Re: [gnet-dev] GNet 2 for Windows?

[Martyn Russell <ginxd btopenworld com>]

On Thu, 2003-05-29 at 04:00, David A. Helder wrote:
> On Wed, 2003-05-28 at 22:27, Andrewlanoix aol com wrote:
> > Charles,
> > 
> > Parts of GNet 2.0 for Win32 is still under development however that does not mean you can not upgrade to 2.0 if necessary. If you do not need any of the new functionality you can use the code from CVS though you will have to make some simple changes to the makefiles. I am not aware of any regressions but have not done any testing besides running some (but not all) of the examples and test code. If you run into any issues I will fix any regressions before getting the new APIs working. 
> >  
> > I am in the process of making major changes to the Glib for Win32 specific iochannel code to support the new GNet’s new “Conn” API. The code has been written and I am trying to track down an access violation when you have multiple clients connecting to a “party line” server then have one disconnect.
> > 
> 
> DanielKO has been helping out with the Win32 compile on #gnet.  The
> latest 2.0 release should at least build on Win32.  Like Andy said, IPv6
> won't work yet.  Resolving a name to multiple addresses won't work
> either, but that doesn't affect most applications upgrading from 1.1.
> 
> 
> > On second thought, since thing are taking longer than expected, would it be helpful to you and others if I did switched gears and did a some testing of the current 2.0 code and then created a snapshot? The snapshot would not have IPv6 or “Conn” API support but may allow you to upgrade to 2.0.
> 
> It sounds like that'd be useful.  DanielKO said that there are bugs
> related destroying the window used for receiving messages.  I imagine
> once that's fixed the non-asynchronous part of GNet will work as
> before.  The Conn part will work once asynchronous connects/reads/writes
> work.  Conn just builds ontop of GLib iochannels.

Do these changes to GLib include fixing problems with async connections
that was reported a while back?  

Plus will this be fixed for the _PROPER_ release of 2.x of gnet? Or has
it been fixed already? Andrew, you were following this up weren't  you?

Regards,
Martyn