Re: Glib::Object::Subclass and properties

["Martin Schlemmer" <Martin Schlemmer nwu ac za>]

> > > On 2011/11/20 at 01:34 AM, Torsten Schoenfeld <kaffeetisch gmx de> wrote:
> On 17.11.2011 12:59, Martin Schlemmer wrote:
> > I have a Glib::Object::Subclass derived class, with a
> > Glib::ParamSpec->boolean() property named 'symbols' that has a
> > default value of TRUE. If I try to access it via $self->{symbols}, it
> > does not exist. If I however do a $self->set( symbols =>
> > $self->get('symbols') ) it does exist ...
> >
> > Feature or bug?
>
> Well, I'd say it's an implementation detail that leaked.
>
> If you don't overwrite them, G:O:Subclass gives you default 
> implementations of the GET_PROPERTY and SET_PROPERTY vfuncs that 
> retrieve/store property values inside the object hash.  (See 'perldoc 
> Glib::Object::Subclass' for more details.)  But if you haven't called 
> set() yet, then nothing will be stored in the object hash.  You could 
> argue that G:O:Subclass should make sure that the value in the hash is 
> initially set to the default value if there is one.  But that doesn't 
> sound too robust.  So I think you, as the class implementor, should 
> handle initializing these values yourself.
>
> Or you could just always go through get()/set() instead of direct hash 
> access.

I do not mind either way, just wanted to verify that it was due to some reason, and not maybe oversight.

On a bit different topic - I guess that if you make something readonly, the correct way is to use hash 
access? The reason I am asking, is because I am not sure if you do that, whether you should signal a notify 
on the property?


Thanks,

Martin



Vrywaringsklousule / Disclaimer:  http://www.nwu.ac.za/it/gov-man/disclaimer.html