Re: libglademm proposed change: arbitrary constructors for get_widget_derived
Murray Cumming <[email protected]> Wed, 04 May 2005 23:05:44 +0200
| Newsgroups | gmane.comp.gnome.gnomemm |
|---|---|
| Message-ID | <[email protected]> |
On Wed, 2005-05-04 at 22:17 +0200, Marek Materzok wrote: > Dnia 04-05-2005, =C5=9Bro o godzinie 18:17 +0200, Murray Cumming napisa= =C5=82(a): >=20 > > I think it's slightly useful. It allows you to use a derived widget b= ut > > do things to that widget's children outside of the class, instead of > > inside. Personally, I think the derived widget should not want you to > > mess around with its (normally protected) child widgets directly like > > this. Can you persuade me that it's useful? >=20 > I've used it to create a pattern, in which all access to children was > inside the class, but I could have additional parameters to the > constructor of that class.=20 >=20 > > If you revise the patch, please put it in bugzilla: > > http://www.gtkmm.org/bugs.shtml#CreatingPatches >=20 > Added, bug #303044. >=20 > > This idea might also be interesting. I _still_ have not looked at it > > properly: > > http://bugzilla.gnome.org/show_bug.cgi?id=3D134161 >=20 > I'm shocked. I'm doing nearly the same thing in my code, but using the > changes to glademm from my patch. This idea seems cleaner than mine, > because in my version I use a static factory function, and here just a > constructor is needed. I'll test this idea. Great. If you can make an actual patch then it will make it quicker for me to review. --=20 Murray Cumming [email protected] www.murrayc.com www.openismus.com