Re: Sinkability: usefull in language bindings ? [was: Re: Sinkability considered harmful]
- From: Tristan Van Berkom <tristan van berkom gmail com>
- To: muppet <scott asofyet org>
- Cc: GTK+ development mailing list <gtk-devel-list gnome org>
- Subject: Re: Sinkability: usefull in language bindings ? [was: Re: Sinkability considered harmful]
- Date: Sat, 07 Jan 2006 15:00:49 -0500
On Jan 4, 2006, at 11:51 PM, Tristan Van Berkom wrote:
Excuse me but I still dont register, why can you assume that
a function returning a GtkObject cant return an added reference ?
Excuse my intermittent "dropping off the planet" ;-)
Such special cases such as that are handled as special cases. I can
recall implementing a few, but can't recall which ones off the top of
Long story short, lest it get lost:
- there is truth to the assertion that floating can aid language bindings.
I do understand that "floating" objects as a concept may be usefull
to language bindings; since it seemingly provides an interface for
"single anonymous ownership", but as you concede, there are still
special-cases to be handled.
- nevertheless, i dislike the inconsistency of semantics and would
rather not have it.
] [Thread Prev