Extract Class A End Device Type functionality into its own class #53
Conversation
| @@ -17,6 +17,8 @@ | |||
| * | |||
| * Author: Davide Magrin <[email protected]> | |||
| * Martina Capuzzo <[email protected]> | |||
| * | |||
| * Modified by: Peggy Anderson <[email protected]> | |||
There was a problem hiding this comment.
I added this line as this refactor impacts the work flow of how a simulation/program would be run with the module as people will now have to use the ClassAEndDeviceLorawanMac class instead of the EndDeviceLorawanMac class.
…wanMac from EndDeviceLorawanMac
…e-lorawan-mac.cc|h
peggyopal
force-pushed
the
extract-class-a-end-device-mac
branch
from
ff888ab
to
ef0da1d
Compare
The fact that this variable was not initialized caused the lengths of receive windows be 0, not allowing the GW to correctly send a message back to the device when ACKs were requested, for instance. Bug found thanks to network-server-example.
|
Peggy, I completed my review, thank you for this contribution! I was able to spot a bug that caused receive windows to be closed immediately, making downlink communication impossible - however, I easily fixed it by initializing the Other than that this pull request looks good to me, but it also highlighted how we need more tests to ensure the MAC works as expected - I might be able to get some students to work on this, we'll see. We'll keep this on the develop branch for a bit, so that if other issues surface we can take care of them without disturbing the main branch - however you can definitely base your further work upon this contribution, with the certainty that it will make it to master eventually. |
|
Rebased on develop and pushed, thanks again! |
This pull request works towards issue #39
This pull request fixes issue #54
Proposed Changes
ClassAEndDeviceLorawanMacclass that extendsEndDeviceLorawanMacEndDeviceLorawanMacto new classEDtoED_Ainlorawan-mac-helper.ccin anticipation for the addition of Class C EDClassAEndDeviceLorawanMacEndDeviceLorawanMacwere made protected