Port to Mbed OS6

[kvkc97]

I would like to port the lorawan fuota code to mbed os6 version. How am I supposed to do that?

[kvkc97]

Can I get an update on this issue @janjongboom

[janjongboom]

@kvkc97 I have left Arm two years ago, and no-one in Arm has taken over the LoRaWAN stack. So I don't think this'll happen anytime soon.

[StefanFirefly]

@janjongboom. I realize that you left ARM/MBED and are sooo past the MBED LoraWAN stack. However, your Lora FUOTA example is the only one around for MBED. It's well written and well commented, so it should be usable in a MBED 6 context too (with some minor changes) - I need this also... but one thing elludes me when going through your code.

In your main.cpp you use the 'get_session/set_session' call when changing back and forth between CLASS-A/C to dynamically update (and restore) the current-session keys with the multicast-keys.
I get that 'get_session/set_session' methods are patches to the original MBED LoRaWAN implementation... but I dont understand why you need to do this.

Why can the obtained multistcast-keys and address not be stored in the multicast-linked-list (multicast_params_t *multicast_channels;) for which there already exists interface-methods?

multicast_channel_link (multicast_params_t *channel_param)   // For adding a set of multicast keys & address
multicast_channel_unlink (multicast_params_t *channel_param)   // For removing

Changing the session parameters directly - as you do - would that also not prevent you from receiving/sending unicasts while in Class-C mode? Maybe it's obvious and I'm missing some point here (wouldn't be the first time), I dont't know. Anyway, I comment on this in an off-chance that you remember you old work and have to time to comment back - if not, no offence will be taken :-)