https://forge.codesys.com/lib/cobolt/tickets/You searched for assigned_to:"aliazzz"enSun, 18 Oct 2020 19:43:12 -0000https://forge.codesys.com/lib/cobolt/tickets/64/If you only call PublishNDATA once, then an NDATA is not sent. If you call PublishNDATA repeatedly though, it continuously sends NDATA. A balance should be found, such that each change is sent only once.i-campbellSat, 03 Oct 2020 22:28:07 -0000https://forge.codesys.com/lib/cobolt/tickets/64/https://forge.codesys.com/lib/cobolt/tickets/63/NDATA uses another list, ChangeList, in addition to the AllMetricsList. The problem is, these elements can only belong to one list. Solution: Rather than how it currently is: 1. Cycle through Every Element 2. Add changed element to ChangeList (side effect: removing element from AllMetricsList) 3. Call the AddListOfMetrics, which cycles through every changed element and adds them to the Encoder Should do: 1. Cycle through Every Element 2. Add changed element to Encoderi-campbellSat, 03 Oct 2020 17:57:07 -0000https://forge.codesys.com/lib/cobolt/tickets/63/https://forge.codesys.com/lib/cobolt/tickets/61/In the code, it asks "What is the maximum size of an NDEATH" Answer: 44 Bytes. (see attachment) Note that changing the size to exactly 44 requires the fix in ticket [#60]. i-campbellSat, 03 Oct 2020 22:28:28 -0000https://forge.codesys.com/lib/cobolt/tickets/61/https://forge.codesys.com/lib/cobolt/tickets/56/Update code to check/replace these illegal characters "/,+,#" from the Edge_Node _Id or Group_id element The illegal characters will be replaced with an underscore "_" aliazzzWed, 30 Sep 2020 16:34:44 -0000https://forge.codesys.com/lib/cobolt/tickets/56/https://forge.codesys.com/lib/cobolt/tickets/55/# Publish interval Definition PublishInterval / PublishRate is default set to 5000 ms. So we cannot publish "updates" any faster then each 5s. This means the publish rate is a kind of "sample-rate" per signal change: * 5000 means 1 signal change per interval of 5 seconds is sent, * 100 means 1 signal change per interval of 100mec is sent, * 10000 means 1 signal change per interval of 10 seconds is sent. # Solution We have a pulse timer (or some other means to detect a time interval), it sets itself as soon as we have sent data and after T#PublishInterval it resets itself. If new data is detected during timer countdown we ignore it. If the timer has expired we can send data immediately and the timer sets itself again after this action. So, the timer "blocks" sending of data during countdown.aliazzzWed, 30 Sep 2020 16:35:09 -0000https://forge.codesys.com/lib/cobolt/tickets/55/https://forge.codesys.com/lib/cobolt/tickets/54/similar to [#53] PublishNodeData() roughly line 17 through 19 NDATATopic := WCONCAT( SPARKPLUGB, GroupId); NDATATopic := WCONCAT( **_LastWillTopic**, NODE_DATA_TOPIC); NDATATopic := WCONCAT( **_LastWillTopic**, NodeName); Should be all NDATATopici-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/54/https://forge.codesys.com/lib/cobolt/tickets/53/copy paste edit error around FB_Edge_Of_Network_Node.PublishNodeBirth line 22-24. **NBIRTHTopic** := WCONCAT( SPARKPLUGB, GroupId); NBIRTHTopic := WCONCAT( **_LastWillTopic**, NODE_BIRTH_TOPIC); NBIRTHTopic := WCONCAT( **_LastWillTopic**, FW_SLASH); NBIRTHTopic := WCONCAT( **_LastWillTopic**, NodeName); _LastWillTopic should be NBIRTHTopic Additionally, as NODE_BIRTH_TOPIC := "/NBIRTH/" you don't need the line with FW_SLASH (otherwise it becomes spBv1.0/NBIRTH//myNodeNamei-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/53/https://forge.codesys.com/lib/cobolt/tickets/52/Similar to [#50], but this time around line 30-32 of FB_MQTT(). _PublishMessageSize := DINT_TO_UDINT(2*Stu.StrLenW( ADR(_PublishMessage))); Should be deleted.i-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/52/https://forge.codesys.com/lib/cobolt/tickets/51/EON.PublishBirth() calls _MQTT.publish() twice. publish() only sets the inputs to _Publisher, it is not until _MQTT() is called that the message is published with a call to _Publisher(). Either have some sort of state machine to 1. Connect 2. Publish NBirth 3. Publish DBirth Or have two publishers (and two blobs, etc)i-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/51/https://forge.codesys.com/lib/cobolt/tickets/50/FB_MQTT Line 2 is _LastWillMessageSize := DINT_TO_UINT(2*Stu.StrLenW(ADR(_LastWillMessage))); This doesn't make sense. _LastWillMessageSize is already set in FB_MQTT.SetLastWill, so you don't need to set it again to twice the length of some random series of bytes before you hit a dword 0x0000 (Pointer to Pointer to a Byte Array which isn't a wstring). I think line 2 has to go! (also line 1)i-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/50/https://forge.codesys.com/lib/cobolt/tickets/49/FB_Edge_Of_Network_Node.ServerURL is not used, so the user cannot connect to a broker of their choice. Ideally this would be passed to _MQTT before connecti-campbellMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/49/https://forge.codesys.com/lib/cobolt/tickets/45/It seems that all unit tests are returning the result of the biggest unit test. ==> SHOULD: Each returns a different payload.i-campbellSun, 27 Sep 2020 16:33:39 -0000https://forge.codesys.com/lib/cobolt/tickets/45/https://forge.codesys.com/lib/cobolt/tickets/41/I gather the description says it all! Add a SparkplugB.T_DateTime ALIAS TYPE for ULINT Usage: TimeStampOld : ULINT; TimeStampNew : SparkplugB.T_DateTime; // Still an ULINT ;-) aliazzzSat, 03 Oct 2020 21:17:15 -0000https://forge.codesys.com/lib/cobolt/tickets/41/https://forge.codesys.com/lib/cobolt/tickets/39/FB_SparkplugB_SimpleEncoder This function block acts as a Helper to reduce work for FB_EoN. The FB contains a protobuf blob encoder and various methods to quickly create (compose and/or encode) various messages. Since this FB is of the 'simple' variant, it can can only deal with the SimpleObjects type. Messages which are mandatory for this FB are: * [x] Node Death Certificate * [x] Node Birth Certificate * [x] Node Data * [x] Device Death Certificate * [x] Device Birth Certificate * [x] Device Data Pass a list of every All Metrics Which Will Ever Reported on during this "Lifetime". (1) * [x] UpdateListOfMetricsWhichWillEverReportedOn( ListOfMetricsWhichWillEverReportedOn ) Note 1 This Method can be used to pass the entire list of metrics which will ever be reported on and send this list ONLY on node / device (re)birth. We can only send values which have been 'announced at birth' of the Node or Device. Since the Device or Node can die and be reborn, elements could be added or removed from the list which will be announced only at the next Birth. Warning! Never send values that have not been announced this Birth! Idea Add a way to let a Node/Device die and be reborn in order to anounce the latest list so we can update 'on-the-fly' Pass a single instance of SimpleObject to the Encoder. (2) * [x] AddSimpleMetric( Timestamp, SimpleObject ) Note 2 This method accepts a single SimpleObject deliberatly, to keep burnerplate code low and added flexibility as it is not mandatory for a SimpleObject to be part of an array or a list. The SimpleObject must have been announced at this Birth. Pass a prepared list of SimpleObjects to the Encoder. (3) * [x] AddMetricList( Timestamp, ListOfSimpleObjects ) Note 3 This method accepts a List of SimpleObject, this way you can send multiple values in one list. This list can not contain other metrics metrics than those announced on birth. The list is thus at most the same list as "ListOfMetricsWhichWillEverReportedOn". Apart from these Methods the FB contains various methods for interaction and diagnosis * [x] Diag // Diagnosis of the internal decoder state * [x] Encode( Enable ) // Start Encoding if Enable = TRUE, Stops encoding if enable = FALSE * [x] Encoded // Is encoding finished? * [x] EncodeBusy // Is the encoder busy? * [x] EncodeError( => eError ) // An error occured * [x] InitEncoderBeforeAddingMetrics( Timestamp, SeqNr) // Initialize the Encoder before adding Metrics * [x] GetBDSeqNum * [x] GetSeqNum Last but not least it should provide the end user with a way to obtain the protobuf encoded (serialized) blob in order to publish it under the correct SparkPlugB Topic * [x] GetEncodedBlob( BlobAdr, BlobSize )aliazzzSat, 03 Oct 2020 23:45:17 -0000https://forge.codesys.com/lib/cobolt/tickets/39/https://forge.codesys.com/lib/cobolt/tickets/37/As MQTT.MQTTParam.g_udiMaxPayloadSize and MQTT.MQTTParam.g_udiMaxPacketSize are library parameters, the end user has to set these, we cannot change them using the library. However, we can raise errors if we try to send something too big. To investigate: what happens when a packet or payload is received which is too big?i-campbellSun, 18 Oct 2020 11:17:31 -0000https://forge.codesys.com/lib/cobolt/tickets/37/https://forge.codesys.com/lib/cobolt/tickets/33/Unit test will be written to cover 50% of the FB's and methods from the library using Cfunit v1.1.0.0 or better. The unittest project in the trunk is to be leading. Any other tests can be written in branches to test personal developments. As the trunk of the library gets updated, ideally the unittest project of the trunk gets updated too. The software will be unit tested and/or integration tested, as much as possible. **My suggestion is that we all take a peek at the unit test code whenever we feel like it to ensure that it passes a basic sanity level** aliazzzSun, 18 Oct 2020 19:43:12 -0000https://forge.codesys.com/lib/cobolt/tickets/33/https://forge.codesys.com/lib/cobolt/tickets/32/PubSub can inform us that a message has been received via .MessageReceived(). The message & topic then can be obtained via .getmessage(). Example ~~~ IF PubSub.MessageReceived() THEN myRecvMessage := PubSub.MyMessage( wsTopic => MyTopic); // pass the message to the payload decoder MyDecodedMessage := Payload.Decode( pBuf:=ADR(myRecvMessage),udiBufSize:=SIZEOF(myRecvMessage)); END_IF; ~~~ => Passing a pointer/length of the blob (the undecoded received message) is far superior in speed and resources as no data is physically copied around. Binding a dynamic buffer is under future consideration. => Interaction/handshaking between PubSub and Payload in the usual way (xbusy, xdone, xerror) and methods for this are provided. => according to the documentation of the CODESYTS IIoT lib, there is a maximum message payload length. In the CODESYS IIot MQTT library, is a single subscriber only capable of just subscribing to a single topic? => I Assume so, as the provided example shows a single topic per subscriber instance. aliazzzMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/32/https://forge.codesys.com/lib/cobolt/tickets/24/~~~ FB_MQTT_PubSub /Client ClientBusy(Private) ClientDone(Private) ClientError() Connect() Disconnect() InitialiseClient(Private) IsConnected() IsConnecting() SetBroker() SetLastWill() SetSessionCredentials() SetSessionTls() GetConnectionState() /Publisher Publish() PublisherBusy() PublisherDone() PublisherError() /Subscriber SetSubscriberBuffer => // !needs implementation! IsMessageReceived() // TRUE if a message is ready to be digested GetMessage() // Provides the message itself => DEBUG! provides NOTHING currently :-( Subscribe() SubscriberBusy() SubscriberDone() SubscriberError() Unsubscribe() ~~~ *The behaviourmodel is phased out as it is overkill because each of the composite FB's allready implemented their own CMBL.IConc statemachine* aliazzzMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/24/https://forge.codesys.com/lib/cobolt/tickets/23/FB_EoN_Node // Represents a Sparkplug Edge of Network Node and is responsible for publishing Node and // Device metrics (thus integration without separate devices is possible) encapsulates FB_MQTT_PubSub // MQTT Client & Publisher FB_Payload // Responsible for payload blob encoding and decoding FB_Sparkplug // Responsible for the Sparkplug B Topics & Messages Optional FB_Device // Represents a single Sparkplug Device. Responsible for publishing Device Metrics // In theory the amount of devices is unlimited and is only limited by the adress namespace and maximum message // length. We will limit the number of devices default to 32 (configurable via a library parameter) aliazzzSat, 03 Oct 2020 23:46:45 -0000https://forge.codesys.com/lib/cobolt/tickets/23/https://forge.codesys.com/lib/cobolt/tickets/18/The cforge svn guidelines are in the logical location: https://forge.codesys.com/forge/wiki/Migrating%20Projects/ so folder structure should be: license.txt STRUCTURE.MD .drone.yml branches/ tags/ trunk/ mqttsparkplugb.package <== created with cforge > build??? mqttsparkplugb/ package.manifest <== create with cforge > create package mqttsparkplugb/ <== Library (created with CODESYS SVN) unittest/ <== project with cf unit unittests (created with CODESYS SVN) mqttsparkplugb-example/ <== project (created with CODESYS SVN) mqttsparkplugb.library <== created with cforge > comit?? unittest.project <== created with cforge > comit?? mqttsparkplugb-example.project <== created with cforge > comit?? i-campbellMon, 14 Sep 2020 07:27:06 -0000https://forge.codesys.com/lib/cobolt/tickets/18/https://forge.codesys.com/lib/cobolt/tickets/17/I haved moved the trunk ./ to trunk/library ./aliazzzSun, 29 Mar 2020 19:52:04 -0000https://forge.codesys.com/lib/cobolt/tickets/17/https://forge.codesys.com/lib/cobolt/tickets/9/Requirement: quick to setup, access to all initial developers Solution: Host: broker.hivemq.com TCP Port: 1883 No authentication or certificate verification possible See http://www.mqtt-dashboard.com/ for their current traffic. For a good MQTT explorer; http://www.jensd.de/apps/mqttfx/1.7.0/mqttfx-1.7.0-windows-x64.exe or (slightly more experimental) http://mqtt-explorer.com/ Opensource, free and available for any OS by Thomas Nordquist i-campbellSun, 29 Mar 2020 19:52:04 -0000https://forge.codesys.com/lib/cobolt/tickets/9/https://forge.codesys.com/lib/cobolt/tickets/7/[Git] seems to point to an old version of repository, with an old version (v2.1) of the SparkPlug B spec. Suggest we upgrade to https://github.com/eclipse/tahu Comparing the spec though, it is just changing the owner from CirrusLink to eclipse.i-campbellMon, 14 Sep 2020 07:37:59 -0000https://forge.codesys.com/lib/cobolt/tickets/7/https://forge.codesys.com/lib/cobolt/tickets/5/This FB should form the basis for every Sparkplug B Node. There can be several Sparkplug B Nodes in a network. Every Sparkplug B Node can have any number of devices (0..N), but the Node itself is singular. It bridges between the MQTT Broker and device(s), sensor(s) and IO(s) An implementation can be made after the IEC "Google Protobufs' ticket has been Addressed. Suggestions for a lean and mean but proper readable implementation are welcome. I'd suggest we try to follow/translate a library example provided by SparkplugB. ~~~ FB_EoN_Node FB_Init() Initialise() onConnect() onMessage() publishBirth() publishDeviceBirth() publishNodeBirth() SessionState() ~~~aliazzzSat, 03 Oct 2020 23:43:38 -0000https://forge.codesys.com/lib/cobolt/tickets/5/https://forge.codesys.com/lib/cobolt/tickets/4/The Eon device interacts with the MQTT server in a session. The session state management is well described in the techref. PDF. This session state management can be implemented via a state machine. Or if the state machine is 'simple' we can choose to implement is via simple binary logic. aliazzzMon, 14 Sep 2020 07:27:14 -0000https://forge.codesys.com/lib/cobolt/tickets/4/