Peter Kümmel syntheticpp at gmx.net
Sun Sep 17 09:26:39 PDT 2006

Havoc Pennington wrote:
> Peter Kümmel wrote:
>> Replacing _dbus_connect_unix_socket with a pure
>> windows-specific version seems to enforce to
>> change the whole DBusTransport infrastructure:
>> - transport.h is full of _unix functions
>> - transport.c uses transport-unix.h and
>>   dbus-transport-protected.h,  both unix
>>   specific files
>> So I don't see a way to implement a clean windows
>> version without touching all the code of DBusTransport,
>> the current design is too unix centric.
>> To have a clean windows port we must have no _unix
>> functions in all non unix header/soucre files, but
>> this is maybe an item for D-Bus 2.0. :)
> Whine whine whine. ;-) I just did this for you, it took a couple hours.
> I did not quite finish; dbus_transport_open needs a similar change to
> dbus_server_listen, will need to add DBusTransportOpenResult like
> DBusServerListenResult, and also _dbus_transport_open_platform_specific
> living in dbus-transport-unix.c.
> _dbus_server_set_bad_address should probably become
> _dbus_address_set_bad() and shared between the server and transport code.
> This remaining change should take about 1/2 hour but is needed for
> Thiago's patch, since he would hook in to the unix
> _dbus_transport_open_platform_specific.
> Havoc

Thanks for the api changes; it's much faster when you change it directly
than waiting for the patches of a whining windows guy. ;)

I've updated our svn repository, but now we get an regression in the bus test.
Have you changed the error handling?


More information about the dbus mailing list