On Wed, Nov 19, 2014 at 02:42:51PM -0700, Eric Blake wrote:
On 11/17/2014 05:36 PM, Eric Blake wrote:
Use of an 'int' to represent a 'bool' value is confusing. Just because dbus made the mistake of cementing their 4-byte wire format of dbus_bool_t into their API doesn't mean we have to repeat the mistake. With a little bit of finesse, we can guarantee that we provide a large-enough value to the DBus code, while still copying only the relevant one-byte bool to the client code, and isolate the rest of our code base from the DBus stupidity.
[Have I ever mentioned that the assymetry between type promotions of values passed through var-args on setting, vs. no promotions when passing pointers through var-args for getting, makes life awkward, not just for DBus interactions, but also for printf vs. scanf? Then again, using scanf is often the wrong thing to do...]
* src/util/virdbus.c (GET_NEXT_VAL): Add parameter. (virDBusMessageIterDecode): Adjust all clients. * src/util/virpolkit.c (virPolkitCheckAuth): Use nicer type. * tests/virdbustest.c (testMessageSimple, testMessageStruct): Test new behavior.
-# define GET_NEXT_VAL(dbustype, vargtype, fmt) \ +# define GET_NEXT_VAL(dbustype, member, vargtype, fmt) \ do { \ dbustype *x; \ + DBusBasicValue v; \
Blech. DBusBasicValue wasn't defined in the older dbus-types.h from 1.1.2 as shipped on RHEL 5, causing a compilation failure there. I'm working on a fix...
Its just a dumb union, so we can either provide the definition ourselves, or just define our of virDBusValue union and ignore DBusBasicValue Regards, Daniel -- |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|