|
Boost : |
From: Frank Mori Hess (fmhess_at_[hidden])
Date: 2008-04-13 11:08:53
On Saturday 12 April 2008 14:37, Peter Dimov wrote:
> I leave the decision (and the actual revert, should you elect to
> proceed) to you. It might be better to wait for a test cycle first. I'll
> also appreciate if you try sp_accept_owner for your existing use cases
> that motivated the enhanced esft base and, if possible, distill their
> essence into a test case that will replace (or complement)
> esft_constructor_test.
Ok. My use case is similar to the classic esft case, where a weak_ptr is
stored when ownership is taken. To be specific, I want to store some
weak_ptrs to this for use in tracking signal/slot connections where a
class connects signals to its member functions during its initialization.
How about making the default implementation of the 3 argument
sp_accept_owner call the 2 argument overload? Then the 3 argument version
can be ignored if the user doesn't care about the deleter.
-- Frank
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk