> On the other hand, QSharedData's ref member is an implementation detail (the > QRefCount class is even private API), > and the semantics of ref/deref on it are undocumnted, so *that* could > bechanged right now.
It is good to mention QRefCount aka QtPrivate::RefCount except it is not used in QSharedData (QSharedData uses a "QAtomicInt ref") All implicitly shared class should rely on a single optimized QtPrivate::RefCount on the line of https://codereview.qt-project.org/c/qt/qtbase/+/66118 Philippe On Wed, 7 Aug 2019 22:26:11 +0200 Giuseppe D'Angelo via Development <development@qt-project.org> wrote: > Il 07/08/19 20:00, Mutz, Marc via Development ha scritto: > > This was discussed in > > https://codereview.qt-project.org/c/qt/qtbase/+/66118. See, in > > particular, Olivier's comment. > > > > TL;DR: ref() is documented to be ordered, so cannot be changed. > > On the other hand, QSharedData's ref member is an implementation detail (the > QRefCount class is even private API), and the semantics of ref/deref on it > are undocumnted, so *that* could be changed right now. Although I'd still > feel very uncomfortable at doing so. > > My 2 c, > -- Giuseppe D'Angelo | giuseppe.dang...@kdab.com | Senior Software Engineer > KDAB (France) S.A.S., a KDAB Group company > Tel. France +33 (0)4 90 84 08 53, http://www.kdab.com > KDAB - The Qt, C++ and OpenGL Experts > _______________________________________________ Development mailing list Development@qt-project.org https://lists.qt-project.org/listinfo/development
