Re: [Boost-users] [signals] thread-safe post-disconnect() cleanup

From: Peter Dimov (pdimov_at_[hidden])
Date: 2007-02-25 12:07:35

Peter Dimov wrote:
> Frank Mori Hess wrote:
>
>> Couldn't tracked objects could be copied from one slot to another by
>> overloading slot::track to accept a slot as an argument?  Then
>> binding a slot to another slot with a different signature would
>> might look like
>>
>> signal.connect(signal_type::slot_type(slot, _2, _1).track(slot));
>>
>> It's a bit murky whether track(slot) would only cause the tracked
>> objects of slot to get tracked, or also track the slot itself (or
>> its pimpl).  I suppose it would depend on whether copies of slots
>> are deep or share the same SlotFunction.
>
> You don't need tracking for the above example; bind( slot, _2, _1 )
> stores a copy of slot.

Now that I think about it, it can be implemented both ways. A slot that 
requires an explicit lock() before operator() would still need the .track. A 
slot that throws bad_weak_ptr from operator() would not. This brings us back 
to the problem of autodisconnects only happening on calls. :-/ It doesn't 
have to be either-or, though.