mapping single-key input devices to buttons

[Richard Hughes]

On Mon, 2006-05-01 at 17:50 +0200, Johannes Berg wrote:
> Hi,
> 
> > Yes, so, just to recap, the case is that the kernel, for each physical
> > button exports a single input device with a single key, yes? 
> 
> That's the question. I'm no longer sure this approach is good, because
> we end up with tons of input devices. I'd probably prefer in the kernel
> to have just a single one for the PMU.

This is what I put in g-p-m:

/** Use when we want to emit a ButtonPressed event and we know the udi
 *
 *  @param	monitor		This monitor instance
 *  @param	udi		The HAL UDI
 *  @param	details		The event details, or "" for unknown or invalid
 *
 * We can get two different types of ButtonPressed condition
 *   1. The old acpi hardware buttons
 *      udi="acpi_foo", details="";
 *      button.type="power"
 *   2. The new keyboard buttons
 *      udi="foo_Kbd_Port_logicaldev_input", details="sleep"
 *      button.type=""
 */
static void
emit_button_pressed (GpmHalMonitor *monitor,
		     const char	   *udi,
		     const char	   *details)
{
	char	   *button_name = NULL;
	gboolean    value;

	if (strcmp (details, "") == 0) {
		/* no details about the event, so we get more info
		   for type 1 buttons */
		gpm_hal_device_get_string (udi, "button.type", &button_name);
	} else {
		button_name = g_strdup (details);
	}

	/* Buttons without state should default to true. */
	value = TRUE;
	/* we need to get the button state for lid buttons */
	if (strcmp (button_name, "lid") == 0) {
		gpm_hal_device_get_bool (udi, "button.state.value", &value);
	}

	/* we now emit all buttons, even the ones we don't know */
	gpm_debug ("emitting button-pressed : %s (%i)", button_name, value);
	g_signal_emit (monitor, signals [BUTTON_PRESSED], 0, button_name,
value);

	g_free (button_name);
}

Does that help? Sorry I can't be more verbose, but I'm right in the
middle of exams.

Richard.