Boost Users :
From: Jean-François Brouillet (verec_at_[hidden])
Date: 2005-05-14 17:06:25
Thanks Stefan & Peter, for your replies.
I tried immediately:
struct MyTaskStruct : public LongTaskStruct
And obviously, I got:
terminate called after throwing an instance of 'boost::bad_weak_ptr'
> but you can't do this in a constructor, because the outer shared_ptr<>
> hasn't been created yet.
But the next solution
> Use a static create() function as explained here :
is not good enough. I've got 150+ Java classes to translate, each with
an average of about 2 constructors. That would mean an extra 300
createXYZ, and apart from being ugly, and an insult to the user, is
a maintenance nightmare.
Ideally, I would want an intrusive_ptr (because no matter how they are
created, they can be made to share a single ref-count per body/
but one that offers the shared_ptr guarantee:
shared_ptr<T> can be implicitly converted to shared_ptr<U> whenever
T* can be implicitly converted to U*. In particular, shared_ptr<T>
is implicitly convertible to shared_ptr<T const>, to shared_ptr<U>
where U is an accessible base of T, and to shared_ptr<void>.
What I am after, is a construct as transparent as possible, for which
I can say, in my documentation, "For every struct XXX -- the body --,
there is a corresponding envelope -- xxxx_ptr<XXX> -- that walks and
talks like a XXX *"
So far, shared_ptr had everything covered except that "sharing" thing,
which is an artefact, not a requirement. intrusive_ptr makes this
"sharing" thingie a non issue (as it should be), but fails on the
"polymorphic covariance" quoted above...
In this project, it is acceptable to rely on a common base class,
(anyway I've got to translate Java's "Object" class :-) so I can
stuff it with whatever machinery is required to make garbage
Boost-users list run by williamkempf at hotmail.com, kalb at libertysoft.com, bjorn.karlsson at readsoft.com, gregod at cs.rpi.edu, wekempf at cox.net