[TransWarp] On the requirements for "Assembly Events"
Phillip J. Eby
pje at telecommunity.com
Mon Apr 28 19:44:27 EDT 2003
At 04:40 AM 4/29/03 +0200, Ulrich Eck wrote:
>A formalized way of doing this provided by the framework will help with
>interoperability of components as well. I think the way you described
>is pretty straigth forward. to be able to avoid memory-leaks (e.g. wxPython)
>i need to destroy components as well. would it make sense to extend
>the Assembly events with destroy events as well (this would target to
>LifeTimeEvents which i'ld find very useful, if implemented in a good manner) ?
I don't have a good understanding of what would be desired for such
"destroy" events, probably because I haven't done something that needed
this yet. Perhaps you could elaborate?
I don't think the assembly events will help you with destruction,
though. The implementation is very specific to the requirements I
described; they take every possible advantage of the specific need, rather
than being some kind of generic event mechanism.
More information about the PEAK