D-Bus windows help

[Jesper Dam]

On 11-07-2012 21:39, Alexander Larsson wrote:
>
> GDbus on win32 doesn't support any of the scopes. install-path for
> instance doesn't seem to be a great choice for gdbus as it ties the
> implementation-independent protocol to an implementation (and gdbus is a
> non-libdbus client implementation).
>

isn't autolaunch: *already* tied to an implementation? On Windows it 
uses shared memory to discover the address, which is really an 
implementation detail that both server and client nevertheless have to 
agree on 
(http://dbus.freedesktop.org/doc/dbus-specification.html#meta-transports-autolaunch 
indicates that a mutex and a shared memory section are used, but nothing 
about how these are named, and how a client locates them. So is 
'autolaunch:*install-path' any "more" implementation-dependent than 
'autolaunch:' is?)

Please correct me if I'm missing something, but as far as I can see, 
these scopes are no more or less implementation-independent than 
'autolaunch:' without a scope.

'autolaunch:*install-path' is safer, because Windows generally isn't 
going to have a single OS-level DBus installation. Applications which 
want to use DBus are going to have to bring their own, which means 
multiple versions may end up having to coexist, and this scope ensures that.

However, 'autolaunch:' without a scope would still work in most cases, 
because DBus isn't that widely used on Windows currently. It is unlikely 
that multiple incompatible versions are going to be installed and used 
simultaneously.

So I would prefer using the install-path scope, but I see no fundamental 
problems in using 'autolaunch:' if necessary. It's much better than the 
existing default, and it'll work out of the box *most of the time*.

If GDbus isn't going to support these scopes, then 'autolaunch:' is the 
only working option, and then I guess that's the end of the story.

Alexander, your thoughts?

@Simon: the patches look good to me. I say apply them, once we know 
which default to use.


/ Jesper