Markus Armbruster <arm...@redhat.com> writes: > Marc-André Lureau <marcandre.lur...@redhat.com> writes: > >> Following a discussion on the mailing list: > > If a reader of this commit message could profit from reading the > discussion, refer to it by URL and/or Message-Id. If not, don't mention > it. > >> while it may be convenient >> to accept NULL value in qobject_unref() (for similar reasons as free() >> accepts NULL), it is a probably a bad idea to accept NULL argument in >> qobject_ref(). > > Yes? What's the patch doing about it? Peeking ahead: it outlaws it. > So say that. > >> Furthermore, > > Commit message smell: two things in one patch. Worth separating them? > >> it is convenient and more clear to call qobject_ref() at >> the time when the reference is associated with a variable, or >> argument. For this reason, make qobject_ref() return the same pointer >> as given. > > Not 100% clear whether the patch merely makes the "convenient and more > clear" way possible, or reality. Peeking ahead: it's the latter. So > say that. > > How did you find the places to change?
Ah, you answered this during review of v3: manual review of qobject_ref() uses. > Do you think you got them all? I guess that means you do. >> Signed-off-by: Marc-André Lureau <marcandre.lur...@redhat.com>