Re: glib+orbit

[Owen Taylor <otaylor redhat com>]

bob@cs.csoft.net writes:

> On 9 Jan 2000, Owen Taylor wrote:
> 
> > 
> > bob@cs.csoft.net writes:
> > 
> > > I am trying to get a g_timeout to work with orbit. I have looked through
> > > the gconf code, and kinda figured out to do it, but it seems very genaric.
> > > why is this code not in orbit itself?
> > > 
> > > also, a GMainLoop question.
> > > in gconfd, the global var main_loops is defined. Why does it exist? I
> > > spent the last few hours trying to figure out how GMainLoop works (there
> > > is like no dox on it), and it looks like there can only ever be one main
> > > loop, so why does main_loops exist? (or was this code copied out of gtk,
> > > the only other place I could find it?)
> > 
> >   http://developer.gnome.org/doc/API/glib/glib-the-main-event-loop.html
> 
> I read that already. desnt say anything but how to create a new main loop.
> (doesnt mention that there is only one ever really created). 

To quote:

   The GMainLoop data type represents a main event loop. A GMainLoop is
   created with g_main_new(). After adding the initial event sources,
   g_main_run() is called. This continuously checks for new events from
   each of the event sources and dispatches them. Finally, the processing
   of an event from one of the sources leads to a call to g_main_quit()
   to exit the main loop, and g_main_run() returns.

   It is possible to create new instances of GMainLoop recursively. This
   is often used in GTK+ applications when showing modal dialog
   boxes. However, all event sources are global; they are not tied to a
   particular GMainLoop.

I'm not really sure isn't there that you'd want to see. But if you have
suggestions on how the text could be improved we'd appreciate hearing
about it. :-)

(Actually, the text is a little unclear, in that the first part sounds
like event sources are associated with a particular GMainLoop, and then
that is only clarified in the second paragraph.)

> I had to go
> through the glib cvs tree with lxr (great little tool btw) and trace half
> a duzzen functions to find out how it worked. wasnt pritty. :)
> 
> Thanks though. I got it to work like I wanted about 10 min ago. :)

Regards,
                                        Owen