'ed it, transplanted and untar/unzipped it into my MinGW dev
> > > > directory.
> > > >
> > > > Thanks,
> > > >
> > > > Philip
> > >
> > > Oh, I don't think that will work Philip. The reason is that
> > >
> > > moc-QTerminal.cc
> > > moc-QTerminalInterface.cc
> > >
> > > are generated files with contents dependent upon the system.
> >
> > That should happen during ./configure, not during .autogen.sh, or have I
> > got that wrong?
>
> No, that's right. But I'm wondering if that makes a difference or not.
> The autogen.sh/bootstrap may be setting some definition as though the
> configure is intended for WIN32, or something. Let me think about this
> one. I see where the MOC/Q_OS_WIN32 definitions are done.

The definitions and such in libgui/Makefile.in end up looking reasonable 
(they are dependent upon WIN32_TERMINAL_TRUE and WIN32_TERMINAL_FALSE). 
 Similarly, my ./configure file tests look proper (they are dependent 
upon win32_terminal from AC:

 if test $win32_terminal = yes; then
  WIN32_TERMINAL_TRUE=
  WIN32_TERMINAL_FALSE='#'
else
  WIN32_TERMINAL_TRUE='#'
  WIN32_TERMINAL_FALSE=
fi
)

To narrow down what we are looking for, could you please check the 
result of your ./configure inside the config.status file for something 
like the following?

S["WIN32_TERMINAL_FALSE"]=""
S["WIN32_TERMINAL_TRUE"]="#"

What that means is a comment character (#) will be put in front of a 
Windows-related definitions.

On your system the result should be the opposite.  Let's confirm that 
first.  (And if you find in your config.status these two lines aren't 
opposite of above, try switching the "#" around and then run "make" just 
to see if compilation finishes.)

Dan