fix reading encrypted frames that don't line up with network frames#64
Conversation
we want to call step again if there's remaining data. pos + toWrite is what we consumed from this network frame to complete the encrypted frame. so if it's less than the total network frame (data.length), then call step again.
|
@beowulfe any chance you can grant me reviewing powers ? I'd like to give this a +1 |
|
Thanks @ccutrer - this is a great fix! Could be the cause of a quite a few odd issues! @timcharper - looks like I can't give granular permissions on a personal repo. But Inacknowledge your +1 :) |
|
I'm really rather wondering if the window covering bug was triggered with this! And removing the other characteristics caused it to stop triggering it. It was failing in fantastic ways. Such a great fix!!! @beowulfe do you feel like we're ready for a new point release? I'd like to start merging my openhab changes and integrating into master. |
|
I'd like to get ccutrer@1544779 (or some variation thereof) merged before a new point release. It was far easier to remove the LowBatteryStatusAccessory rather than trying to keep it compatible-but-deprecated. See #70. |
|
I'll move the discussion there :) |
we want to call step again if there's remaining data. pos + toWrite
is what we consumed from this network frame to complete the encrypted
frame. so if it's less than the total network frame (data.length),
then call step again.