Getting device name

Firmware
1

I have the following code which I like to include in all of my applications so I know what application each device is running and what version(date) of the application and OS by simply requesting the associated Particle variable:

#define __FILENAME__ (strrchr(__FILE__, '/') ? strrchr(__FILE__, '/') + 1 : __FILE__)
char gDeviceInfo [200] = "";
void updateDeviceInfo() {
  snprintf(gDeviceInfo, sizeof(gDeviceInfo)
          ,"Application: %s, Date: %s, Time: %s, System firmware: %s, SSID: %s"
          ,__FILENAME__
          ,__DATE__
          ,__TIME__
          ,(const char*)System.version()  // cast required for String
          ,(const char*)WiFi.SSID()       // cast not required but always safe when in doubt if String or char*
          );
}

I also have this in setup()

updateDeviceInfo();
Particle.variable("deviceInfo",gDeviceInfo);

I would like to include the actual name of the device in this Particle variable and so I made the following changes:

gDeviceName  char [40] = "";
void nameHandler(const char *topic,const char *data) {
  strncpy(gDeviceName,data,sizeof(gDeviceName-1));
  updateDeviceInfo();
}

I also added the following to setup();

Particle.subscribe("particle/device/name",nameHandler);
Particle.publish("particle/device/name",PRIVATE);

and modified my updateDeviceInfo to this:

void updateDeviceInfo() {
  snprintf(gDeviceInfo, sizeof(gDeviceInfo)
          ,"Name: %s, Application: %s, Date: %s, Time: %s, System firmware: %s, SSID: %s"
          ,(const char*)gDeviceName
          ,__FILENAME__
          ,__DATE__
          ,__TIME__
          ,(const char*)System.version()  // cast required for String
          ,(const char*)WiFi.SSID()       // cast not required but always safe when in doubt if String or char*
          );
}

I’m finally getting SOMETHING out for Name, but only 4 characters. The actual device name is MailboxChimeMonitor, but here’s the output when I look at deviceInfo:

Name: Mail, Application: MailboxChimeMonitor.ino, Date: Jan  5 2021, Time: 15:44:29, System firmware: 2.0.1, SSID: HVAC Star 2.4

I’ve been looking at this too long and can’t see the forest for the trees, as they say. Suggestions would be greatly appreciated!

2

Not sure what this syntax is.
Didn't you rather mean to write

char gDeviceName[40] = "";

Also in order to debug your subscribe handler you may want to add some Serial.print() statements to monitor the incoming and outgoing data :wink:

3

Hahaha…typed too fast. It’s actually what you had suggested…my brain was not engaged with my fingers.

The device is remote, so I might have to try with one that’s local. However, other than my mis-typing above, everything else must look “reasonable”. Thanks!

4

I think you want:

strncpy(gDeviceName,data,sizeof(gDeviceName)-1);

not

strncpy(gDeviceName,data,sizeof(gDeviceName-1));

I’m kind of surprised that compiled.

5

:man_facepalming:…I just couldn’t see it. Thanks! Yay!!! It’s working now!

6

By the way, does this technique work for all device types (Argon, Boron, Electron and Photon)?

Also, what are the timing considerations? I have an Electron and used the above code, but seemed to have problems when I had the Particle.subscribe() in setup() followed by the Particle.publish(). Does best practice require some time pass between the two statement executions?

I ended up putting the Particle.publish() in a portion of the loop() function and that resolved the issue.

7

Yes, putting Particle.subscribe() in setup() and calling Particle.publish() from loop() after Particle.connected() returns true is the recommended method when using system thread mode.

If you use non-threaded AUTOMATIC mode the example in the docs works as-is, but it’s on the list to fix the basic example so it works properly if threading is enabled. Essentially, it’s possible to get into a situation where the publish goes out and the result comes back before the subscription is complete if you do both back-to-back in certain modes.