GSignal documentation
- From: Joshua N Pritikin <vishnu pobox com>
- To: gtk-devel-list gnome org
- Subject: GSignal documentation
- Date: Sun, 19 Aug 2001 19:19:26 +0530
After a few weeks of working with signals, i'm beginning to understand
how they work -- very cool. However, i feel the documentation is
misleading or something. The confusion arises here:
guint g_signal_newc (const gchar *signal_name,
GType itype,
GSignalFlags signal_flags,
guint class_offset,
GSignalAccumulator accumulator,
gpointer accu_data,
GSignalCMarshaller c_marshaller,
GType return_type,
guint n_params,
...);
Now, most of this is good stuff and self explanatory. However, the
naming of `itype' is a mystery. Instead of the generic term `itype',
the parameter name should be suggestive of the signal's subscription
scope or subscription granularity. Perhaps `scope_type' is better.
The same misconception can be seen in the g_signal_connect_* prototype:
gulong g_signal_connect_data (gpointer instance,
const gchar *detailed_signal,
GCallback c_handler,
gpointer data,
GClosureNotify destroy_data,
gboolean swapped,
gboolean after);
Everything is fine, except that the name `instance' is too generic. It's
really a `scope_instance' -- since everything that connects to a signal's
scope receives emissions. Contrary to the documentation, i feel like the
signal system doesn't have much to do with objects. Rather it bridges
scopes and pierces abstraction layers -- i think of it as a run-time
scope system.
--
Get self-realization at <http://sahajayoga.org> ... <http://why-compete.org> ?
Victory to the Divine Mother!!
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]