Alejandro Piñeiro authored Mar 14, 2014 and Mike Gorse committed Mar 17, 2014

https://bugzilla.gnome.org/show_bug.cgi?id=726243

Per the spec, org.a11y.Atspi.Component.Contains takes a uint32 for the coord
type, not an int16

https://bugzilla.gnome.org/show_bug.cgi?id=710730

Apparently, the call to register with gnome-session-manager can fail without
setting a GError, and, in any case, it doesn't hurt to check that the error
is set before using it.

https://bugzilla.gnome.org/show_bug.cgi?id=710689

Since g_source_attach adds a ref to a source and we keep the source stored
in a global variable and, therefore, do not unref it after attaching it, we
thus need to unref it when it is being destroyed and we clear the global
variable.

https://bugzilla.gnome.org/show_bug.cgi?id=709625

https://bugzilla.gnome.org/show_bug.cgi?id=700865

dbind isn't currently running a GMainLoop, so it shouldn't call
atspi_dbus_connection_setup_with_g_main, and calling this is harmful if
it is being called from a context other than the default context, since
D-Bus messages will no longer be delivered.

https://bugzilla.gnome.org/show_bug.cgi?id=707218

Rather than use complicated gboolean-cast-to-gpointer to make
decisions, simply use two different callbacks.

If at-spi2-atk enters a new main context and calls
atspi_set_main_context and process_deferred_messages has already been
scheduled for the original context, then it would not be called until
the original main loop is re-entered, meaning that Orca will sometimes
deadlock in its preferences dialog, since its key listener won't get the
notification for its own key event, as it will be deferred, while
at-spi2-atk waits for the listener to be processed.

https://bugzilla.gnome.org/show_bug.cgi?id=707218

atspi-misc.c was not including config.h anywhere, resulting in HAVE_X11
not being defined, so X would not be queried when fetching the
accessibility bus address.

https://bugzilla.gnome.org/show_bug.cgi?id=707084

Calling dbus_connection_dispatch isn't really needed when fetching the
bus address--this is only done once.

dbus_connection_dispatch is trying to lock the connection when it is
already locked, resulting in a deadlock, so just don't call it
recursively for now.

https://bugzilla.gnome.org/show_bug.cgi?id=707010

That unref is done inside an if that starts with a !parent condition, so
unrefing it doesn't makes sense

https://bugzilla.gnome.org/show_bug.cgi?id=706893

If no main loop is running, then the dbus connection's incoming message
queue was never emptied, so with, eg, dogtail,
dbus_connection_send_with_reply_and_block would take longer and longer
to look through the message queue to find a reply. This also effectively
led to a memory leak.

https://bugzilla.gnome.org/show_bug.cgi?id=701145

Previously, AT-SPI would hard-code a default display name of ":0" if
neither DISPLAY nor AT_SPI_DISPLAY were set. This seems unnecessary--the
environment variable ought to be set--and causes a warning to be
generated whenever a Gtk app is started under Wayland.

https://bugzilla.gnome.org/show_bug.cgi?id=701378

Separate out X11-specific code in preparation for Wayland support and to
avoid calling X under Mac OS. It should now be possible to build without
X, although keyboard and mouse-related functionality will likely not
work.

https://bugzilla.gnome.org/show_bug.cgi?id=682199

These are intended as C convenience functions. Gjs doesn't support
having a GDestroyNotify without a user_data and prints out a warning, so
might as well skip for purposes of introspection.

When timing out without having received a reply from a pending call,
then
cancel the call so that it will be removed from the DBusConnection's
queue. This will hopefully prevent the call and the
SpiReentrantCallClosure from being leaked.

Note that this code is still behaving oddly (why isn't a notification
being sent by libdbus when the specified amount of time has ellapsed?)

Only dup the string once.

Add a function to allow at-spi2-atk to set the main context that is used
when adding idle handlers (ie, process_deferred_messages, which is used
to prevent new events from being sent to the AT while other events are
being processed).

Also bump version to 2.9.4.

Added a function to query whether libatspi has been initialized. Used by
at-spi2-atk. Also, bump version.

Atspi_role_get_name is used (by Orca, for instance) and was
inadvertently removed with the enum refactor.

Having functions with common suffixes (ie, atspi_text_get_attributes and
atspi_document_get_attributes) results in problems for language
bindings. Deprecating these functions in favor of functions with new
names to avoid these collisions.

https://bugzilla.gnome.org/show_bug.cgi?id=700243