Re: [Gtk-osx-users] Notes on handling Apple events with gtk_osxapplication and PyGTK

[Richard Procter <richard n procter gmail com>]

On 27/08/2010, at 4:14 AM, John Ralls wrote:
> That doesn't sound like a race condition (which usually means that  
> two threads are writing to a shared resource that lacks mutex or  
> semaphore protection).

I guess it comes down to semantics - I use it in the more general  
sense of an error due to (possibly random) differences in the  
relative speed of at least two processes: in this case OS X and the  
application. Semaphores are a means to ensure correctness without  
having to account for their relative speed (and which, incidentally,  
may therefore miss 'optimisation' opportunities as a result: a false  
economy but something held against them early on, as mentioned by  
Dijkstra in answer to a question at the end of this talk: http:// 
userweb.cs.utexas.edu/users/EWD/videos/EWD2.mpg).

> Rather, it sounds like the notification handler isn't turned on  
> until NSApplication::FinshedLaunching is called, and that OSX  
> doesn't queue the notification until then. I'll have to dig into  
> that a bit, but ISTM the event loop is going to be required for  
> properly handling the notification.

Yes. Since writing, I've dug into it a little further and decided the  
most convenient thing from the point of view of the user is for the  
signal to be emitted only after ready() has been called as otherwise  
all sorts of hoop jumping is required. The docs suggest there is a  
way to holdoff the OS X notification on startup, if I'm reading them  
right, which I'm looking into more deeply at the moment.

> I don't understand what you mean by "multiple instances of the  
> process". That's not possible. Do you mean that if you have  
> multiple copies of the program running, only the first-launched  
> gets the notification? That's the way OSX works. It expects there  
> to be only one running instance of a program, which may have open  
> multiple documents.

Probably sloppy terminology on my part. What I meant is running  
multiple instances of the same application bundle, which is possible  
from OS X 10.5 onwards by passing the '-n' argument to open(1).

Just to be clear, I wasn't suggesting any of this was necessarily a  
bug in gtk_osxapplication - only to let others know of what I had  
found in using it.

regards,
Richard.