- From: Murray Cumming <murrayc murrayc com>
- To: bob fis-cal com
- Cc: gtkmm-list gnome org, Ole Laursen <olau hardworking dk>
- Subject: Re: printing
- Date: Thu, 16 Feb 2006 17:52:34 +0100
On Thu, 2006-02-16 at 09:37 -0600, Bob Caryl wrote:
> viewcvs doesn't list libgnomeprintuimm,
It's in gnomemm:
> so I cannot link to that.
> However, the situation to which I allude above is with
> In the file "printdialog.h" in the libgnomeprintuimm library I
> downloaded (version 2.5.2), the run method is listed in the class
> declaration but it is commented out. The method as it appears (also
> commented out) in printdialog.cc is shown below:
> int PrintDialog::run()
This is just some old commented-out code - I've removed it now to
prevent confusion. It is actually implemented if you search further
(generated by a WRAP_METHOD() call in the .hg file). It doesn't do
anything interesting compared to Gtk::Dialog::run() anyway.
> As you can see, the function prototype specifies a return value, but the
> function returns nothing. Because the method wrapper is commented out,
> and since the class is derived from Gtk::Dialog I made the assumption
> that the run method would return something meaningful, but as I said in
> my previous message, it does not seem to do that.
Are you talking about the result of a test, or just what you thought the
source code did?
> My perusal of the
> gedit source code indicated that a callback function was connected to
> the print dialog object to handle the different button presses in the
In general, people seem to prefer to avoid Gtk::Dialog::run() because it
forces the dialog to be modal, which some people don't like, so they
handle the button presses asynchronously like this. But I doubt it's
worth the bother in most cases so I recommend starting simple.
> I am guessing that this is probably what should have been done
> in the wrapped version of the dialog since the buttons displayed by the
> print dialog are "PRINT PREVIEW", "PRINT", and "CANCEL".
> Gnome::Print::UI::PrintDialog::run returns the same value regardless of
> which of those buttons are pushed.
It seems to return different values for each button in the example. The
possible values are:
I will wrap that enum in cvs. I have already added tests for the
response value in the example.
> >We are all waiting for the perfect cairo-based printing API to arrive in
> >a future version of GTK+.
> Is there any forecast about when the new cairo-based printing API will
Maybe in GTK+ 2.10, which should be stable within 6 months:
People are working on it (see gtk-devel-list), and it looks like they
will succeed. It's a difficult job because it needs to be suitable for
Windows and Mac as well as Linux.
murrayc murrayc com
] [Thread Prev