Re: [PATCH] adapter: Add ConnectDevice method

[Szymon Janc <szymon.janc@codecoup.pl>]

Hi Marcel,

On Friday, 19 January 2018 18:38:18 CET Marcel Holtmann wrote:
> Hi Szymon,
> 
> > This allows to connect device without doing general discovery. This is
> > needed for some of qualification tests where there is no general
> > discovery upfront and we need to do connection to device with provided
> > address.
> > 
> > Another usecase is for scenario where scanning happen on one controller
> > but
> > connection handling on another.
> > 
> > Implementation wide this results in new temporary object being created and
> > connect being called on it which behaves exactly the same as if Connect
> > method from Device1 interface was called on discovered device. If
> > connection fails, created object is removed. On success this method
> > returns path to created device object.
> > 
> > This patch implements bare minimum properties needed for connection -
> > address and address type. If needed eg. for non-NFC based OOB it could be
> > extended with more options.
> > ---
> > doc/adapter-api.txt | 36 ++++++++++++++++++++++
> > src/adapter.c       | 89
> > +++++++++++++++++++++++++++++++++++++++++++++++++++++ src/device.c       
> > | 64 +++++++++++++++++++++++++++++++-------
> > src/device.h        |  2 ++
> > 4 files changed, 180 insertions(+), 11 deletions(-)
> > 
> > diff --git a/doc/adapter-api.txt b/doc/adapter-api.txt
> > index 0533b674a..67b7ca487 100644
> > --- a/doc/adapter-api.txt
> > +++ b/doc/adapter-api.txt
> > @@ -145,6 +145,42 @@ Methods		void StartDiscovery()
> > 
> > 			Possible errors: None
> > 
> > +		object ConnectDevice(dict properties) [experimental]
> > +
> > +			This method creates new temporary device with defined
> > +			properties and issues connection to that device without
> > +			need of performing General Discovery first. Connection
> > +			mechanism is the same as if Connect method from Device1
> > +			interface was called. If connection failed, created
> > +			device is removed and this method fails. If connection
> > +			was successful this method returns object path to
> > +			created device object.
> 
> actually I would prefer if device is never announced if connection fails.
> Otherwise we might cause a bit of D-Bus object signaling overhead. So
> unless successfully connect, the device object should be a shallow object.
> It might exist internally, but is not yet exposed over D-Bus.

OK, I've change the code so that device is announced only after physical 
connection was succesfull. I also modified description that this method 
returns when physical link was created but it will continue with services 
discovery and connection of any supported profiles.

> 
> Also we really need to talk about if the object manager update is triggered
> before the return message. So which message comes first.

RFC v2 implements that OM is updated before this method returns (mostly due to 
code simplicity).

> > +
> > +			Parameters that may be set in the filter dictionary
> > +			include the following:
> > +
> > +			string Address
> > +
> > +				The Bluetooth device address of the remote
> > +				device. This parameter is mandatory.
> > +
> > +			string AddressType
> > +
> > +				The Bluetooth device Address Type. This is
> > +				address type that should be used for initial
> > +				connection. If this parameter is not present
> > +				BR/EDR device is created.
> > +
> > +				Possible values:
> > +					"public" - Public address
> > +					"random" - Random address
> > +
> > +			Possible errors: org.bluez.Error.InvalidArguments
> > +					 org.bluez.Error.AlreadyExists
> > +					 org.bluez.Error.NotSupported
> > +					 org.bluez.Error.NotReady
> > +					 org.bluez.Error.Failed
> > +
> > Properties	string Address [readonly]
> > 
> > 			The Bluetooth device address.
> > 
> > diff --git a/src/adapter.c b/src/adapter.c
> > index 0a25ae27e..31686bc04 100644
> > --- a/src/adapter.c
> > +++ b/src/adapter.c
> > @@ -3080,6 +3080,92 @@ static DBusMessage
> > *get_discovery_filters(DBusConnection *conn,> 
> > 	return reply;
> > 
> > }
> > 
> > +static DBusMessage *connect_device(DBusConnection *conn,
> > +					DBusMessage *msg, void *user_data)
> > +{
> > +	struct btd_adapter *adapter = user_data;
> > +	DBusMessageIter iter, subiter, dictiter, value;
> > +	uint8_t addr_type = BDADDR_BREDR;
> > +	bdaddr_t addr = *BDADDR_ANY;
> > +	struct btd_device *device;
> > +
> > +	DBG("sender %s", dbus_message_get_sender(msg));
> > +
> > +	if (!(adapter->current_settings & MGMT_SETTING_POWERED))
> > +		return btd_error_not_ready(msg);
> > +
> > +	dbus_message_iter_init(msg, &iter);
> > +	if (dbus_message_iter_get_arg_type(&iter) != DBUS_TYPE_ARRAY ||
> > +	    dbus_message_iter_get_element_type(&iter) != DBUS_TYPE_DICT_ENTRY)
> > +		return btd_error_invalid_args(msg);
> > +
> > +	dbus_message_iter_recurse(&iter, &subiter);
> > +	do {
> 
> Why not while (true) { } here? Is there some affinity towards do-loops ;)
> 
> > +		int type = dbus_message_iter_get_arg_type(&subiter);
> > +		char *key;
> > +		char *str;
> > +
> > +		if (type == DBUS_TYPE_INVALID)
> > +			break;
> > +
> > +		dbus_message_iter_recurse(&subiter, &dictiter);
> > +
> > +		dbus_message_iter_get_basic(&dictiter, &key);
> > +		if (!dbus_message_iter_next(&dictiter))
> > +			return btd_error_invalid_args(msg);
> > +
> > +		if (dbus_message_iter_get_arg_type(&dictiter) !=
> > +							DBUS_TYPE_VARIANT)
> > +			return btd_error_invalid_args(msg);
> > +
> > +		dbus_message_iter_recurse(&dictiter, &value);
> > +
> > +		if (!strcmp(key, "Address")) {
> > +			if (dbus_message_iter_get_arg_type(&value) !=
> > +							DBUS_TYPE_STRING)
> > +				return btd_error_invalid_args(msg);
> > +
> > +			dbus_message_iter_get_basic(&value, &str);
> > +
> > +			if (str2ba(str, &addr) < 0 )
> > +				return btd_error_invalid_args(msg);
> > +		} else if (!strcmp(key, "AddressType")) {
> > +			if (dbus_message_iter_get_arg_type(&value) !=
> > +							DBUS_TYPE_STRING)
> > +				return btd_error_invalid_args(msg);
> > +
> > +			dbus_message_iter_get_basic(&value, &str);
> > +
> > +
> > +			if (!strcmp(str, "public"))
> > +				addr_type = BDADDR_LE_PUBLIC;
> > +			else if (!strcmp(str, "random"))
> > +				addr_type = BDADDR_LE_RANDOM;
> > +			else
> > +				return btd_error_invalid_args(msg);
> > +		} else {
> > +			return btd_error_invalid_args(msg);
> > +		}
> > +
> > +		dbus_message_iter_next(&subiter);
> > +	} while (true);
> > +
> > +	if (!bacmp(&addr, BDADDR_ANY))
> > +		return btd_error_invalid_args(msg);
> > +
> > +	device = btd_adapter_find_device(adapter, &addr, addr_type);
> > +	if (device)
> > +		return btd_error_already_exists(msg);
> > +
> > +	device = adapter_create_device(adapter, &addr, addr_type);
> > +	if (!device)
> > +		return btd_error_failed(msg, "Failed to create new device");
> > +
> > +	device_update_last_seen(device, addr_type);
> > +
> > +	return device_connect(conn, msg, device);
> > +}
> > +
> > static const GDBusMethodTable adapter_methods[] = {
> > 
> > 	{ GDBUS_ASYNC_METHOD("StartDiscovery", NULL, NULL, start_discovery) },
> > 	{ GDBUS_METHOD("SetDiscoveryFilter",
> > 
> > @@ -3091,6 +3177,9 @@ static const GDBusMethodTable adapter_methods[] = {
> > 
> > 	{ GDBUS_METHOD("GetDiscoveryFilters", NULL,
> > 	
> > 			GDBUS_ARGS({ "filters", "as" }),
> > 			get_discovery_filters) },
> > 
> > +	{ GDBUS_EXPERIMENTAL_ASYNC_METHOD("ConnectDevice",
> > +				GDBUS_ARGS({ "properties", "a{sv}" }), NULL,
> > +				connect_device) },
> > 
> > 	{ }
> > 
> > };
> > 
> > diff --git a/src/device.c b/src/device.c
> > index 1acecce33..5efde4b54 100644
> > --- a/src/device.c
> > +++ b/src/device.c
> > @@ -624,6 +624,11 @@ static void browse_request_cancel(struct browse_req
> > *req)> 
> > 	attio_cleanup(device);
> > 
> > +	if (req->msg && dbus_message_is_method_call(req->msg,
> > +				"org.bluez.Adapter1", "ConnectDevice"))
> > +		g_dbus_send_message(dbus_conn, btd_error_failed(req->msg,
> > +							strerror(ECANCELED)));
> > +
> > 
> > 	browse_request_free(req);
> > 
> > }
> > 
> > @@ -1577,9 +1582,14 @@ done:
> > 	if (!dev->connect)
> > 	
> > 		return;
> > 
> > -	if (!err && dbus_message_is_method_call(dev->connect, 
DEVICE_INTERFACE,
> > -								"Connect"))
> > +	if (!err) {
> > +		if (dbus_message_is_method_call(dev->connect, DEVICE_INTERFACE,
> > +								"Connect") ||
> > +				dbus_message_is_method_call(dev->connect,
> > +					"org.bluez.Adapter1", "ConnectDevice"))
> > 
> > 		dev->general_connect = TRUE;
> > 
> > +		btd_device_set_temporary(dev, false);
> > +	}
> > 
> > 	DBG("returning response to %s", dbus_message_get_sender(dev->connect));
> > 
> > @@ -1599,7 +1609,15 @@ done:
> > 		/* Start passive SDP discovery to update known services */
> > 		if (dev->bredr && !dev->svc_refreshed)
> > 		
> > 			device_browse_sdp(dev, NULL);
> > 
> > -		g_dbus_send_reply(dbus_conn, dev->connect, DBUS_TYPE_INVALID);
> > +
> > +		if (dbus_message_is_method_call(dev->connect,
> > +					"org.bluez.Adapter1", "ConnectDevice"))
> > +			g_dbus_send_reply(dbus_conn, dev->connect,
> > +						DBUS_TYPE_OBJECT_PATH,
> > +						&dev->path, DBUS_TYPE_INVALID);
> > +		else
> > +			g_dbus_send_reply(dbus_conn, dev->connect,
> > +							DBUS_TYPE_INVALID);
> > 
> > 	}
> > 	
> > 	dbus_message_unref(dev->connect);
> > 
> > @@ -1773,11 +1791,15 @@ static DBusMessage *connect_profiles(struct
> > btd_device *dev, uint8_t bdaddr_type> 
> > 	if (!btd_adapter_get_powered(dev->adapter))
> > 	
> > 		return btd_error_not_ready(msg);
> > 
> > -	btd_device_set_temporary(dev, false);
> > +	if (!dbus_message_is_method_call(msg, "org.bluez.Adapter1",
> > +							"ConnectDevice"))
> > +		btd_device_set_temporary(dev, false);
> > 
> > 	if (!state->svc_resolved)
> > 	
> > 		goto resolve_services;
> > 
> > +	/* will never happen for ConnectDevice call */
> > +
> > 
> > 	dev->pending = create_pending_list(dev, uuid);
> > 	if (!dev->pending) {
> > 	
> > 		if (dev->svc_refreshed) {
> > 
> > @@ -1864,7 +1886,7 @@ static uint8_t select_conn_bearer(struct btd_device
> > *dev)> 
> > 	return dev->bdaddr_type;
> > 
> > }
> > 
> > -static DBusMessage *dev_connect(DBusConnection *conn, DBusMessage *msg,
> > +DBusMessage *device_connect(DBusConnection *conn, DBusMessage *msg,
> > 
> > 							void *user_data)
> > 
> > {
> > 
> > 	struct btd_device *dev = user_data;
> > 
> > @@ -1889,10 +1911,13 @@ static DBusMessage *dev_connect(DBusConnection
> > *conn, DBusMessage *msg,> 
> > 	if (bdaddr_type != BDADDR_BREDR) {
> > 	
> > 		int err;
> > 
> > +		/* This will never happen for ConnectDevice call */
> > 
> > 		if (dev->le_state.connected)
> > 		
> > 			return dbus_message_new_method_return(msg);
> > 
> > -		btd_device_set_temporary(dev, false);
> > +		if (!dbus_message_is_method_call(msg, "org.bluez.Adapter1",
> > +							"ConnectDevice"))
> > +			btd_device_set_temporary(dev, false);
> > 
> > 		if (dev->disable_auto_connect) {
> > 		
> > 			dev->disable_auto_connect = FALSE;
> > 
> > @@ -2272,12 +2297,15 @@ static void browse_request_complete(struct
> > browse_req *req, uint8_t type,> 
> > 	 * device->browse is set and "InProgress" error is returned instead
> > 	 * of actually connecting services
> > 	 */
> > 
> > +
> > 
> > 	msg = dbus_message_ref(req->msg);
> > 	browse_request_free(req);
> > 	req = NULL;
> > 
> > -	if (dbus_message_is_method_call(msg, DEVICE_INTERFACE, "Connect"))
> > -		reply = dev_connect(dbus_conn, msg, dev);
> > +	if (dbus_message_is_method_call(msg, DEVICE_INTERFACE, "Connect") ||
> > +		dbus_message_is_method_call(msg, "org.bluez.Adapter1",
> > +							"ConnectDevice"))
> > +		reply = device_connect(dbus_conn, msg, dev);
> > 
> > 	else if (dbus_message_is_method_call(msg, DEVICE_INTERFACE,
> > 	
> > 							"ConnectProfile"))
> > 		
> > 		reply = connect_profile(dbus_conn, msg, dev);
> > 
> > @@ -2624,7 +2652,7 @@ static DBusMessage *cancel_pairing(DBusConnection
> > *conn, DBusMessage *msg,
> > 
> > static const GDBusMethodTable device_methods[] = {
> > 
> > 	{ GDBUS_ASYNC_METHOD("Disconnect", NULL, NULL, dev_disconnect) },
> > 
> > -	{ GDBUS_ASYNC_METHOD("Connect", NULL, NULL, dev_connect) },
> > +	{ GDBUS_ASYNC_METHOD("Connect", NULL, NULL, device_connect) },
> 
> Looks like you took an easy path here and hacked this into the same message
> handler. So no, do a clean separation and do that the right way. I really
> dislike conditional handling like above. And then especially separating the
> return messages based on how something is called.
> > 	{ GDBUS_ASYNC_METHOD("ConnectProfile", GDBUS_ARGS({ "UUID", "s" }),
> > 	
> > 						NULL, connect_profile) },
> > 	
> > 	{ GDBUS_ASYNC_METHOD("DisconnectProfile", GDBUS_ARGS({ "UUID", "s" }),
> > 
> > @@ -4998,16 +5026,30 @@ done:
> > 		device_browse_gatt(device, NULL);
> > 	
> > 	if (device->connect) {
> > 
> > -		if (err < 0)
> > +		if (err < 0) {
> > 
> > 			reply = btd_error_failed(device->connect,
> > 			
> > 							strerror(-err));
> > 
> > -		else
> > +		} else {
> > 
> > 			reply = dbus_message_new_method_return(device->connect);
> > 
> > +			if (dbus_message_is_method_call(device->connect,
> > +					"org.bluez.Adapter1", "ConnectDevice"))
> > +				dbus_message_append_args(reply,
> > +							DBUS_TYPE_OBJECT_PATH,
> > +							&device->path,
> > +							DBUS_TYPE_INVALID);
> > +		}
> > +
> > 
> > 		g_dbus_send_message(dbus_conn, reply);
> > 		dbus_message_unref(device->connect);
> > 		device->connect = NULL;
> > 	
> > 	}
> > 
> > +
> > +	/* This is the case only for error in ConnectProfile, otherwise device
> > +	 * is not temporary
> > +	 */
> > +	if (err && device_is_temporary(device))
> > +		btd_adapter_remove_device(device->adapter, device);
> > }
> > 
> > int device_connect_le(struct btd_device *dev)
> > diff --git a/src/device.h b/src/device.h
> > index bc046f780..608bc27d3 100644
> > --- a/src/device.h
> > +++ b/src/device.h
> > @@ -151,6 +151,8 @@ void btd_device_set_pnpid(struct btd_device *device,
> > uint16_t source,> 
> > 			uint16_t vendor, uint16_t product, uint16_t version);
> > 
> > int device_connect_le(struct btd_device *dev);
> > +DBusMessage *device_connect(DBusConnection *conn, DBusMessage *msg,
> > +							void *user_data);
> 
> And I do not feel that DBusMessage or DBusConnection should bleed through
> into *.h here.
> 
> Regards
> 
> Marcel


-- 
pozdrawiam
Szymon Janc