glib bindings and memory management

[Havoc Pennington]

On Wed, 2005-05-04 at 16:09 -0400, Colin Walters wrote:
> 
> In particular I feel fairly strongly that "a{s?}" must map to
> GHashTable, or we haven't actually gained a lot from the glib bindings
> except that stuff is allocated using g_malloc and variants are GValue,
> which to me is nearly negligible.

I'm not sure this is true (you've still avoided a bunch of manual
boilerplate code - "native types" isn't the whole point of a binding)
but I don't disagree with the substance of your point, just footnoting
that there is other value.

> for (i = 0; i < arr->len; i++) {
>    FooValue *elt = &(g_array_index (arr, i]));
>    g_free (elt->name);
>    g_value_unset (elt->val);
>    g_hash_table_destroy (elt->table);
> }

What if we introduce a dbus_g_free_complex_value() (OK, let's work on
the name) and put the above code snippet in a destructor at the front of
the malloc block? i.e. for simple cases where there's an existing GLib
call to free (g_free, g_hash_table_destroy()) then that existing call
has to be used. For complex cases you have to call the special dbus free
thing.

I pretty strongly agree with your point that we should be sure the basic
cases are really nice, even if it's at the expense of the complex
cases. 

Havoc