On Sun, Aug 18, 2013 at 8:15 AM, Tanu Kaskinen
<tanu.kaskinen at linux.intel.com> wrote:
> On Tue, 2013-08-13 at 01:54 -0300, jprvita at gmail.com wrote:
>> From: Jo?o Paulo Rechi Vita <jprvita at openbossa.org>
>>
>> ---
>> src/modules/bluetooth/bluez5-util.c | 28 ++++++++++++++++++++++++++--
>> 1 file changed, 26 insertions(+), 2 deletions(-)
>>
>> diff --git a/src/modules/bluetooth/bluez5-util.c b/src/modules/bluetooth/bluez5-util.c
>> index 1194503..1d98174 100644
>> --- a/src/modules/bluetooth/bluez5-util.c
>> +++ b/src/modules/bluetooth/bluez5-util.c
>> @@ -697,11 +697,35 @@ fail:
>> }
>>
>> static DBusMessage *endpoint_clear_configuration(DBusConnection *conn, DBusMessage *m, void *userdata) {
>> + pa_bluetooth_discovery *y = userdata;
>> + pa_bluetooth_transport *t;
>> DBusMessage *r;
>> + DBusError err;
>> + const char *path;
>>
>> - pa_assert_se(r = dbus_message_new_error(m, BLUEZ_MEDIA_ENDPOINT_INTERFACE ".Error.NotImplemented",
>> - "Method not implemented"));
>> + dbus_error_init(&err);
>>
>> + if (!dbus_message_get_args(m, &err, DBUS_TYPE_OBJECT_PATH, &path, DBUS_TYPE_INVALID)) {
>> + pa_log_error("Endpoint ClearConfiguration(): %s", err.message);
>> + dbus_error_free(&err);
>> + goto fail;
>> + }
>> +
>> + if ((t = pa_hashmap_get(y->transports, path))) {
>> + pa_log_debug("Clearing transport %s profile %s", t->path, pa_bluetooth_profile_to_string(t->profile));
>> + transport_state_changed(t, PA_BLUETOOTH_TRANSPORT_STATE_DISCONNECTED);
>> + t->device->transports[t->profile] = NULL;
>> + /* TODO: check if there is no risk of calling
>> + * transport_connection_changed_cb() when the transport is already freed */
>> + pa_bluetooth_transport_free(t);
>
> Similar to what I said about device_free(), the core code shouldn't call
> pa_bluetooth_transport_free() here either. I think
> pa_bluetooth_transport should have a configuration_cleared() callback,
> and the transport backend can then call pa_bluetooth_transport_free() if
> it wants.
>
All endpoint methods on the D-Bus interface are specific to the BlueZ
transport-backend, so it's not the core calling
pa_bluetooth_transport_free(). If we ever decide to separate the BlueZ
5 transport backend to a different file, the endpoint functions should
be moved altogether.
--
Jo?o Paulo Rechi Vita
http://about.me/jprvita
