ooc-dev team mailing list archive

[Scott Olson <scott@xxxxxxxxxxxxxxx>]

It was my understanding that nddrylliog's plan was to generate 
getter/setter methods in the generated C for all properties, including the 
ones that look just like plain old attributes. This is for ease of C 
calling ooc code. The C generated from the ooc code will still use the 
fastest way of accessing the variable (direct-access if there is no custom 
accessor).

As for the syntax, we could do:

age: Int {
    set: func (.age) {
        // ...
    }
    get: func -> Int {
        // ...
    }
}

But personally, I prefer the short form.

On Wed, 31 Mar 2010, Friedrich Weber wrote:

> Hi all,
>
> looks good to me, too, thanks for this!
>
> Some things:
>
> > name: String {
> >     set
> >     get
> > }
> >
> > // The above example is exactly equivalent to this:
> > name: String
>
> IMHO the variable might be equivalent on the ooc level, but not on the C
> level: The property would create a setter and a getter method, which
> would create ..._setName and ..._getName functions on the C level. This
> means: These two examples generate different C APIs.
>
> Also, I just want to note that this getter / setter syntax does not
> conform to our original `name: Type` or `name: func ...` syntax - we're
> missing the colon. But I see no better solution, since we couldn't use
> empty getters / setters with the colon syntax. So, +1 for keeping your
> suggested syntax :)
>
> I think allowing multiple values for setters is an interesting idea, but
> I'm not sure if we should do this. `1, 2` is not an expression in
> "normal" ooc, making it an expression that is only used in the context
> of properties would lead to confusion IMHO.
>
> What Do You Think? ;)
>
> _______________________________________________
> Mailing list: https://launchpad.net/~ooc-dev
> Post to     : ooc-dev@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~ooc-dev
> More help   : https://help.launchpad.net/ListHelp