On Fri, Jan 22, 2021 at 11:05:22AM -0500, Laine Stump wrote: > (Thought I sent this 7 hours ago before I went to sleep, but when I sat down > this morning I saw it was still sitting there as a draft :-/) > > On 1/21/21 1:50 PM, Matt Coleman wrote: > > This series of patches simplifies the code in several ways and makes a > > few changes required by the next round of patches that I'll submit. > > > > Simplifications: > > > > * add a macro to cut down on repetitive SettingData code > > * enable GLib auto-cleanup for hypervObject and several OpenWSMAN types > > > > Changes: > > > > * store the version in hypervPrivate, which will be used to handle > > breaking changes in the Hyper-V API: despite 2012R2 and 2016+ all > > using Hyper-V's "V2" API, backwards-incompatible changes were made in > > 2016 > > * add inheritance to the WMI generator to simplify handling of the > > backwards-incompatible changes introduced in Hyper-V 2016 > > > I've gone through all of these, and just have two questions that affect > multiple patches each (I've replied to the associated patches): > > > 1) There are several cleanup functions in external libraries that in the > past were only called after checking that the pointer was != NULL. g_autoptr > cleanups need to handle being called with NULL as a NOP, and I'm concerned > that these functions may not behave properly in that case. Can you either > verify that it's safe to call them with NULL, or provide a wrapper function > that checks for NULL and use that as the cleanup? The G_DEFINE_AUTOPTR_* macros alrady define wrappers that include a NULL check I believe. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
