Dear Frederick,

 

in my humble opinion I would reorder the first half page.

Instead of starting by chronological order, how you got the idea, and which function you wanted to write in the beginning, I would start with describing the typical class hierarchy of wxWidget controls and how it would look like handling those kinds of hierarchy. That would make it easier to follow your function Red shown as directly next topic and the proposal would be a bit more general (for users of wxWidget) than for a use case seemingly having occured only once (for you).

 

Another point is the member resolution. I would suggest to make members defined twice an error (if the base class is not specified as in the GodBolt code, but automatically), similar to the error with ambiguous members, when inheriting from both base classes (the wxControlWithTextEntry, if it would exist in between the inheritance hierarchy).

 

Best,

Sebastian
 

-----Ursprüngliche Nachricht-----
Von: Frederick Virchanza Gotham via Std-Proposals <std-proposals@lists.isocpp.org>
Gesendet: So 27.11.2022 14:22
Betreff: Re: [std-proposals] Chimeric Pointer
Anlage: chimeric_pointer2.pdf
An: std-proposals@lists.isocpp.org;
CC: Frederick Virchanza Gotham <cauldwell.thomas@gmail.com>;
I have updated the paper, it's now Draft No. 2. It's attached to this
email and you can also download it from:

   http://www.virjacode.com/download/chimeric_pointer2.pdf

Changes:
* Thorough explanation of how member objects/function are looked up in
the base classes
* Discussion of the need for a type that alleviates the need for a
template function
* Discussion of Martin Darius's implementation of chimeric_ptr (along
with a link to GoldBolt)
--
Std-Proposals mailing list
Std-Proposals@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals