Angus Leeming <[EMAIL PROTECTED]> writes:
| On Tuesday 02 July 2002 8:26 am, Lars Gullik Bjønnes wrote:
>> Angus Leeming <[EMAIL PROTECTED]> writes:
>> | On Monday 01 July 2002 6:11 pm, Andre Poenitz wrote:
>> >> On Mon, Jul 01, 2002 at 05:24:13PM +0100, Angus Leeming wrote:
>> >> > > [If do, cant' this be made tmore transparent somehow?]
>> >> >
>> >> > Wrap you signal inside LyXSignal ?
>> >> >
>> >> > template<typename Signal>
>> >> > class LyXSignal : public Signal{
>> >> > LyXSignal(LyXSignal const &) {}
>> >> > };
>> >>
>> >> Do we have that already?
>> >>
>> >> Andre'
>> |
>> | No, I just made it up. Would it not do the trick though?
>>
>> But you loost all the semantics for this particular signal right?
>
| What do you lose? (Genuinely mystified.) This class derives from
| boost::signalX, so we retain all that beast's semantics. All we do is state
| explicitly that copy-construction does not copy at all but generates a fresh,
| empty instance.
Exactly... and that should be stated explictly. Sometimes you really
want to rebind the signal upon copy at other times not. We should hide
these facts inside some mystical LyXSignal.
>> I'd prefere the explicit copy constructor.
still.
--
Lgb
