Class Application

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.

In essence, this results in the #GApplication::activate signal being emitted in the primary instance.

The application must be registered before calling this function.

If the action is expecting a parameter, then the correct type of parameter must be given as parameter. If the action is expecting no parameters then parameter must be %NULL. See g_action_group_get_action_parameter_type().

If the #GActionGroup implementation supports asynchronous remote activation over D-Bus, this call may return before the relevant D-Bus traffic has been sent, or any replies have been received. In order to block on such asynchronous activation calls, g_dbus_connection_flush() should be called prior to the code, which depends on the result of the action activation. Without flushing the D-Bus connection, there is no guarantee that the action would have been activated.

The following code which runs in a remote app instance, shows an example of a "quit" action being activated on the primary app instance over D-Bus. Here g_dbus_connection_flush() is called before exit(). Without g_dbus_connection_flush(), the "quit" action may fail to be activated on the primary instance.

// call "quit" action on primary instance
g_action_group_activate_action (G_ACTION_GROUP (app), "quit", NULL);

// make sure the action is activated now
g_dbus_connection_flush (...);

g_debug ("application has been terminated. exiting.");

exit (0);

If the action map already contains an action with the same name as action then the old action is dropped from the action map.

The action map takes its own reference on action.

Each action is constructed as per one #GActionEntry.

static void
activate_quit (GSimpleAction *simple,
               GVariant      *parameter,
               gpointer       user_data)
{
  exit (0);
}

static void
activate_print_string (GSimpleAction *simple,
                       GVariant      *parameter,
                       gpointer       user_data)
{
  g_print ("%s\n", g_variant_get_string (parameter, NULL));
}

static GActionGroup *
create_action_group (void)
{
  const GActionEntry entries[] = {
    { "quit",         activate_quit              },
    { "print-string", activate_print_string, "s" }
  };
  GSimpleActionGroup *group;

  group = g_simple_action_group_new ();
  g_action_map_add_action_entries (G_ACTION_MAP (group), entries, G_N_ELEMENTS (entries), NULL);

  return G_ACTION_GROUP (group);
}

Calling this function is the equivalent of calling g_application_add_main_option_entries() with a single #GOptionEntry that has its arg_data member set to %NULL.

The parsed arguments will be packed into a #GVariantDict which is passed to #GApplication::handle-local-options. If %G_APPLICATION_HANDLES_COMMAND_LINE is set, then it will also be sent to the primary instance. See g_application_add_main_option_entries() for more details.

See #GOptionEntry for more documentation of the arguments.

This function is comparable to g_option_context_add_main_entries().

After the commandline arguments are parsed, the #GApplication::handle-local-options signal will be emitted. At this point, the application can inspect the values pointed to by arg_data in the given #GOptionEntrys.

Unlike #GOptionContext, #GApplication supports giving a %NULL arg_data for a non-callback #GOptionEntry. This results in the argument in question being packed into a #GVariantDict which is also passed to #GApplication::handle-local-options, where it can be inspected and modified. If %G_APPLICATION_HANDLES_COMMAND_LINE is set, then the resulting dictionary is sent to the primary instance, where g_application_command_line_get_options_dict() will return it. This "packing" is done according to the type of the argument -- booleans for normal flags, strings for strings, bytestrings for filenames, etc. The packing only occurs if the flag is given (ie: we do not pack a "false" #GVariant in the case that a flag is missing).

In general, it is recommended that all commandline arguments are parsed locally. The options dictionary should then be used to transmit the result of the parsing to the primary instance, where g_variant_dict_lookup() can be used. For local options, it is possible to either use arg_data in the usual way, or to consult (and potentially remove) the option from the options dictionary.

This function is new in GLib 2.40. Before then, the only real choice was to send all of the commandline arguments (options and all) to the primary instance for handling. #GApplication ignored them completely on the local side. Calling this function "opts in" to the new behaviour, and in particular, means that unrecognised options will be treated as errors. Unrecognised options have never been ignored when %G_APPLICATION_HANDLES_COMMAND_LINE is unset.

If #GApplication::handle-local-options needs to see the list of filenames, then the use of %G_OPTION_REMAINING is recommended. If arg_data is %NULL then %G_OPTION_REMAINING can be used as a key into the options dictionary. If you do use %G_OPTION_REMAINING then you need to handle these arguments for yourself because once they are consumed, they will no longer be visible to the default handling (which treats them as filenames to be opened).

It is important to use the proper GVariant format when retrieving the options with g_variant_dict_lookup():

• for %G_OPTION_ARG_NONE, use b
• for %G_OPTION_ARG_STRING, use &s
• for %G_OPTION_ARG_INT, use i
• for %G_OPTION_ARG_INT64, use x
• for %G_OPTION_ARG_DOUBLE, use d
• for %G_OPTION_ARG_FILENAME, use ^&ay
• for %G_OPTION_ARG_STRING_ARRAY, use ^a&s
• for %G_OPTION_ARG_FILENAME_ARRAY, use ^a&ay

This function is comparable to g_option_context_add_group().

Unlike g_application_add_main_option_entries(), this function does not deal with %NULL arg_data and never transmits options to the primary instance.

The reason for that is because, by the time the options arrive at the primary instance, it is typically too late to do anything with them. Taking the GTK option group as an example: GTK will already have been initialised by the time the #GApplication::command-line handler runs. In the case that this is not the first-running instance of the application, the existing instance may already have been running for a very long time.

This means that the options from #GOptionGroup are only really usable in the case that the instance of the application being run is the first instance. Passing options like --display= or --gdk-debug= on future runs will have no effect on the existing primary instance.

Calling this function will cause the options in the supplied option group to be parsed, but it does not cause you to be "opted in" to the new functionality whereby unrecognised options are rejected even if %G_APPLICATION_HANDLES_COMMAND_LINE was given.

The binding holds a reference to application while it is active, but not to object. Instead, the binding is destroyed when object is finalized.

Whenever the source_property is changed the target_property is updated using the same value. For instance:

  g_object_bind_property (action, "active", widget, "sensitive", 0);

Will result in the "sensitive" property of the widget #GObject instance to be updated with the same value of the "active" property of the action #GObject instance.

If flags contains %G_BINDING_BIDIRECTIONAL then the binding will be mutual: if target_property on target changes then the source_property on source will be updated as well.

The binding will automatically be removed when either the source or the target instances are finalized. To remove the binding without affecting the source and the target you can just call g_object_unref() on the returned #GBinding instance.

Removing the binding by calling g_object_unref() on it must only be done if the binding, source and target are only used from a single thread and it is clear that both source and target outlive the binding. Especially it is not safe to rely on this if the binding, source or target can be finalized from different threads. Keep another reference to the binding and use g_binding_unbind() instead to be on the safe side.

A #GObject can have multiple bindings.

This function is the language bindings friendly version of g_object_bind_property_full(), using #GClosures instead of function pointers.

The action must be stateful and value must be of the correct type. See g_action_group_get_action_state_type().

This call merely requests a change. The action may refuse to change its state or may change its state to something other than value. See g_action_group_get_action_state_hint().

If the value GVariant is floating, it is consumed.

This is necessary for accessors that modify multiple properties to prevent premature notification while the object is still being modified.

An action must be enabled in order to be activated or in order to have its state changed from outside callers.

When activating the action using g_action_group_activate_action(), the #GVariant given to that function must be of the type returned by this function.

In the case that this function returns %NULL, you must not give any #GVariant, but %NULL instead.

The parameter type of a particular action will never change but it is possible for an action to be removed and for a new action to be added with the same name but a different parameter type.

If the action is not stateful then %NULL will be returned. If the action is stateful then the type of the return value is the type given by g_action_group_get_action_state_type().

The return value (if non-%NULL) should be freed with g_variant_unref() when it is no longer required.

If %NULL is returned it either means that the action is not stateful or that there is no hint about the valid range of values for the state of the action.

If a #GVariant array is returned then each item in the array is a possible value for the state. If a #GVariant pair (ie: two-tuple) is returned then the tuple specifies the inclusive lower and upper bound of valid values for the state.

In any case, the information is merely a hint. It may be possible to have a state value outside of the hinted range and setting a value within the range may fail.

The return value (if non-%NULL) should be freed with g_variant_unref() when it is no longer required.

If the action is stateful then this function returns the #GVariantType of the state. All calls to g_action_group_change_action_state() must give a #GVariant of this type and g_action_group_get_action_state() will return a #GVariant of the same type.

If the action is not stateful then this function will return %NULL. In that case, g_action_group_get_action_state() will return %NULL and you must not call g_action_group_change_action_state().

The state type of a particular action will never change but it is possible for an action to be removed and for a new action to be added with the same name but a different state type.

If #GApplication is using its D-Bus backend then this function will return the #GDBusConnection being used for uniqueness and communication with the desktop environment and other instances of the application.

If #GApplication is not using D-Bus then this function will return %NULL. This includes the situation where the D-Bus backend would normally be in use but we were unable to connect to the bus.

This function must not be called before the application has been registered. See g_application_get_is_registered().

If #GApplication is using its D-Bus backend then this function will return the D-Bus object path that #GApplication is using. If the application is the primary instance then there is an object published at this path. If the application is not the primary instance then the result of this function is undefined.

If #GApplication is not using D-Bus then this function will return %NULL. This includes the situation where the D-Bus backend would normally be in use but we were unable to connect to the bus.

This function must not be called before the application has been registered. See g_application_get_is_registered().

See #GApplicationFlags.

This is the amount of time (in milliseconds) after the last call to g_application_release() before the application stops running.

An application is registered if g_application_register() has been successfully called.

If application is remote then it means that another instance of application already exists (the 'primary' instance). Calls to perform actions on application will result in the actions being performed by the primary instance.

The value of this property cannot be accessed before g_application_register() has been called. See g_application_get_is_registered().

The value can be:

• an empty #GValue initialized by %G_VALUE_INIT, which will be automatically initialized with the expected type of the property (since GLib 2.60)
• a #GValue initialized with the expected type of the property
• a #GValue initialized with a type to which the expected type of the property can be transformed

In general, a copy is made of the property contents and the caller is responsible for freeing the memory by calling g_value_unset().

Note that g_object_get_property() is really intended for language bindings, g_object_get() is much more convenient for C programming.

See g_application_set_resource_base_path() for more information.

Use this function to indicate that the application has a reason to continue to run. For example, g_application_hold() is called by GTK+ when a toplevel window is on the screen.

To cancel the hold, call g_application_release().

The caller is responsible for freeing the list with g_strfreev() when it is no longer required.

If no such action exists, returns %NULL.

Use this function to indicate that the application is busy, for instance while a long running operation is pending.

The busy state will be exposed to other processes, so a session shell will use that information to indicate the state to the user (e.g. with a spinner).

To cancel the busy indication, use g_application_unmark_busy().

The application must be registered before calling this function.

When possible, eg. when signaling a property change from within the class that registered the property, you should use g_object_notify_by_pspec() instead.

Note that emission of the notify signal may be blocked with g_object_freeze_notify(). In this case, the signal emissions are queued and will be emitted (in reverse order) when g_object_thaw_notify() is called.

This function omits the property name lookup, hence it is faster than g_object_notify().

One way to avoid using g_object_notify() from within the class that registered the properties, and using g_object_notify_by_pspec() instead, is to store the GParamSpec used with g_object_class_install_property() inside a static array, e.g.:

  enum
  {
    PROP_0,
    PROP_FOO,
    PROP_LAST
  };

  static GParamSpec *properties[PROP_LAST];

  static void
  my_object_class_init (MyObjectClass *klass)
  {
    properties[PROP_FOO] = g_param_spec_int ("foo", "Foo", "The foo",
                                             0, 100,
                                             50,
                                             G_PARAM_READWRITE);
    g_object_class_install_property (gobject_class,
                                     PROP_FOO,
                                     properties[PROP_FOO]);
  }

and then notify a change on the "foo" property with:

  g_object_notify_by_pspec (self, properties[PROP_FOO]);

In essence, this results in the #GApplication::open signal being emitted in the primary instance.

n_files must be greater than zero.

hint is simply passed through to the ::open signal. It is intended to be used by applications that have multiple modes for opening files (eg: "view" vs "edit", etc). Unless you have a need for this functionality, you should use "".

The application must be registered before calling this function and it must have the %G_APPLICATION_HANDLES_OPEN flag set.

This function acquires the information available from g_action_group_has_action(), g_action_group_get_action_enabled(), g_action_group_get_action_parameter_type(), g_action_group_get_action_state_type(), g_action_group_get_action_state_hint() and g_action_group_get_action_state() with a single function call.

This provides two main benefits.

The first is the improvement in efficiency that comes with not having to perform repeated lookups of the action in order to discover different things about it. The second is that implementing #GActionGroup can now be done by only overriding this one virtual function.

The interface provides a default implementation of this function that calls the individual functions, as required, to fetch the information. The interface also provides default implementations of those functions that call this function. All implementations, therefore, must override either this function or all of the others.

If the action exists, %TRUE is returned and any of the requested fields (as indicated by having a non-%NULL reference passed in) are filled. If the action doesn't exist, %FALSE is returned and the fields may or may not have been modified.

Upon return to the mainloop, g_application_run() will return, calling only the 'shutdown' function before doing so.

The hold count is ignored. Take care if your code has called g_application_hold() on the application and is therefore still expecting it to exist. (Note that you may have called g_application_hold() indirectly, for example through gtk_application_add_window().)

The result of calling g_application_run() again after it returns is unspecified.

Since GLib 2.56, if GLIB_VERSION_MAX_ALLOWED is 2.56 or greater, the type of object will be propagated to the return type (using the GCC typeof() extension), so any casting the caller needs to do on the return type must be explicit.

In other words, if the object is floating, then this call "assumes ownership" of the floating reference, converting it to a normal reference by clearing the floating flag while leaving the reference count unchanged. If the object is not floating, then this call adds a new normal reference increasing the reference count by one.

Since GLib 2.56, the type of object will be propagated to the return type under the same conditions as for g_object_ref().

This is the point at which the application discovers if it is the primary instance or merely acting as a remote for an already-existing primary instance. This is implemented by attempting to acquire the application identifier as a unique bus name on the session bus using GDBus.

If there is no application ID or if %G_APPLICATION_NON_UNIQUE was given, then this process will always become the primary instance.

Due to the internal architecture of GDBus, method calls can be dispatched at any time (even if a main loop is not running). For this reason, you must ensure that any object paths that you wish to register are registered before calling this function.

If the application has already been registered then %TRUE is returned with no work performed.

The #GApplication::startup signal is emitted if registration succeeds and application is the primary instance (including the non-unique case).

In the event of an error (such as cancellable being cancelled, or a failure to connect to the session bus), %FALSE is returned and error is set appropriately.

Note: the return value of this function is not an indicator that this instance is or is not the primary instance of the application. See g_application_get_is_remote() for that.

When the use count reaches zero, the application will stop running.

Never call this function except to cancel the effect of a previous call to g_application_hold().

If no action of this name is in the map then nothing happens.

This function is intended to be run from main() and its return value is intended to be returned by main(). Although you are expected to pass the argc, argv parameters from main() to this function, it is possible to pass %NULL if argv is not available or commandline handling is not required. Note that on Windows, argc and argv are ignored, and g_win32_get_command_line() is called internally (for proper support of Unicode commandline arguments).

#GApplication will attempt to parse the commandline arguments. You can add commandline flags to the list of recognised options by way of g_application_add_main_option_entries(). After this, the #GApplication::handle-local-options signal is emitted, from which the application can inspect the values of its #GOptionEntrys.

#GApplication::handle-local-options is a good place to handle options such as --version, where an immediate reply from the local process is desired (instead of communicating with an already-running instance). A #GApplication::handle-local-options handler can stop further processing by returning a non-negative value, which then becomes the exit status of the process.

What happens next depends on the flags: if %G_APPLICATION_HANDLES_COMMAND_LINE was specified then the remaining commandline arguments are sent to the primary instance, where a #GApplication::command-line signal is emitted. Otherwise, the remaining commandline arguments are assumed to be a list of files. If there are no files listed, the application is activated via the #GApplication::activate signal. If there are one or more files, and %G_APPLICATION_HANDLES_OPEN was specified then the files are opened via the #GApplication::open signal.

If you are interested in doing more complicated local handling of the commandline then you should implement your own #GApplication subclass and override local_command_line(). In this case, you most likely want to return %TRUE from your local_command_line() implementation to suppress the default handling. See [gapplication-example-cmdline2.c][https://gitlab.gnome.org/GNOME/glib/-/blob/HEAD/gio/tests/gapplication-example-cmdline2.c] for an example.

If, after the above is done, the use count of the application is zero then the exit status is returned immediately. If the use count is non-zero then the default main context is iterated until the use count falls to zero, at which point 0 is returned.

If the %G_APPLICATION_IS_SERVICE flag is set, then the service will run for as much as 10 seconds with a use count of zero while waiting for the message that caused the activation to arrive. After that, if the use count falls to zero the application will exit immediately, except in the case that g_application_set_inactivity_timeout() is in use.

This function sets the prgname (g_set_prgname()), if not already set, to the basename of argv[0].

Much like g_main_loop_run(), this function will acquire the main context for the duration that the application is running.

Since 2.40, applications that are not explicitly flagged as services or launchers (ie: neither %G_APPLICATION_IS_SERVICE or %G_APPLICATION_IS_LAUNCHER are given as flags) will check (from the default handler for local_command_line) if "--gapplication-service" was given in the command line. If this flag is present then normal commandline processing is interrupted and the %G_APPLICATION_IS_SERVICE flag is set. This provides a "compromise" solution whereby running an application directly from the commandline will invoke it in the normal way (which can be useful for debugging) while still allowing applications to be D-Bus activated in service mode. The D-Bus service file should invoke the executable with "--gapplication-service" as the sole commandline argument. This approach is suitable for use by most graphical applications but should not be used from applications like editors that need precise control over when processes invoked via the commandline will exit and what their exit status will be.

This function should only be called from object system implementations.

Notifications may persist after the application exits. It will be D-Bus-activated when the notification or one of its actions is activated.

Modifying notification after this call has no effect. However, the object can be reused for a later call to this function.

id may be any string that uniquely identifies the event for the application. It does not need to be in any special format. For example, "new-message" might be appropriate for a notification about new messages.

If a previous notification was sent with the same id, it will be replaced with notification and shown again as if it was a new notification. This works even for notifications sent from a previous execution of the application, as long as id is the same string.

id may be %NULL, but it is impossible to replace or withdraw notifications without an id.

If notification is no longer relevant, it can be withdrawn with g_application_withdraw_notification().

The application id can only be modified if application has not yet been registered.

If non-%NULL, the application id must be valid. See g_application_id_is_valid().

If the object already had an association with that name, the old association will be destroyed.

Internally, the key is converted to a #GQuark using g_quark_from_string(). This means a copy of key is kept permanently (even after object has been finalized) — so it is recommended to only use a small, bounded set of values for key in your program, to avoid the #GQuark storage growing unbounded.

This function does not take its own reference on application. If application is destroyed then the default application will revert back to %NULL.

The flags can only be modified if application has not yet been registered.

See #GApplicationFlags.

This is the amount of time (in milliseconds) after the last call to g_application_release() before the application stops running.

This call has no side effects of its own. The value set here is only used for next time g_application_release() drops the use count to zero. Any timeouts currently in progress are not impacted.

See g_option_context_set_description() for more information.

This function registers the argument to be passed to g_option_context_new() when the internal #GOptionContext of application is created.

See g_option_context_new() for more information about parameter_string.

See g_option_context_set_summary() for more information.

The path is used to automatically load various [application resources][gresource] such as menu layouts and action descriptions. The various types of resources will be found at fixed names relative to the given base path.

By default, the resource base path is determined from the application ID by prefixing '/' and replacing each '.' with '/'. This is done at the time that the #GApplication object is constructed. Changes to the application ID after that point will not have an impact on the resource base path.

As an example, if the application has an ID of "org.example.app" then the default resource base path will be "/org/example/app". If this is a #GtkApplication (and you have not manually changed the path) then Gtk will then search for the menus of the application at "/org/example/app/gtk/menus.ui".

See #GResource for more information about adding resources to your application.

You can disable automatic resource loading functionality by setting the path to %NULL.

Changing the resource base path once the application is running is not recommended. The point at which the resource path is consulted for forming paths for various purposes is unspecified. When writing a sub-class of #GApplication you should either set the #GApplication:resource-base-path property at construction time, or call this function during the instance initialization. Alternatively, you can call this function in the #GApplicationClass.startup virtual function, before chaining up to the parent implementation.

void
object_add_to_user_list (GObject     *object,
                         const gchar *new_string)
{
  // the quark, naming the object data
  GQuark quark_string_list = g_quark_from_static_string ("my-string-list");
  // retrieve the old string list
  GList *list = g_object_steal_qdata (object, quark_string_list);

  // prepend new string
  list = g_list_prepend (list, g_strdup (new_string));
  // this changed 'list', so we need to set it again
  g_object_set_qdata_full (object, quark_string_list, list, free_string_list);
}
static void
free_string_list (gpointer data)
{
  GList *node, *list = data;

  for (node = list; node; node = node->next)
    g_free (node->data);
  g_list_free (list);
}

Using g_object_get_qdata() in the above example, instead of g_object_steal_qdata() would have left the destroy function set, and thus the partial string list would have been freed upon g_object_set_qdata_full().

Duplicate notifications for each property are squashed so that at most one #GObject::notify signal is emitted for each property, in the reverse order in which they have been queued.

It is an error to call this function when the freeze count is zero.

When the busy count reaches zero, the new state will be propagated to other processes.

This function must only be called to cancel the effect of a previous call to g_application_mark_busy().

If the pointer to the #GObject may be reused in future (for example, if it is an instance variable of another object), it is recommended to clear the pointer to %NULL rather than retain a dangling pointer to a potentially invalid #GObject instance. Use g_clear_object() for this.

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.

This function should only be called by #GActionGroup implementations.