Support interfaces

[romgrk]

css = new Gtk.CssProvider();
css.load_from_path('./style.css');
context = win.get_style_context();
context.add_provider(css, 200);

The last call is the one producing this:

(Red text is debug info; just a printf of base_info and type in method V8ToGIArgument)

[magcius]

We don't support interfaces yet.

[romgrk]

All good, just reporting.

[kidwm]

@romgrk how about changing the title of this issue to support interfaces?

[romgrk]

Hello, @magcius I have a question and I was wondering if you could help (but don't feel obligated to).

On a local branch, I have the following:

// value.cc
void V8ToGIArgument(Isolate *isolate, GIBaseInfo *base_info, GIArgument *arg, Handle<Value> value) {
    GIInfoType type = g_base_info_get_type (base_info);

    switch (type) {
    case GI_INFO_TYPE_OBJECT:
        arg->v_pointer = GObjectFromWrapper (value);
        break;
    case GI_INFO_TYPE_INTERFACE: // added
        arg->v_pointer = GObjectFromWrapper (value);
        break;
    // …

Is it safe to assume that the GIArgument will be a GObject? If not, in which case would it be problematic?

(P.S. I recently read your “xplain” articles, that's a very great work! Very interesting)

[friktor]

Hi, I understand the problem is not only in the concrete support of interfaces, but also in the branching units of method calls. Will you soon Fix this problem? Or At least basic support CssProvider & StyleContext, it is quite common basic functionality for Gtk. (At the moment to test the styles only in Gtk-Inspector)

[magcius]

I am not actively working on node-gtk at the moment.

[WebReflection]

So we're done here?

[magcius]

I hope to eventually work on it in the future, but I'm swamped with real work right now.

[WebReflection]

Me too, I was just hoping for some little interest and little progress from time to time. Maybe it's time for me to learn C++ and move on here. Cheers