Another case for GCentral

[The Q]

I'm working on a personal project (more information will be shared about this later) that needs Message Queue Telemetry Transport (MQTT).  While searching for LabVIEW libraries for MQTT I found 1 on VIPM, 2 in the NI Forums, and 1 through Goggle on GitHub, as follows:

• WireQueue-MQTT Driver for LabVIEW by WireFlow AB (this one costs $550)
• MQTT Client API in native LabVIEW by Peter - daq.io (also on GitHub as LVMQTT)
• MQTT-LabVIEW by Michal Radziwon
• Quaxo MQTT LabVIEW by Stefan May

This is not unusual for just about anything you might be looking for.  In fact searching on GitHub there are 13 results for LabVIEW+MQTT.  What was weird is that two of them were almost completely the same, yet neither attribute the other.  I don't know which came first.  I ended up forking from one of them but I guess I'll attribute both to be safe if I end up using it.

However, talking about code confidence, I just found this one: LV-MQTT-Broker by @Francois Normandin.  I know Francois, he is a LabVIEW Champion.  He has included unit tests.  It has full documentation as well as an NIWeek presentation by him and Sarah Zalusky, both of whom are Certified LabVIEW Architects (CLAs).  From GitHub I can see he has been actively contributing to it and its open source (which most of them were).  Honestly, I wish I had found this one first.

Just some words for thought...

 

 

Edited January 29, 2020 by The Q

[Jordan Kuehn]

Honestly MQTT should have a native implementation for as much as NI has been hyping IIoT over the last years IMHO. 

I do get the overall point about making a case for GCentral though.

[Phillip Brooks]

We should be getting TLS support based on this:

https://forums.ni.com/t5/LabVIEW-Idea-Exchange/SSL-TLS-Support/idi-p/3314187

Maybe 2020 is the year!

[The Q]

23 hours ago, Jordan Kuehn said:

Honestly MQTT should have a native implementation for as much as NI has been hyping IIoT over the last years IMHO. 

That’s probably true but we can’t rely on NI to produce a native implementation of every protocol out there. That’s why we need to make it easier to help each other. 

[Francois Normandin]

@The Q My library is not complete by any means, but what works really works... I've followed a TDD approach. So far I've covered only 79 requirements out of 141, and I haven't started any branches to support MQTT 5.0 yet... So it covers the basic needs such as shown in the presentation you're referring to, but nothing fancy like QoS 1 and 2 and the likes. (or TLS for that matter). If you want something more complete, I'd check Wireflow's implementation first.

Now, I'd really like to get other folks to contribute to my Open Source Project. It would give me a moral boost to continue pushing it further. In the meantime, thanks for the advertisement!!!

Edited January 31, 2020 by Francois Normandin

[The Q]

@Francois Normandin,

I'm finishing another project now but I will be returning to this soon.  If it doesn't do what I need yet, I'll work with you on it to accomplish more of the features.